KR20180064969A - RIDAR Apparatus Based on Time of Flight and Moving Object - Google Patents

Abstract

Embodiments of the present invention provide a flight time-based lidar apparatus and a moving object. The flight time-based lidar apparatus and the moving object transmit and receive a light to output an electric signal, analyze the electric signal to generate a control signal, calculate flight time of the light based on a control signal to measure a pin point distance, and process point cloud data generated in a basis of the measured distance, thereby capable of precisely constructing information regarding a surrounding environment.

Description

BACKGROUND OF THE INVENTION 1. Field of the Invention [0001] The present invention relates to a RIDAR-

The technical field to which this embodiment belongs is a flight time-based Lada device and a moving object.

The contents described in this section merely provide background information on the present embodiment and do not constitute the prior art.

LIght Detection And Ranging (LIDAR), which is based on Time Of Flight, is a device that measures the time of reflected light by shooting an optical signal and measuring the distance of the reflector using the speed of light. The received optical signal is converted into an electrical signal through the photodiode.

The embodiments of the present invention provide a method and apparatus for transmitting and receiving light, outputting an electric signal, generating a control signal by analyzing the electric signal, calculating a flight time of the light based on the control signal, The main purpose of the invention is to construct information about the surrounding environment by processing the point cloud data generated based on the measured distance.

Embodiments of the present invention may include (i) a first mirror including a hole through which light emitted from a light source passes, the first mirror having a focal point at which light reflected by a curved reflective surface converges, and (ii) By including the second mirror which reflects the light to the object and receives the light reflected from the object and reflects it to the first mirror and moves in the vertical direction, it is possible to minimize the size and cost of the product without using the array type, There is another purpose of the invention in measuring points.

The embodiments of the present invention do not need to wait for the output signal to disappear by converting the received light into an electrical signal and outputting an electrical signal during the detection period in the sampling period to uniformize the irregular signal processing time Another object of the invention is to stabilize the operation of the system and improve the processing speed of the system.

The embodiments of the present invention can be implemented by converting the electrical signal so that the signal point having the maximum signal magnitude in the electrical signal has a predetermined magnitude, adjusting the magnitude of the converted electrical signal, There is another purpose of the invention to improve the error and to measure the exact time with a comparator alone.

Embodiments of the present invention have another object of the present invention to calculate an accurate flight time by correcting the flight time using the pulse width of the reflected signal.

It is another object of the present invention to improve time resolution of a time-digital converter by changing the position of a logic element included in a plurality of oscillators to adjust the clock width of a plurality of oscillators.

Embodiments of the present invention are directed to a method of calculating an accurate flight time by correcting a flight time using a ratio of the number of reference clocks received from an external clock generator and the number of internal clocks generated in an internal oscillator, have.

Embodiments of the present invention provide an artificial landmark that is visible to the user but which recognizes the artificial landmarks and provides a recognizable artificial landmark, Another purpose of the invention is to recognize the artificial markers by analyzing the scanned points during the scanning process.

The embodiments of the present invention generate a key frame related to a node by using scan information, calculate an odd metric edge between consecutive nodes, update a key frame to estimate a local map, The present invention provides a map that covers a large area within a certain time under a certain condition by detecting a loop closure edge between nodes that do not exist and correcting the position of the node based on the edge of the odometry and the loop closure, There is another purpose.

Embodiments of the present invention introduce a cleaning robot equipped with a multi-channel laser (LiDAR) for acquiring the surrounding environment as three-dimensional spatial information, and use it to acquire and display environment information. An object of the present invention is to provide an obstacle detecting device and method for a vehicle, and a mobile object having the obstacle detecting device and method.

Embodiments of the present invention provide an apparatus and method for detecting obstacles of a multi-channel lidar-based mobile object that distinguish walls, obstacles (including low obstacles), and vertical environments from obtained three-dimensional environment information, There is another purpose.

 Embodiments of the present invention provide an apparatus and method for detecting obstacles of a multi-channel lidar based mobile object having a collision path avoidance algorithm that increases the efficiency of a cleaning robot while avoiding collision based on previously measured environmental obstacle information, There is another purpose of the invention to provide.

Embodiments of the present invention include a cleaning function control device for acquiring information about each point in a room as three-dimensional information to determine the state of the floor and controlling the cleaning function for the room based on the state of the floor Another object of the present invention is to provide a cleaning robot having such a cleaning robot.

Other and further objects, which are not to be described, may be further considered within the scope of the following detailed description and easily deduced from the effects thereof.

According to an aspect of the present invention, there is provided a navigation system including a Lada device for measuring a distance from a target based on a flight time, a path controller for processing a point cloud data generated based on the measured distance to generate a travel route, And a mobile device implemented to move the mobile device based on the mobile device.

According to another aspect of the present invention, there is provided an optical transceiver including: an optical transceiver for transmitting and receiving light and outputting an electric signal; a signal discriminator for analyzing the electric signal to generate a control signal; And a distance measuring unit for measuring the distance.

According to another aspect of the present invention, there is provided a light-emitting device including: a light source that emits light; a first mirror that includes a hole through which the emitted light passes; light that passes through the hole of the first mirror, A second mirror for receiving the reflected light and reflecting the reflected light to the first mirror, and a photodiode for receiving the light reflected from the second mirror and converting the received light into an electric signal.

According to another aspect of the present invention, there is provided an optical transceiver including: an optical transceiver that emits light to a target object by a start control signal and receives light reflected by the target object and converts the received light into an electrical signal; And a distance measuring unit for measuring a distance by calculating a flight time based on a time difference between the start control signal and the stop control signal, wherein the optical transceiver includes a light source for emitting light and a hole for passing the emitted light, A first mirror having a focal point at which a light beam reflected by a curved surface is gathered; a light source that reflects light passing through the hole of the first mirror to the object, receives light reflected from the object, A reflector positioned at a focal point of the first mirror for receiving and reflecting the light reflected from the first mirror, And a rotation unit for rotating the light source, the first mirror, the second mirror, the reflector, and the photodiode with respect to a rotation axis, characterized by comprising: a photodiode for receiving light reflected from a carcass and converting the light into an electric signal; To provide a Lidar device.

According to another aspect of the present invention, there is provided a mobile device comprising: a Lada device for calculating a flight time between the mobile object and a target object to measure a distance to the target object; And an optical transceiver for outputting light to the object by a start control signal and for receiving the light reflected by the object and converting the received light into an electric signal, And a distance measurer for calculating a distance by calculating a flight time based on a time difference between the start control signal and the stop control signal, wherein the optical transceiver comprises: a light source for emitting light; A first mirror including a hole, a first mirror having a focal point at which a ray reflected by a curved reflecting surface is collected, A second mirror that reflects light passing through the hole of the wool to the object, receives light reflected from the object and reflects the light to the first mirror, and a second mirror positioned at a focal point of the first mirror, A reflector for receiving and reflecting the reflected light, a photodiode for receiving the reflected light from the reflector and converting the reflected light into an electric signal, and a control unit for controlling the light source, the first mirror, the second mirror, the reflector, And a rotation unit that rotates by a reference.

According to another aspect of the present invention, there is provided an optical transceiver for emitting light to a target object by a start control signal, receiving light reflected by the target object, converting the received light into an electric signal, and outputting the electric signal for a predetermined detection time, And a distance measurer for calculating a flight time based on a time difference between the start control signal and the stop control signal to measure a distance.

According to another aspect of the present invention, there is provided a light source device including a light source that emits light to a target object based on a predetermined sampling period, a photodiode that receives light reflected from the target object and converts the light into an electrical signal, And an optical transceiver including a signal converter for outputting an electrical signal.

According to another aspect of the present invention, there is provided a mobile device comprising: a Lada device for calculating a flight time between the mobile object and a target object to measure a distance to the target object; Wherein the light emitting device emits light to a target object by a start control signal and receives light reflected by the target object and converts the received light into an electric signal, A transceiver, a signal discriminator for converting the electric signal to generate a stop control signal, and a distance measurer for calculating a flight time based on the time difference between the start control signal and the stop control signal, As shown in FIG.

According to another aspect of the present invention, there is provided an optical transceiver that emits light to a target object by a start control signal, receives light reflected by the target object and converts the light into an electrical signal, And a distance measuring unit for measuring a distance by calculating a flight time based on a time difference between the start control signal and the stop control signal.

According to another aspect of the present invention, there is provided a signal processing apparatus including: a first conversion unit for converting an input signal having a predetermined size to a signal point having a maximum signal size with respect to an ascending and descending input signal; And a signal detector for detecting at least one time point having a predetermined reference magnitude from the magnitude-adjusted input signal to generate an output signal.

According to another aspect of the present invention, there is provided a moving object, comprising: a ladder device for calculating a flying time between the moving object and a target object to measure a distance to the target object; and a moving device for moving the moving object based on the distance to the object An optical transceiver for emitting light to a target object by a start control signal and for receiving the light reflected by the target object and converting the received light into an electrical signal; And a distance measuring unit for measuring the distance by calculating the flight time based on a time difference between the start control signal and the stop control signal, Thereby providing a moving body.

According to another aspect of the present invention, there is provided an optical communication system including: an optical transceiver that emits light to a target object by a start control signal and receives light reflected by the target object and converts the light into an electrical signal; And a distance measurer for calculating a flight time using a time digital converter that adjusts the position of a logic element included in the oscillator based on a time difference between the start control signal and the stop control signal, Lt; / RTI > device.

According to another aspect of the present invention, there is provided an oscillator circuit comprising: a slow oscillator generating a first clock; a fast oscillator generating a second clock less than the first clock; a normal counter counting the first clock of the slow oscillator; A precise counter for counting the second clock, and a phase detector for detecting when the first clock and the second clock are synchronized.

According to another aspect of the present invention, there is provided a mobile device comprising: a Lada device for calculating a flight time between the mobile object and a target object to measure a distance to the target object; And an optical transceiver for outputting light to the object by a start control signal and for receiving the light reflected by the object and converting the received light into an electric signal, And a distance measurer for calculating a flight time using a time digital converter that adjusts the position of a logic element included in the oscillator based on a time difference between the start control signal and the stop control signal, And a plurality of moving objects.

According to another aspect of the present invention, there is provided an optical transceiver for emitting light to a target object, receiving light reflected by the target object and converting the received light into an electric signal, measuring a distance between the points of the target object using the time difference, A light intensity meter for measuring the intensity of light reflected at the points of the object, and a controller for determining whether the object is an artificial marker by analyzing a change in the distance and the light intensity, And an artificial landmark detector for analyzing the invisible bar code to generate a digital code.

According to another aspect of the present embodiment, there is provided an optical device including a base and an invisible bar code connected to the base and having a code sequence formed of a material that reflects, absorbs, or emits light in a first wavelength band, Wherein the light source is hidden by a material that reflects, absorbs, or emits light in a two-wavelength band and is rendered invisible.

According to another aspect of this embodiment, there is provided a computer program for artificial landmark recognition, recorded in a non-transitory computer readable medium comprising computer program instructions executable by a processor, Measuring a distance between points of the object using a time difference when the light is emitted to the object and the reflected light is received by the object; And analyzing changes in the distance and the light intensity to determine whether the object is an artificial marker and analyzing an invisible bar code included in the artificial marker to generate a digital code, A computer program for executing the program.

According to another aspect of the present invention, there is provided a mobile device including a Lada device for emitting and receiving light to a target object to recognize an artificial landmark, and a mobile device implemented to move the mobile object based on a distance to the target object , The lidar apparatus includes an optical transceiver for emitting light to the object and receiving the light reflected by the object and converting the received light into an electric signal, a distance measuring unit for measuring a distance between the points of the object using the time difference, A light intensity meter for measuring the intensity of light reflected at the points of the object, and a controller for determining whether the object is an artificial marker by analyzing a change in the distance and the light intensity, Characterized in that it comprises an artificial landmark detector for analyzing the invisible bar code to produce a digital code It provides.

According to another aspect of the present invention, there is provided a method for locating and mapping a mobile robot, the method comprising: acquiring scan information of a space in which the mobile robot is located; Generating a key frame, calculating an odometry edge between consecutive nodes, updating the key frame to estimate a local map, and estimating a local map based on the set of updated key frames And estimating a global map by detecting a loop closure edge between nodes that do not exist and correcting the position of the node based on the odometry edge and the loop closure edge, It provides a method of recognition and mapping.

According to another aspect of the present invention, there is provided a map generator for a mobile robot, comprising: a scanner for acquiring scan information of a space in which the mobile robot is located; a key frame for a node using the scan information; A local map estimator for calculating an odometry edge between consecutive nodes and updating the key frame to estimate a local map, and a local map estimator for estimating a local map, And a global map estimator for detecting a loop closure edge between the nodes and estimating a global map by correcting the position of the node based on the odometry edge and the loop closure edge, Provides a map generator.

According to another aspect of this embodiment, there is provided a computer program for location awareness and mapping written to a non-transitory computer readable medium comprising computer program instructions executable by a processor, The method comprising: acquiring scan information of a space in which the mobile robot is located when the mobile robot is executed by at least one processor of the robot; generating a key frame for the node using the scan information; Calculating an odometry edge between consecutive nodes and updating the key frame to estimate a local map, and determining a loop closure edge between nodes that is not contiguous to the updated set of key frames (Loop Closure Edge) is detected, and the odometry edge and the loop closure edge are detected To provide a computer program for performing the operations, including the step of estimating a global map (Global Map) to correct the position of the node.

In order to detect three-dimensional environment information including obstacle information of a moving object, an obstacle detecting apparatus for a multi-channel lidar-based moving object according to another aspect of the present invention includes a multi- Generating three-dimensional environment information for defining an obstacle on three-dimensional basis based on reception information of the at least two optical transceivers, and generating a projection map including the three-dimensional environment information on a two-dimensional space map And a traveling route generating unit for determining a position on the space of the obstacle based on the projection map and generating a traveling route for avoiding the obstacle.

In another aspect of the present invention, there is provided an obstacle detection method for a multi-channel lidar-based moving object, comprising the steps of: acquiring three-dimensional environment information of a mobile object using at least two optical transceivers having different emission angles of a transmission signal; Generating three-dimensional environment information for defining obstacles in three dimensions on the basis of the three-dimensional environment information, generating a projection map including the three-dimensional environment information in a two-dimensional space map, Based wall extraction algorithm to extract wall information and to detect low obstacle information using the three-dimensional environment information acquired by the at least two optical transceivers.

According to another aspect of the present invention, there is provided an indoor-room-information obtaining apparatus including an indoor-information obtaining unit for obtaining information on each point of a room as three-dimensional information, a state information obtaining unit for obtaining a state of the floor based on information about first points related to the floor of the indoor space And a cleaning function control unit for controlling the cleaning function for the room based on the state of the floor surface.

According to another aspect of the present invention, there is provided an indoor-room-information obtaining apparatus including an indoor-information obtaining unit for obtaining information on each point of a room as three-dimensional information, a state information obtaining unit for obtaining a state of the floor based on information about first points related to the floor of the indoor space A cleaning function control unit for controlling the cleaning function for the room based on the condition of the floor surface, and a cleaning function performing unit for performing a cleaning function for the room, We propose a robot.

As described above, according to the embodiments of the present invention, the light is transmitted and received to output an electric signal, the control signal is generated by analyzing the electric signal, and the flight time of the light is calculated based on the control signal And the point cloud distance is measured, and the point cloud data generated based on the measured distance is processed, so that the information on the surrounding environment can be accurately constructed at a low cost.

According to embodiments of the present invention, there is provided an optical system including: (i) a first mirror having a hole through which light emitted from a light source is passed, the first mirror having a focal point at which light reflected by a curved reflective surface converges; and (ii) And a second mirror that reflects light passing through the first mirror and reflects the light reflected from the object to the first mirror and moves in the vertical direction, thereby minimizing the size and cost of the product without using the array type, The pin point can be measured with high performance.

According to the embodiments of the present invention, it is unnecessary to wait until the output signal disappears by receiving the reflected light and converting it into an electrical signal and outputting an electrical signal during the detection time in the sampling period. The operation of the system can be stabilized and the processing speed of the system can be improved.

According to the embodiments of the present invention, the electric signal is transformed so that the signal point having the maximum signal magnitude in the electric signal has a predetermined magnitude, the magnitude of the converted electric signal is adjusted, Thus, it is possible to improve the work error and to measure the accurate time by using only the comparator.

According to the embodiments of the present invention, it is possible to calculate the accurate flight time by correcting the flight time using the pulse width of the reflected signal.

According to the embodiments of the present invention, the time resolution of the time-digital converter can be improved by adjusting the clock width of a plurality of oscillators by adjusting the position of logic elements included in a plurality of oscillators.

According to embodiments of the present invention, an accurate flight time can be calculated by correcting the flight time using the ratio of the number of reference clocks received from the external clock generator and the number of internal clocks generated in the internal oscillator .

According to embodiments of the present invention, an apparatus that recognizes an artificial landmark, which is not visible to the user's eyes, provides an artificial landmark that can be recognized, and a lidar system needs to have a separate camera for recognizing the artificial landmark It is possible to recognize the artificial landmarks by analyzing the scanned points in the process of scanning the surroundings without the scanner.

According to embodiments of the present invention, a key frame related to a node is generated using scan information, an odd map tree is calculated between successive nodes, a key frame is updated to estimate a local map, It is possible to create a map covering a wide area within a predetermined time under a certain condition by detecting a loop closure edge between non-contiguous nodes and correcting the position of the node based on the odometry edge and the loop closure edge, There is an effect.

According to the embodiments of the present invention, it is possible to three-dimensionally acquire obstacle information for the surrounding environment of the moving object by providing a multi-channel structure capable of rotating and transmitting / receiving signals at various angles, It is possible to detect low-level obstacles at the bottom.

According to the embodiments of the present invention, it is possible to generate a projection map including three-dimensional information on the surrounding environment acquired through the multi-channel ladder in a grid map, so that even with low storage capacity and computing power, It is possible to sufficiently acquire obstacle information at a lower level in the lower front.

According to the embodiments of the present invention, even when acquiring three-dimensional information about the surrounding environment of a moving object through multi-channel rendering, it is possible to apply the techniques applied to the existing two- .

According to the embodiments of the present invention, even in the case of acquiring three-dimensional information on the surrounding environment of the moving object through the multi-channel line, the projection map is generated, thereby securing the safety of the moving object and securing the maximum cleaning area.

According to the embodiments of the present invention, it is possible to accurately detect the state of the bottom surface, to detect low obstacles scattered on the floor surface, to accurately detect the slip of the cleaning robot, And the coverage performance can be improved.

Even if the effects are not expressly mentioned here, the effects described in the following specification which are expected by the technical characteristics of the present invention and their potential effects are handled as described in the specification of the present invention.

1 to 3 are views of a moving body according to embodiments of the present invention.
4 and 5 are diagrams of a laddering device according to embodiments of the present invention.
6 to 15 are views of an optical transceiver of a lidar device according to embodiments of the present invention.
16 and 17 are views of point cloud data generated by a moving object according to embodiments of the present invention.
18 to 21 are views showing signal conversion units of an optical transceiver of a radar apparatus according to embodiments of the present invention.
22 to 27 are diagrams illustrating a signal discriminator of a radar device according to embodiments of the present invention.
28 to 34 are diagrams of a time digital converter of a laddering device according to embodiments of the present invention.
FIG. 35 is a view for explaining an operation for correcting the flight time by the Lada apparatus according to the embodiments of the present invention. FIG.
FIGS. 36 to 51 are views for explaining an operation of recognizing artificial landmarks according to the embodiments of the present invention.
52 is a view illustrating how moving objects interact according to embodiments of the present invention.
Figs. 53 to 64 are diagrams for explaining an operation in which the Lada device according to the embodiments of the present invention recognizes a position and generates a map. Fig.
65-75 are views of a path controller of a laddering device according to embodiments of the present invention.
Figures 76-83 are diagrams of a projection map generator of a laddering device in accordance with embodiments of the present invention.
84 to 88 are views of a cleaning part of a moving object according to embodiments of the present invention.
FIGS. 89 to 96 are views for explaining an operation of moving a moving object according to the embodiments of the present invention.
97 is a diagram exemplarily showing a moving path of a moving object according to embodiments of the present invention.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Will be described in detail with reference to exemplary drawings.

<Mobile body>

1 to 3 are views of a moving body according to embodiments of the present invention.

As shown in Fig. 1, the mobile device 1 includes a ladder device 10, a mobile device 20, and a path controller 30. As shown in Fig. The moving object 1 may omit some of the various elements exemplarily shown in Fig. 1 or may further include other elements. For example, the moving body may further include a cleaning unit.

The moving body 1 means an apparatus designed to be movable from a specific position to another position according to a predetermined method and can be moved from a specific position to another position by using a moving means such as a wheel, a rail, a walking leg, . The moving object 1 may be moved according to information collected after collecting external information using a sensor or the like, or may be moved by a user using a separate operating means.

Examples of the moving body 1 include a robot cleaner, a toy car, a mobile robot which can be used for an industrial purpose or a military purpose, and the moving body 1 can be driven by wheels, walked using one or more legs , Combinations thereof, and the like.

The robot cleaner is a device that automatically cleans the cleaning space by suctioning foreign substances such as dust accumulated on the floor while traveling in the cleaning space. Unlike a general vacuum cleaner moving by an external force by a user, the robot cleaner cleans the cleaning space while moving using external information or a predefined movement pattern.

The robot cleaner may automatically move using a predefined pattern or may detect an external obstacle by a detection sensor and then move according to the detected motion or may be moved according to a signal transmitted from a remote control device operated by a user It is movable.

The sensing sensor may be implemented as a LIDAR. Lidar is a device that shoots a laser signal, measures the return time of reflected light, and measures the distance of the reflector using the speed of light. The laser signal is converted into an electrical signal through the photodiode. The laser signal may have a predetermined wavelength band.

Referring to FIGS. 2 and 3, the Lada device 10 for calculating the distance between the moving object and the object and measuring the distance to the object is located at the upper end of the main body. However, the present invention is not limited thereto. Lt; RTI ID = 0.0 &gt; and / or &lt; / RTI &gt;

The lidar apparatus 10 transmits and receives light using a pair of light sources and photodiodes, and scans the periphery by using a movable mirror and a rotating body.

The lidar apparatus 10 may operate in a time of flight (TOF) manner. The time of flight method measures the distance between the object to be measured and the Lidar device by measuring the time that the laser emits a pulse or a square wave signal to arrive at the receiver the reflected pulse or square wave signals from objects within the measurement range.

The mobile device 20 calculates a traveling route based on the distance to the object or detects an obstacle to move the moving object. The mobile device 20 may move the mobile object based on the relative position of the artificial landmarks. The mobile device 20 can be implemented as a moving means such as a wheel, a rail, a walking leg, and the like.

&Lt;

4 and 5 are diagrams of a laddering device according to embodiments of the present invention. The lidar device is implemented in a mobile object or operates independently. As shown in FIG. 4, the Lidar apparatus 10 includes an optical transceiver 100, a signal discriminator 200, and a distance measurer 300. The lidar device 10 may omit some of the various components illustrated in FIG. 4 or may additionally include other components. For example, the Lidar device 10 may further include an interface 400. [

The optical transceiver 100 transmits a laser signal and receives a reflected signal. The optical transceiver 100 emits light to a target object in response to a start control signal, receives the light reflected by the target object, and converts the received light into an electric signal. The optical transceiver 100 outputs an electric signal for a predetermined detection time.

The optical transceiver 100 converts light into current or voltage, which requires a circuit for buffering and scaling the output of the photodiode. For example, a transimpedance amplifier (TIA) may be connected to the photodiode. The transimpedance amplifier amplifies the current of the photodiode, converts it into a voltage, and outputs it. Transimpedance amplifiers can be classified into R-TIA (Resistive Feedback TIA) and C-TIA (Capacitive Feedback TIA).

The optical transceiver 100 may include a signal converter. A signal conversion unit may be connected to the photodiode of the optical transceiver 100, and a transimpedance amplifier may be connected to the signal conversion unit.

The light source emits light to the object based on a predetermined sampling period. The sampling period can be set by the control unit of the lidar apparatus 10. The sampling period is a period of time from when the optical transceiver 100 emits light, receives the reflected light, and converts the light into an electrical signal in accordance with the start control signal. The optical transceiver 100 may repeat these operations in the next sampling period.

The photodiode receives the light reflected by the object and converts it into an electric signal. The photodiode may be implemented as a PN junction photodiode, a PIN photodiode, an avalanche photodiode (APD), or the like. The photodiode outputs an electrical signal until the optical carrier disappears. In addition, as the size of the output signal increases, the time required until the signal disappears increases.

The signal discriminator 200 converts an electric signal to measure an accurate time and outputs a stop control signal. The signal discriminator 200 converts the electrical signal so that the signal point having the maximum signal magnitude has a predetermined magnitude, adjusts the magnitude of the converted electrical signal, and detects a point of time having a predetermined magnitude. The signal discriminator converts the electrical signal to generate a stop control signal.

The signal discriminator 200 receives electrical signals from a photodiode or a transimpedance amplifier. The received electrical signal, that is, the input signal, has a form of rising and falling by the reflected light. The signal discriminator accurately measures the desired point of view of the input signal and outputs an electrical signal.

The signal discriminator 200 differentiates the input signal or converts the input signal using a constant fraction discriminator (CFD). The constant fraction discrimination is a method of finding a point at which a point at which the signal delayed by the original signal is equal to a signal adjusted by a constant magnitude ratio becomes a certain ratio of the maximum magnitude.

The signal discriminator 200 converts the slope of the input signal so that the signal point having the maximum signal size has a predetermined size. The signal discriminator 200 adjusts the size of the converted input signal. The signal discriminator amplifies the magnitude of the converted input signal by N (where N is a natural number). The signal discriminator performs a plurality of amplification processes to convert the slope of the signal to be close to vertical. Since the slope is large, accurate timing can be obtained even if a circuit is simply implemented with a comparator.

The distance measuring device 300 measures time and distance in a time-of-flight manner. The distance measuring device 300 measures the distance by calculating the flight time based on the time difference between the start control signal and the stop control signal. The distance measuring device 300 calculates the distance from the time using the speed of light.

The distance measurer 300 may include one or more time digital converters 310 and 312 for converting the two time differences to digital values. The input signal of the time digital converter may be in the form of a pulse of the same signal source, or may be the edge of another signal source. For example, the radar apparatus 10 can calculate the time difference based on the rising edge or the falling edge of the start control signal, the rising edge or the falling edge of the stop control signal.

The interface 400 is a communication path for transmitting / receiving information to / from another device (or host). Other devices may be connected to the Raid device 10 via the interface to set parameters. The lidar device 10 can transmit the time and distance measured by the interface to another device.

&Lt; Structure of optical transceiver >

6 to 15 are views of an optical transceiver of a lidar device according to embodiments of the present invention.

The three-dimensional distance measurement system measures the distance of space using various sensors such as a CCD (Charge Coupled Device) image sensor, a CMOS (Complementary Metal Oxide Semiconductor) image sensor, an ultrasonic sensor, and a laser sensor.

A typical three-dimensional distance measurement system scans a space by rotating a two-dimensional distance sensor that scans a plane including the center of the sensor. Since the device using such a two-dimensional distance sensor does not limit the cost, the size, and the sampling rate, there is a limit in producing the product as a commercial product rather than a research purpose.

A device using a two-dimensional photodiode array measures the distance using Structure Light or Time of Flight. The structure light is a method of projecting a unique pattern and calculating a depth by detecting a corresponding point, and the flight time is a method of measuring a time difference or a phase difference and converting it into a distance. A device to which a two-dimensional photodiode array is applied has a problem that it is difficult to widen the angle of view, and it is difficult to measure pinpoint because of a large amount of three-dimensional information per pixel.

A lidar apparatus to which a one-dimensional photodiode array is applied includes a photodiode array and a laser diode array (or a laser diode and a diffuser). The photodiode array has a structure in which hundreds to several thousands of photodiodes are arranged linearly on the silicon crystal. Since the Lidar device with a one-dimensional photodiode array is difficult to widen the angle of view, and the high efficiency diffuser, sensor array, and MEMS mirror required for implementation are expensive, it is difficult to produce them as a commercial product have.

In order to solve such a problem, the optical transceiver 100 is formed in a structure capable of three-dimensional scanning.

The optical transceiver 100 can detect the obstacles in the horizontal direction and the ground direction simultaneously by setting the angles of the plurality of mirrors differently. The optical transceiver 100 connects the mirrors to the transmission optical unit 120 and the reception optical unit 130 and rotates the transmission optical unit 120 and the reception optical unit 130 to detect obstacles in all directions have. For example, the scan lines may be set to 45 degrees and 60 degrees, respectively, and may be configured to be two or more.

The optical transceiver 100 may be positioned above the moving direction of the moving object 200 and the at least two optical transceivers 101 and 102 may be mounted to have different transmission signal emission angles from each other in the horizontal downward direction. In the present embodiment, the multi-channel laser diode means one laser diode (LiDAR) including an optical transceiver that emits or receives a plurality of lasers.

9 is an operational example of a moving object 1 equipped with a two-channel distance measuring instrument using a high beam and a low beam, and Fig. 10 is a diagram showing an operation example of a high obstacle 2 and a low obstacle channel Lidar apparatus 10 capable of detecting all of the obstacles 4, 4, 5, 6, 7, 8, 10, 11, When the mobile body 1 uses a two-channel distance measuring device, one channel (ex. Low beam) is used in a direction parallel to the ground, and another channel (ex. High beam) is used in the direction of the ground and oblique lines .

Referring to FIG. 11, the optical transceiver 100 includes a light source 110, a first mirror 120, a second mirror 130, and a photodiode 140. Optical transceiver 100 may omit some of the various components illustrated in FIG. 11 or may additionally include other components. 12, the optical transceiver 100 includes a reflector 150, a rotation unit 160, a point control unit 112, a reference angle control unit 132, a movement angle control unit 134, a vertical frequency control unit 136, a horizontal frequency adjuster 162, or a combination thereof.

The light source 110 emits light, and may be implemented as a laser diode (LD) or the like. The light source can generate laser pulse signals in nanoseconds. The laser signal may have a predetermined wavelength band. The light source 110 may be connected to a point adjuster 112 for adjusting the number of point cloud data acquired per unit time by adjusting a speed at which the light source emits light based on a predetermined sampling period. For example, the point adjusting unit 112 may set the output speed of the light source 110 to obtain 10 K points per second.

The first mirror 120 includes a hole through which the emitted light passes. That is, a hole penetrating the reflecting surface is formed. The optical transceiver 100 minimizes the size of the optical transceiver 100 while maintaining the straight path of the light from the light source 110 to the second mirror 130 by forming a light path to pass through the first mirror 120 . The reflection surfaces of the first mirror 120 are not employed because optical components such as a two-way mirror or a half mirror that transmits a part of light are generally lower in sensitivity, And a light path of several millimeters passes through the second mirror 130 to form a light path.

The first mirror 120 is formed into a curved surface and has a focal point at which the rays reflected by the reflecting surface of the first mirror 120 converge. The size of the reflective surface of the first mirror 120 has an appropriate size based on the distribution of the reflected light according to the angle of the moving second mirror 130. The first mirror 120 performs a condensing and reflecting function using a parabolic reflecting mirror (a lens and a mirror are integrated into one). That is, the light reflected again through the second mirror 130 is collected and sent to the reflector 150.

The reflector 150 is positioned at the focus by the curved surface of the first mirror 120. The reflector 150 receives the light reflected from the first mirror 120 and reflects the light to the photodiode 140. The reflected light from the curved surface of the first mirror 120 may be directly received by the photodiode 140 while the reflected light from the curved surface of the first mirror 120 may be received directly by the photodiode 140. [ It can move in the vertical direction. That is, the reflected rays may form a parallel straight path or a straight path before entering the photodiode 140. The photodiode 140 is located on a virtual straight path passing through the focal point of the first mirror 120.

The second mirror 130 reflects the light passing through the hole of the first mirror to the object, receives the light reflected from the object, and reflects the light to the first mirror. The second mirror 130 moves in a predetermined cycle to change the slope of the normal of the second mirror. The second mirror 130 is a moving mirror and can be moved by bending, tremor, reciprocating, seesaw, rotating, or a combination thereof. For example, it can be implemented as a swing mirror.

The second mirror 130 may be connected to a reference angle adjusting unit 132 for adjusting the angle at which the second mirror 130 is installed. For example, the reference angle adjusting unit 132 may set the normal of the second mirror 130 to -55 degrees with respect to the horizontal plane of the ground, and the second mirror 130 may be inclined at 45 degrees.

The second mirror 130 may be connected to a movement angle adjusting unit 134 that changes the angle at which the second mirror 130 moves. For example, the movement angle adjuster 134 may be configured to cause the second mirror 130 to swing +/- 10 degrees.

A vertical frequency controller 136 may be connected to the second mirror 130 to change the period in which the second mirror 130 moves vertically. For example, the vertical frequency adjuster 136 may set the second mirror 130 to oscillate at 200 Hz.

The photodiode 140 is a device that receives light reflected from the second mirror 130 or a reflector and converts the light into an electric signal. The photodiode 140 can be applied to the principle that electrons act when a photon energy of light strikes a diode due to positive electron holes and positive electrons. The photodiode 140 may be implemented as a PN junction photodiode, a PIN photodiode, an avalanche photodiode (APD), or the like.

The optical transceiver 100 adjusts the movement path and angle of the light by using the second mirror 130 to ensure a vertical field of view so that the optical transceiver 100 can be operated with a conventional single lens and a photodiode array Unlike the implemented device, pinpoint measurement is possible.

The optical transceiver 100 may include a transmission optical unit and a reception optical unit. The transmission optical section and the reception optical section are paths of the laser signal and can be formed in a barrel structure. The optical transceiver 100 can detect the obstacles in the horizontal direction and the ground direction simultaneously by setting the angles of the plurality of mirrors differently. It is possible to connect mirrors to the transmission optical section and the reception optical section, respectively, and to detect the obstacles in all directions by rotating the transmission optical section and the reception optical section. For example, the scan lines may be set to 45 degrees and 60 degrees, respectively, and may be configured to be two or more.

The optical transceiver 100 may include a rotation unit 160. The optical transceiver 100 performs horizontal scanning through the rotation unit 160. [ The rotation unit 160 rotates the light source 110, the first mirror 120, the second mirror 130, and the photodiode 140 with respect to the rotation axis. The light source 110 and the photodiode 140 are mounted on a support and operate a driving device such as a motor connected to the support.

The rotation unit 160 may be connected to a horizontal frequency control unit 162 for controlling the rotation speed of the rotation unit 160. For example, the horizontal frequency adjusting unit 162 may be configured to rotate the rotation unit 160 at 5 Hz.

Referring to FIG. 13, the light emitted from the light source 110 can form a movement path from 1) to 6).

1) Light emitted from the light source 110 passes through the hole of the first mirror 120 and moves to the second mirror 130 in a linear path. Light emitted from the light source 110 may be collimated through a collimator. The collimator makes the incident rays parallel.

2) The light reflected from the moving second mirror 130 moves to the object 2 according to the angle of the second mirror 130.

3) The light reflected from the object 2 moves to the second mirror 130 in a linear path.

4) The light reflected from the moving second mirror 130 moves to the first mirror 120.

5) The light collected in the first mirror 120 moves to the reflector 150.

6) The light reflected from the reflector 150 moves to the photodiode 140 in a straight path.

13, the light source and the photodiode are positioned adjacent to each other by disposing the first mirror hole, the second mirror, the reflector, and the photodiode so as to adjust the movement path of the light, so that the light transceiver 100 and the rotating body The size of the rotating body 160 can be minimized and the turning radius of the rotating body 160 can be minimized.

FIGS. 14 and 15 are diagrams illustrating a movement method of a movable mirror of an optical transceiver. FIG.

Referring to FIG. 14, the second mirror 130, which is a movable mirror, can be moved using electromagnetic force generated by a magnet and a coil. There is a soft hinge on the center of the mirror and a permanent magnet on both ends. Position the coil at the back end (or near end) of the reflective surface. The second mirror is shaken when the current is periodically turned on the coil. Here, since the force generated by the magnet and the coil is small, in order to move the second mirror 130 or the normal 135 of the second mirror 130 at a high frequency, it is necessary to use a material capable of moving the hinge smoothly . The stronger the tension of the hinge, the more the second mirror 130 can move with less force, but it is difficult to make high frequency movement.

It is preferable that the movable mirror use a MEMS (MEMS) technology or an ultrasonic motor (Piezo Motor), and operate in a structure as shown in FIG. 7 in consideration of a pseudo ratio.

The structure shown in FIG. 14 may have a low vertical scanning speed, and a method of rotating the polyhedron by connecting the polyhedron to the motor may be applied. In FIG. 15, the second mirror 130, which is a movable mirror, is formed of a polygonal column, and a structure is shown in which the rotating shaft can be rotated and moved. The optical transceiver can adjust the slope of the normal of the reflection surface at different periodic points by adjusting the relationship between the rotation speed of the second mirror 130 and the output speed of the light source.

The laser device 10 adjusts the vertical scanning movement of the second mirror through a control unit implemented by an FPGA or the like. The control unit periodically transmits +/- signals to swing the second mirror. If the signal is a periodic waveform, the angle of the mirror is constant according to the periodic point of view. If necessary, you can mount the PSD sensor on the back of the mirror to measure the angle.

The lidar device 10 adjusts the horizontal rotation movement through a control unit implemented by an FPGA or the like. The control unit controls the rotation speed of the rotation unit and measures the rotation angle through the encoder inside or outside the rotation unit.

The control unit for controlling the vertical scanning movement of the second mirror and the control unit for controlling the horizontal rotation movement may be implemented as independent modules.

The distance measuring device 300 receives a vertical angle from a control unit that controls the vertical scanning movement of the second mirror and receives a horizontal angle from a control unit that controls the horizontal rotation movement and stores the vertical angle and the horizontal angle.

The lidar apparatus 10 receives the light emitted from the light source at the photodiode, and calculates the flight period ToF. The lidar device 10 transmits the vertical angle, the horizontal angle, and the flight time to the host via the interface. The flight duration can be calibrated or calibrated. The lidar device 10 may perform filtering to remove noise for at least one of vertical angle, horizontal angle, and flight time, and then transmit data to the host.

<Point cloud data>

16 and 17 are views of point cloud data generated by a moving object according to embodiments of the present invention. The location unit of the point cloud data is the meter.

The point cloud data acquired using the optical transceiver having the structure as shown in Fig. 13 is shown in Fig. (i) setting the normal of the movable mirror to -55 degrees relative to the horizontal plane of the ground to set the movable mirror to be inclined at 45 degrees, (ii) rotating the rotating part at 5 Hz, and (iii) Setting the output speed of the light source or the sampling rate of the ladder device such that it swings at a 10-degree angle and oscillates at 200 Hz, and (iv) obtains 10K points per second, The point cloud data of FIG.

An application that must measure the bottom surface together can acquire the point cloud data as shown in Fig. (ii) the rotating part is set to rotate at 10 Hz, (iii) the movable mirror is swung at an angle of +/- 10 degrees, and the normal of the movable mirror is set at -55 degrees relative to the horizontal plane of the ground, (Iv) setting the output speed of the light source or the sampling rate of the ladder device to obtain 10 K points per second, and (v) setting the height of the optical transceiver to 0.1 meter, Dimensional point cloud data shown in Fig.

The values set by the point adjusting unit 112, the reference angle adjusting unit 132, the moving angle adjusting unit 134, the vertical frequency adjusting unit 136, and the horizontal frequency adjusting unit 162 may be adjusted Numerical values can be used.

<Signal Conversion Unit of Optical Transceiver>

18 to 21 are views showing signal conversion units of an optical transceiver of a radar apparatus according to embodiments of the present invention.

The electrical signal output from the photodiode not only contains noise according to the circuit characteristics of the photodiode, but also has a variety of signal sizes, and the signal output time is not uniform.

There is a problem that the radar must wait until the optical carrier of the photodiode disappears even after the reflected laser signal passes through the photodiode. In Fig. 18, an electric signal outputted from the photodiode is illustrated. As shown in Fig. 18, the electric signal output from the photodiode requires a considerable time until the signal disappears. In particular, as the size of the output signal increases, the time required until the signal disappears increases.

The optical transceiver 100 may be implemented in a mobile or Lada device or may operate independently.

FIG. 19 is a block diagram illustrating an optical transceiver, FIG. 20 is a circuit diagram illustrating an optical transceiver, and FIG. 21 is a diagram illustrating signals output from the optical transceiver.

19, the optical transceiver 100 includes a light source 110, a photodiode 140, and a signal conversion unit 150. [ Optical transceiver 100 may omit some of the various components illustrated in FIG. 19, or may additionally include other components. For example, a transimpedance amplifier may be connected to the signal conversion unit 150.

The light source 110 emits light to the object based on a predetermined sampling period. The sampling period can be set by the control unit of the lidar apparatus 10. The sampling period is a period of time from when the optical transceiver 100 emits light, receives the reflected light, and converts the light into an electrical signal in accordance with the start control signal. The optical transceiver 100 may repeat these operations in the next sampling period.

The photodiode 140 receives the light reflected by the object and converts the light into an electrical signal. The photodiode 140 may be implemented as a PN junction photodiode, a PIN photodiode, an avalanche photodiode (APD), or the like. As shown in FIG. 1, the photodiode 140 outputs an electrical signal until the optical carrier disappears. In addition, as the size of the output signal increases, the time required until the signal disappears increases.

The signal converting unit 150 outputs the electric signal during the detection time in the sampling period so as not to be limited by the time of extinction of the output signal. Referring to FIG. 8, the signal conversion unit 150 may include a resistor 151, a switch 152, and a capacitor 153.

The resistor 151 is connected to the photodiode 140. One end of the resistor 151 is connected to the photodiode 140 and the other end of the resistor 151 is connected to the ground. The resistor 151 may be connected to the anode or the cathode of the photodiode 140.

If the resistance value is small, there is a problem that the waveform has a value other than 0 for a time similar to the time when light passes through the photodiode 140, but the output signal is small in size. Therefore, it is necessary to amplify the magnitude of the electric signal using a resistor having a value larger than a predetermined value for the resistor 151. [ In this case, as shown in FIG. 1, a phenomenon of trailing of a signal occurs.

And changes the propagation path of the electric signal through the switch 152 to solve the back drag phenomenon of the signal. The optical transceiver 100 can output a signal in which a part of the area where the magnitude of the electrical signal decreases is removed. Even if the rear end of the electric signal is removed, the radar apparatus 10 can measure the distance. The signal discriminator 200 detects the start point and the maximum point of magnitude of the electric signal and outputs the rising edge and the falling edge without detecting the end point of the electric signal.

The switch 152 is connected in parallel to the resistor 151 to change the propagation path of the electric signal. For example, the switch 152 may be implemented as a transistor or the like.

Referring to Figure 21, switch 152 is in off time (i) the sampling period (T _s) detected time (T _d) passing an electrical signal into a first path and, (ii) the sampling period (T _s) while in Lt; RTI ID = _0.0 &gt; ( _Tc ). &Lt; / RTI &gt; The first path is the path through which the signal is passed through the capacitor 153 and the second path is the path through which the signal is passed through the switch 152 to ground.

In the present embodiments, even if the electric decay time (T1, T2, T3) is required due to the trailing phenomenon of the electric signal outputted from the photodiode 140, the signal does not need to wait until the signal disappears, Can be processed.

The lidar device 10 adjusts the sampling period, calculates and sets an appropriate detection time according to the sampling period, and controls the ON / OFF operation of the switch 152. [ The control unit of the lidar apparatus 10 controls the sampling period, the detection time, the cutoff time, the waveform of the emitted light, the ON / OFF time interval of the light source, the pulse width of the start control signal, the pulse width of the stop control signal, The ON / OFF operation of the switch can be controlled with reference to the signal processing and the waiting time of the signal discriminator and the time calculator.

The capacitor 153 is connected to the point where the photodiode 140 and the resistor 151 are connected to output an electric signal. The capacitor 153 functions to remove the DC component of the electric signal. A non-inverting amplifier circuit may be connected to the rear end of the capacitor 153.

<Signal discriminator>

22 to 27 are diagrams illustrating a signal discriminator of a radar device according to embodiments of the present invention.

There is a method of detecting a zero by converting a signal by calculating a peak point of a reflected signal in the ladder. In reality, the reflected signal contains noise, and the point when the signal size becomes zero can not be used as the target point. An additional zero crossing detector should be implemented.

There is a method of using one threshold value as a method of calculating the viewpoint of the reflected signal in the lidar. This method has a problem in that it is difficult to calculate an accurate viewpoint because the signal form varies depending on the amount of reflected light. That is, a walk error occurs. Referring to FIG. 22, it can be easily understood that different time points (T ₁ , T ₂ ) are calculated depending on the signal form.

There is a method of using a plurality of threshold values in a method of calculating the viewpoint of the reflection signal in the case of the conventional method. However, this method has a problem that the circuit complexity increases because the output signal is fed back and the signal is delayed.

23 is a block diagram illustrating a signal discriminator.

23, the signal discriminator 200 includes a first conversion unit 210, a second conversion unit 220, and a signal detection unit 230. As shown in FIG. The signal discriminator 200 may omit some of the various elements illustrated in FIG. 7 or may further include other elements.

The signal discriminator 200 receives electrical signals from the photodiode 140 or the transimpedance amplifier. The received electrical signal, that is, the input signal, has a form of rising and falling by the reflected light. The signal discriminator 200 accurately measures a desired point in time with respect to an input signal and outputs an electric signal.

Referring to FIG. 24, the input signal has a front end time T _front , a target time point T ₁ , T ₂ , and a peak time point T _max , which meet the set threshold value, depending on the type of the input signal. The signal discriminator 200 performs a two-stage conversion process to detect a point of time that is closest to the _front end time T _front and the peak time T _max .

The first conversion unit 210 converts the input signal so that the signal point having the maximum signal size has a predetermined size. The first conversion unit 210 converts the signal point having the maximum signal size to zero. For example, the first conversion unit 210 differentiates an input signal or converts an input signal using a constant fraction discriminator (CFD). The constant fraction discrimination is a method of finding a point at which a point at which the signal delayed by the original signal is equal to a signal adjusted by a constant magnitude ratio becomes a certain ratio of the maximum magnitude.

In Fig. 25, a signal obtained by differentiating an input signal with respect to time is shown. Referring to Figure 25, the converted signal is the front end point (T _front), rise time meets the predetermined threshold (T _rising1, T _rising2), fall time meets the predetermined threshold (T _falling1, T _falling2), the rear end point (T _end ). The rear end point (T _end ) is the same point as the peak point (T _max ) of the signal before conversion. As shown in FIG. 25, when the first converter 210 converts the slope of the input signal so that the signal point having the maximum signal size has a predetermined size, the rising points (T _rising1 , T _rising2 ) T _front ) and the descending _timings (T _falling1 , T _falling2 ) are close to the rear end timing (T _end ).

If the signal is differentiated or the fractional discrimination method is applied to the signal, the dynamic range, which is the ratio of the jitter to the maximum signal amplitude to the minimum signal amplitude, may become narrow. Since the differential method is implemented by the RC circuit, the frequency characteristic of the signal changes according to the change of the distance, and a time error is generated. Since the slope of the signal is different, the charging time of the capacitor of the comparator is different and the response time of the comparator is changed to generate a time error. Therefore, it is necessary to convert the converted signal again.

The second conversion unit 220 adjusts the size of the converted input signal. The second converter amplifies the magnitude of the converted input signal by N (N is a natural number).

In Fig. 26, a signal obtained by amplifying the magnitude of the input signal with the slope converted is shown. The second conversion unit 220 when amplifying the signal the slope of conversion, the slope is so close to the vertical, the rise time (T _rising1, T _rising2) as shown in Figure 26 is the front end point (T _front) , And the falling _timings (T _falling1 , T _falling2 ) are closer to the rear end time (T _end ).

The present embodiment can accurately acquire the _front end timing T _front and the rear end timing T _end even if a circuit for comparing the noise included signal with a threshold value is implemented due to the two stage conversion process.

The signal detection unit 230 detects at least one time point having a predetermined reference size from the scaled input signal to generate an output signal. The signal detector 230 outputs a rising edge and a falling edge based on one threshold value from the scaled input signal. The stop control signal may be a pulse that matches the rising edge, a pulse that matches the falling edge, or a pulse that matches both the rising edge and the falling edge.

The radar apparatus 10 corrects the flight time using the pulse width corresponding to the rising edge and the falling edge.

<Time digital converter>

Figures 27-34 are diagrams of a time digital converter of a laddering device in accordance with embodiments of the present invention.

Examples of the method of measuring the flight time include a phase shifting method, an equivalent time sampling method, a direct measuring method using a high-resolution clock, and a time measuring method using a plurality of delay elements.

The phase deviation method transmits the sine wave continuously at the transmitter and measures the flight time using the phase deviation at the receiver. This method has a problem that the sampling rate is limited according to the cycle of the sine wave, and there is a problem of calculating the erroneous flight time by the crosstalk.

The equivalent time sampling method is a method that is applied to the oscilloscope and reads the signal repeatedly at a time difference to reconstruct the entire signal. This approach is limited to detecting obstacles moving at high speeds or using them for moving objects because of the low sampling rate.

A direct measurement method using a high resolution clock measures the flight time using a clock operating at several GHz. This approach can not physically raise the clock speed sufficiently, so there is a limit to improving the time resolution.

A time measurement method using a plurality of delay elements calculates a time difference using a time to digital converter (TDC). In FIG. 27, a conventional time-to-digital converter is shown, and in FIG. 28, a signal of a conventional time-to-digital converter is illustrated.

27, the buffer has a time delay of several tens to several hundred picoseconds (ps). When the flip-flop is operated using the stop signal, the time delay becomes equal to the number of flip-flops having a value of 1 in Fig. That is, the sum of the delay times has the same value as the flight time. This scheme is dependent on the time delay through the buffer, and there is a problem that the FPGA can not have a linear time resolution. In addition, since a large number of delay lines must be sequentially placed in the FPGA, there is a limitation in the number of spaces and devices to be implemented in the FPGA.

FIG. 29 is a view for explaining an operation of time-digital converter to measure time, FIG. 30 is a block diagram illustrating a time-digital converter, and FIG. 31 is a block diagram illustrating a ring oscillator of a time-to-digital converter.

The distance meter 300 uses a time digital converter to convert the two time differences into digital values.

Time digital converter is a device that converts time information into a digital code. The time digital converter generates a digital code corresponding to the time difference between the two input signals.

The input signal of the time digital converter may be in the form of a pulse of the same signal source, or may be the edge of another signal source. For example, the radar apparatus 10 can calculate the time difference based on the rising edge or the falling edge of the start control signal, the rising edge or the falling edge of the stop control signal.

29, the time-to-digital converter includes (i) a number (N ₁ , N ₂ ) counted by a coarse counter and a fine counter, and (ii) Time is measured using a small clock. The time difference between the large clock of the normal counter and the small clock of the precision counter determines the time resolution of the time digital converter.

Referring to FIG. 30, the time-to-digital converter includes a slow oscillator 510, a fast oscillator 520, a normal counter 530, a precision counter 540, and a phase detector 550.

The slow oscillator 510 generates the first clock by the start control signal. The fast oscillator 520 generates a second clock smaller than the first clock by the stop control signal. That is, the slow oscillator 510 generates a large clock, and the fast oscillator 520 generates a small clock.

The normal counter 530 is coupled to the slow oscillator 510 to count the first clock of the slow oscillator 510. Precision counter 540 is coupled to fast oscillator 520 to count the second clock of fast oscillator 520.

The phase detector 550 is connected to the slow oscillator 510 and the fast oscillator 520 and is connected to the normal counter 530 and the precision counter 540 to detect when the first clock and the second clock are synchronized .

The slow oscillator 510 and the fast oscillator 520 may be implemented with a ring oscillator. A ring oscillator is an oscillator in which an inverter and / or a buffer are connected in series in a loop circulation form. The slow oscillator 510 and the fast oscillator 520 may be comprised of a time delay element and a buffer. The time delay element can be realized by a digital element using an inverter or an analog element using a current source.

In FIG. 31, a ring oscillator in which three inverters are connected in series is exemplified. However, the present invention is not limited thereto, and a suitable combination of logic elements may be used according to the design to be implemented.

Conventional slow oscillators and fast oscillators adjust the clock width by adjusting the number of buffers. Conventional time digital converters have a resolution of about 80 picoseconds (ps) due to the signal delay time of the buffer itself.

These embodiments can adjust the clock width of the slow oscillator 510 by changing the position and signal path of the logic elements of the slow oscillator 510 on the circuit. The clock width of the fast oscillator 520 can be adjusted by changing the position and the signal path of the logic elements of the fast oscillator 520. That is, it is possible to change the slower oscillator 510 to operate faster and change the faster oscillator 520 more slowly. The position and routing path of each gate can be adjusted directly using the FPGA tool's Manual Gate Location adjustment function. Slow oscillator 510 and fast oscillator 520 can be combined into the same logic elements.

This embodiment can improve the difference between the clock width of the slow oscillator and the clock width of the fast oscillator, that is, the time resolution, by changing the position of the gates and the routing path of the signal on the circuit with slow oscillators and fast oscillators. The time-to-digital converter according to the present embodiment has a resolution of about 10 picoseconds (ps).

FIG. 32 is a block diagram illustrating a time-to-digital converter based on two stop control signals, and FIG. 33 is an exemplary implementation of a time-to-digital converter in an FPGA.

The signal discriminator 200 may output the first signal and the second signal. For example, the first signal may be a stop control signal along a rising edge and the second signal may be a stop control signal along a falling edge.

Referring to Fig. 32, a time-to-digital converter calculates a first time difference based on a start control signal and a first signal. The time digital converter calculates a second time difference based on the start control signal and the second signal. Since this embodiment treats the rising edge and the falling edge together, it can be designed by sharing a slow oscillator or a fast oscillator. That is, it is possible to calculate a first time difference and a second time difference by sharing a slow oscillator or sharing a fast oscillator on a circuit. The phase detectors 552 and 554 may be located closer than a predetermined distance from a shared slow oscillator or a shared fast oscillator, thereby improving time resolution.

In FIG. 33, a time-digital converter sharing a slow oscillator is illustrated. In order to process a rising edge and a falling edge, gates constituting three ring oscillators, two phase detectors, and four counters are connected As shown in Fig.

<Flight time correction>

34 is a block diagram illustrating a Lada device according to another embodiment of the present invention.

The time delay of each gate varies according to the temperature and the applied voltage of the FPGA. The frequency of the ring oscillator is changed due to the temperature and the applied voltage, and a slight deviation occurs in the resolution. This deviation causes an error in flight time.

The distance meter includes a reference clock counter 320 for counting the reference clock received from the external clock generator 30 and counting the internal clock generated in the internal oscillator, in order to monitor and correct the error of the flight time.

The clock generator 30 may be implemented with a high precision crystal oscillator. The high-precision crystal oscillator transfers the generated clock to the reference clock counter 320. The reference clock counter 320 stores the number of clocks calculated through the ring oscillator in the internal buffer every time the clock input is received. The Ladder device 10 reads the number of clocks stored periodically and corrects the flight time. The algorithm to correct flight time is as follows.

The distance measurer corrects the flight time using the ratio of the number of reference clocks and the number of internal clocks. ticks_per_x_crystal_clock is calculated in real time and stored at the calibration stage performed immediately prior to shipment of the implementation product. The correction of the flight time using the ratio of the number of stored reference clocks and the number of internal clocks is expressed by Equation (1).

The correction factor alpha is a value obtained by dividing ticks_per_x_crystal_clock_runtime by ticks_per_x_crystal_clock_stored_during_calibration_process.

13, the Lidar device 10 includes a plurality of time digital converters 310 and 312, and the reference clock counter 320 is located between the plurality of time digital converters 310 and 312. [ The clock generator 30 that generates the reference clock may be included in the mobile 1. [ Because the reference clock counter 320 is in contact with or located within a predetermined distance, it has approximately the same temperature and voltage specifications as the oscillator used in the time-to-digital converter. The internal clock also varies with temperature and voltage. Since the present embodiment does not use the internal clock in a divided manner, it has an effect of maintaining high accuracy.

FIG. 35 is a diagram for explaining a time correction operation of the Lada apparatus according to another embodiment of the present invention. FIG.

If the differential amplifier implemented by the RC circuit is applied in the process of converting the slope of the signal by the ladder device 10, the frequency characteristic of the signal changes according to the change of the distance, and a time error occurs. When a certain fraction discrimination method is applied in the process of converting the slope of the signal, the slope of the signal is different, so that the charging time of the capacitor of the comparator becomes different, and the response time of the comparator changes. Therefore, the radar apparatus 10 performs a process of correcting the time error.

The distance measuring device 300 corrects the flight time using the pulse width of the stop control signal. Since the output signal of a general photodiode has a large change in the pulse width, there is a problem that it is difficult to use it unless the pulse width to work error matches one to N and is not in a close range. Since the present embodiment has undergone a process of converting a signal, the relationship between the pulse width and the work error can be simply modeled.

The distance measurer 300 models a function between the work error and the pulse width, and measures the correction factor in advance. The correction factor according to the pulse width is shown in Fig. As shown in FIG. 14, the distance measuring device 300 applies a correction factor inversely proportional to the pulse width to correct the flight time. Since the intensity of the reflected signal is weak and the pulse width is narrowed, the work error becomes large, so that the distance measuring device 300 sets the correction factor to be large. If the intensity of the reflected signal is strong and the pulse width is widened, the work error becomes small. Therefore, the distance measuring device 300 sets the correction factor small.

The relational expression concerning the flight time is expressed by Equation (2).

In Equation (2), t _tof is the corrected flight time, and t _falling is the flight time before correction. The flight time is the time difference between the stop control signal and the start control signal. The ladder device can calculate the time difference based on the rising edge or the falling edge of the start control signal, the rising edge or the falling edge of the stop control signal. f _comp is a function of pulse width versus work error, and t _pulse is the pulse width of the signal. The radar apparatus can calculate the pulse width based on the rising edge or the falling edge of the stop control signal.

<Artificial marker recognition>

FIGS. 36 to 51 are views for explaining an operation of recognizing artificial landmarks according to the embodiments of the present invention.

Moving objects that move in space use markers (lankmarks) to locate themselves or other objects or to recognize other objects. The markers are divided into natural markers and artificial markers. The method of using a natural marker is a method of extracting a specific component from a structure, a ceiling, and a wall of a space, and the method using an artificial marker is a method of attaching a pattern or tag having a specific meaning to a space or another object, Is an artificial marker recognition method.

Unlike natural markers, artificial markers have various problems. It induces cosmetic rejection from the standpoint of the general user. In order to increase the recognition rate of artificial landmarks, the artificial landmarks must be made over a certain size, because of direct reflection of light, dark environment, or error in distance. Due to the unfavorable shape of the space, the position where attachment is possible is restricted in the space where the appearance is emphasized. For producers, a separate camera for artificial marker recognition should be used.

38 is a block diagram illustrating an artificial marker detector.

38, the lidar apparatus 10 includes an optical transceiver 100, a signal discriminator 200, a distance measuring instrument 300, a luminous intensity measuring instrument 400, and an artificial landmark detector 500 . The lidar device 10 may omit some of the various components illustrated in FIG. 38 or may additionally include other components. For example, the lidar apparatus 10 may further include a position estimator.

The lidar apparatus 10 does not have a separate camera for recognizing the artificial landmarks but analyzes the points scanned by the object to recognize artificial landmarks that are not visible to the user.

The optical transceiver 100 emits light to a target object, receives the light reflected from the target object, and converts the received light into an electric signal. The optical transceiver may include a light source, a transmission optical section, a reception optical section, and a photodiode. In photodiodes, when the light of photon energy strikes a diode, the principle that electrons act because electrons and positively charged holes are generated can be applied. The photodiode may be implemented as a PN junction photodiode, a PIN photodiode, an avalanche photodiode (APD), or the like.

The signal discriminator 200 accurately measures a desired point in time with respect to an input signal and outputs an electric signal. The signal discriminator 200 detects at least one time point having a predetermined reference size from the scaled input signal to generate an output signal.

The distance measuring unit 300 measures the distance between the points of the object using the time difference between the emitted light and the received time. The distance measurer 300 may include a time to digital converter. The time digital converter generates a digital code corresponding to the time difference between the two input signals.

The distance measuring device 300 may operate in a time of flight (TOF) manner. The time of flight method measures the distance from the object to be measured by measuring the time that the laser emits a pulse or a square wave signal to arrive at the receiver with reflected pulses or square wave signals from objects within the measurement range.

The luminosity meter 400 measures the intensity of the light reflected at the points of the object. The intensity meter 400 can measure the intensity of the electric signal received from the optical transceiver 100 or the signal discriminator 200.

The artificial marker detector 500 analyzes the change of the distance of the points of the object and (ii) the brightness of the points of the object to determine whether or not the object is an artificial marker, and analyzes the invisible barcode included in the artificial marker Generate digital code.

Artificial markers recognized by the artificial marker detector 500 include a base and an invisible bar code. The invisible bar code includes a code sequence formed of a material that reflects, absorbs, or emits light in the first wavelength band. The code sequence is hidden by the material that reflects, absorbs, or emits light in the second wavelength band and becomes invisible.

The position calculator calculates the relative position between the lidar device and the artificial landmarks.

FIG. 39 shows an artificial marker 51 having a white appearance as a result of visual observation, an artificial marker 52 having a black appearance as a result of visual observation, an artificial marker 53 having discontinuous stripes identified by the artificial marker detector 500, Respectively. When a whitened artificial landmark 51 is attached to a structure or space having a white background or an artificial landmark 52 having a black background attached to a structure or space having a black background, the user recognizes the artificial landmarks 51 and 52 Can not. On the other hand, an artificial landmark detector 500 that receives light of a specific wavelength band can identify stripes of the artificial landmark 53. [

40 to 43, artificial markers 51, 52 and 53 include bases 610a, 610b, 610c and 610d and invisible barcodes 620a, 620b, 620c and 620d. The artificial indicia 51, 52, 53 may further include covers 630c, 630d, an adhesive, or a combination thereof.

The invisible bar code is connected to the base and has a code sequence formed of a material that reflects, absorbs, or emits light in the first wavelength band. The code sequence is hidden by the material that reflects, absorbs, or emits light in the second wavelength band and becomes invisible. The first wavelength band may be a wavelength band of infrared rays and the second wavelength band may be a wavelength band of visible light rays. However, the present invention is not limited thereto.

The code sequence has bright regions 622a, 622b, 622c, 622d and dark regions 624a, 624b, 624c, 624d based on the difference in reflectance, absorption, or luminescence rate for light in the first wavelength band.

The code sequence may be divided into a start pattern, an information pattern, a stop pattern, an error correction pattern, or a combination thereof using (i) the number or width of light domain, (ii) the number or width of dark regions, May be a one-dimensional sequence represented by a combination.

The bright region and the dark region of the code sequence can be distinguished by the reflection wavelength of light in the first wavelength band, the absorption wavelength, the emission wavelength, the reflectance, the absorption rate, the emission rate, the reflection angle, the hue, the surface property, or a combination thereof.

The code sequence is invisible by the reflection wavelength of the light in the second wavelength band, the absorption wavelength, the emission wavelength, the reflectance, the absorption rate, the emission rate, the reflection angle, the hue, the surface property, or a combination thereof.

Referring to FIG. 40, the base may include a material that reflects, absorbs, or emits light in the second wavelength band. For example, an omnidirectional reflective material or paint may be patterned on a white paper to form a code sequence.

Referring to FIG. 41, the code sequence may include a material that reflects, absorbs, or emits light in the second wavelength band. For example, a substance or paint having a different reflectance may be patterned to form a code sequence.

Referring to Figs. 42 and 43, the covers 630c and 630d are located on the transmission path of light in the first wavelength band. The covers 630c and 630d include a material that reflects, absorbs, or emits light in the second wavelength band. The covers 630c and 630d can filter a specific wavelength band. For example, an infrared filter can be used. Since the infrared filter is dark red, the user will recognize it as a device that is simply black rather than striped.

The artificial marker may further include an adhesive portion (not shown) having an adhesive layer formed on the back surface of the base.

Hereinafter, with reference to FIG. 44 to FIG. 46, the operation of recognizing the artificial landmarks will be described.

In step S2110, the artificial landmark detector segments the candidate region in which the luminances of the points of the object discontinuously change. The discontinuity is that the distance of the points of the object is within a certain range, the deviation of the luminosity of the points is over a certain range, some of the luminances of the points are smaller than the minimum value, others are larger than the maximum value and others are smaller than the minimum value Another part means repeating something larger than the maximum value. For example, the candidate region has an area similar to the stripe of the artificial marker 53. [

In step S2120, the artificial landmark detector normalizes the luminous intensity with respect to the distances of the points of the object based on the predetermined reference distance. For example, normalize with respect to x meters, and the reference distance may be a suitable number depending on the design being implemented.

In step S2130, the artificial landmark detector divides the normalized luminous intensity into a bright region and a dark region, and determines whether the bright region and dark region of the candidate region are code sequences. The artificial landmark detector can be divided into a bright region and a dark region using a moving average that continuously calculates the average of two or more consecutive data. The artificial marker detector can use local data to determine the relative magnitude of the surroundings.

In step S2130, a criterion for determining whether the artificial landmark detector is a code sequence is determined according to the design being implemented. For example, the artificial landmark detector may be configured to detect (i) the distance between the first dark region and the second dark region, (ii) the last dark region and the second dark region from the end, (iii) Or (iv) combinations thereof. In Fig. 8A, an artificial landmark attached to a bent face at a certain angle is shown, and an artificial landmark attached to a curved face is shown in Fig. 8B. Since the distance value is different for each point, the moving object can be docked on the basis of the vertex or the like or can be brought close to a desired position precisely. If artificial markers are attached to more than one plane, false positives or Type I errors can be minimized.

In step S2140, the artificial landmark detector converts the code sequence into a binary code to extract identification information. The criterion that the artificial landmark detector converts to the binary code applies the appropriate conversion criterion according to the design being implemented. For example, the artificial marker detector may read the luminous intensity in each of the left and right directions at regular intervals based on the center, and if the luminous intensity is larger than the threshold value, it is set to 1, and if it is smaller, the binary code can be generated.

Hereinafter, with reference to FIG. 47 and FIG. 48, the operation of calculating the relative position by the radar apparatus will be described. The lidar device uses the attitude information and the distance information to determine the position of the artificial marker.

In step S2210, the position calculator line-fitting the points of the recognized artificial marker in one direction. The position estimator can calculate the linear equation through line fitting. Line fitting You can calibrate one point and correct the angle based on the horizontal or vertical line. The position calculator can correct the position of the points of the artificial marker through pattern matching.

In step S2220, the position calculator calculates a relative angle between (i) the line-fitting points and the line virtually extending between the ladder device and (ii) the line extending virtually in the traveling direction of the device. The position calculator compares the angles of the rotating optical transceivers to calculate the relative angle. In Fig. 10, the relative angle [theta] is shown.

In step S2230, the position estimator calculates the line-fitting points and the relative distance between the line devices. In Fig. 10, there is a relative vector V, V _1, V ₂ is shown.

In step S2240, the position calculator calculates the relative position based on the relative angle and the relative distance. For example, the distance can be calculated using a trigonometric function or the like.

<Docking station>

The moving object provided with the TOF-based 360 degree apparatus according to the embodiment of the present invention can be configured as follows. A docking station is made of a material with a high reflectance and an artificial marker with an invisible bar code is applied.

As shown in FIG. 50, a docking station is formed of a material having a high reflectivity on one side of a wall. At this time, the docking station 450 may be configured by applying various patterns.

As shown in FIG. 51, since the reflectivity of the docking portion is high, the position of the docking station can be found. Since the output information of the TOF lidar device is distance, the exact position of the docking station can be known, thereby improving the docking performance.

In the intensity graph measured by the laser receiver mounted on the moving object 1, the docking station can be located because the reflectivity of the docking area is high as shown in FIG.

Since the output information of the TOF-based range finder indicates the distance value, the accurate position of the docking station can be known, thereby improving the docking performance.

&Lt; Interaction between plural mobile bodies >

52 is a view illustrating how moving objects interact according to embodiments of the present invention.

A plurality of small mobile robots are applied to distinguish cleaning spheres by communicating with mobile robots performing cleaning for a wide space such as a building. Mobile robots recognize artificial markers with invisible barcodes attached to the body. Conventional building cleaning robots are very large and expensive, but they are not only slow to clean, but also difficult to maintain. Accordingly, the moving body according to the embodiment of the present invention can clean a wide area by applying a plurality of cleaning robots of the size of a conventional household cleaning robot.

(Current, impact amount, attitude information, and the like) for each operation performed according to the case where the mobile object moves up the slope or rises on the threshold or the carpet according to an embodiment of the present invention, Make sure to distinguish whether you are moving this slope or moving the threshold.

<Location recognition and map generation>

Figs. 53 to 64 are diagrams for explaining an operation in which the Lada device according to the embodiments of the present invention recognizes a position and generates a map. Fig.

In order to move a moving object such as a robot or a vehicle in an unknown environment, there is no information about the surrounding environment. Therefore, a map of the environment should be created using the sensor information, and the current position of the moving object should be estimated from the created map. Simultaneous Localization and Mapping (SLAM) is a method of recognizing such a location and creating a map of the surrounding environment.

FIG. 53 is a graphical model of information processed by the Simultaneous Localization And Mapping (SLAM) method. 53, x denotes the position of the robot, u denotes odometry information, z denotes an observation value, and m denotes an estimation map. There are various methods for SLAM such as filter-based SLAM and graph-based SLAM.

The graph-based SLAM represents the robot's position and movement with graph nodes and edges. A node is the location of a robot or artificial marker at a particular point in time. An edge is a relationship between two nodes, which means space constraint between two nodes. The measured edges include errors. Therefore, as the travel distance of the robot becomes longer or the number of nodes increases, there arises a problem that errors are accumulated in the process of estimating the position.

54 and 55 are diagrams illustrating a map generator.

54 and 55, the mobile device 1 includes a ladder device 10, a mobile device 20, a path controller 30, and a map generator 50. As shown in Fig. The map generator 50 includes an area map estimating unit 710, and a global map estimating unit 720.

The map generator 50 expresses the position (Pose) where the scanning is performed as a node and calculates the relative position between the nodes (Relative Pose). The map generator 50 updates the map by optimizing the position of each node according to the relative position between the nodes. The map generator 50 finds a node value that minimally satisfies the constraint conditions or a node value that minimally satisfies the constraints, and corrects the errors at the positions of the nodes to optimize the position of the node. That is, the nodes converge to the optimal node value based on the constraint.

The map generator 50 estimates the current position using the sensor data obtained in the nearby area when estimating the current position. The scanning period and the key frame generation period are set to appropriate values according to the design to be implemented and can be set differently as needed.

The lidar apparatus 10 acquires scan information of a space where the mobile robot is located. The lidar apparatus 10 can acquire omnidirectional information or distance information of a partial area using a Lidar sensor. The map generator 50 can additionally obtain the odometry information of the mobile robot together with the distance information. The map generator can acquire odometry information such as the number of revolutions, the tilt, and the amount of rotation from an encoder (IMU) connected to the mobile device of the mobile robot or an IMU (Inertial Measurement Unit). The IMU can be implemented as an acceleration sensor and a gyro sensor.

The position of the mobile robot can be expressed by a three-dimensional vector. The three-dimensional vector can be represented by an X-coordinate and a Y-coordinate from the origin of the reference coordinate system, and an angle formed by the X-axis of the robot coordinate system and the X-axis of the reference coordinate system.

The area map estimating unit 710 generates a key frame for the node using the scan information, and calculates an odometry edge between consecutive nodes. Observations between nodes have uncertainty. The uncertainty of the odometry edge can be expressed as a covariance matrix.

The area map estimating unit 710 stores a key frame among a plurality of scan information acquired for each scanning period. The scan information may be represented by a point cloud. The key frame includes distance information and time information. If there is no registered key frame, the area map estimating unit 710 generates a key frame using the currently inputted distance information, and then updates the key frame to estimate the area map. The area map estimating unit 710 can improve the accuracy by performing scan matching based on the area map including a plurality of pieces of scan information.

The area map estimating unit 710 calculates the odometry edge by rotating or moving the scan information and performing scan matching or measuring the odometry information of the mobile robot. The local map estimator 710 corrects the error between the predicted edge of the odd metric and the measured odd metric edge based on the estimated value of the node through scan matching.

Global map estimator 720 detects loop closure edges between nodes that are not consecutive with respect to the updated set of key frames. When the mobile robot visits the visited area again, the error of the nodes can be corrected more accurately. Unlike the odometry edge between temporally consecutive nodes, the loop closure edge forms a spatial relationship between nodes that are not contiguous in time. However, it is inefficient to generate a loop closure edge between all edges, so it is necessary to detect an appropriate loop closure edge in consideration of constraints.

The global map estimator 720 estimates the global map by correcting the position of the node based on the odometry edge and the loop closure edge. The global map estimator 720 combines the regional map to expand and update the global map.

An interface is a communication path for transmitting and receiving information to and from other devices. Other devices may establish parameters by accessing the map generator 50 via the interface. The map generator 50 can transmit the measurement location and the map to another device via the interface.

Hereinafter, referring to FIG. 56, the operation of the map generator for scan matching will be described. FIG. 56 is a diagram illustrating data that is scanned and matched by the map generator. FIG.

p is the position of the node according to the time and position change, and e is the spatial relationship between the nodes. The scan information acquired by the map generator includes distance information from the fixed structure.

When the scan information before and after movement of the robot is compared, the odometer errors are accumulated so that the common parts do not overlap. The accurate position of the robot can be estimated by correcting the odometer error through scan matching that rotates or moves the scan information and calculates the probability of the observed value.

Hereinafter, the operation of filtering by the map generator will be described with reference to FIG. 57 is a diagram illustrating data filtered by a map generator.

Unlike the fixed structure, the scan information acquired at a specific time or at a specific location may include noise such as a moving obstacle, and the area map that the mobile robot moves and gradually updates may be changed to an actual structure or another structure. That is, when the scan information acquired at different time periods at the same position due to the moving obstacle is scanned and matched, an error occurs in the estimated map.

The area map estimating unit identifies moving obstacles and fixed obstacles from a plurality of scan information acquired in a predetermined time interval, compares newly acquired scan information, and updates the area map by removing moving obstacles from the key frame . Recently, the area map is updated by removing noise using N scan information. The noise may be point cloud data that has been shifted by a predetermined movement distance or rotated by a predetermined rotation angle.

The local map estimator can filter the key frame if the first condition regarding the moving distance of the mobile robot is satisfied in order to reduce calculation load and obtain meaningful information. The first condition is set based on the travel distance. For example, the local map estimator performs filtering when moving beyond a certain distance from the reference point.

Hereinafter, the operation of estimating the global map by the map generator will be described. FIG. 58 is a flowchart illustrating an operation of the map generator to estimate a global map, and FIGS. 59 to 63 illustrate data processed by the map generator.

The global map estimator estimates a global map for a set of updated key frames if the second condition regarding the movement distance of the mobile robot or the change of the surrounding environment is satisfied. The second condition is set based on a change in the moving distance or the surrounding environment. For example, the second condition may be set to travel for a certain distance or more, to depart from a certain distance range, or to set the matching rate of the surrounding structures out of a certain range.

The N scans in recent years are all filtered data and have data relating to the fixed structure. The global map estimator stores the filtered key frames.

In step S2410, the global map estimator corrects the error of each key frame using the uncertainty propagation on the basis of the specified key frame, and determines the position of the key frame using the Mahalanobis distance The first candidate group related to the key frame is extracted. The Mahalanobis distance is a value indicating how many times the distance from the mean is the standard deviation. It is a numerical value that indicates which value is hard to generate or how strange it is. Figure 9 of a first candidate node extracted by the node (p _n) on a key frame into the last _{(p k-1, p k} , p k + 1, p m-1, p m, p m ₊₁ ) is shown.

In step S2420, the global map estimator compares the key frame of the first candidate group and the key frame obtained in the predetermined time interval, and extracts a second candidate group related to the key frame matching the predetermined probability range. In Figure 10 the nodes on the last N number of frame keys _{(p n, p n -1,} p n - 2) the nodes of the second candidate is extracted based on _{(p k, p m-1} , p a _m, p _{m +1} ) is shown.

In step S2430, the global map estimator extracts a third candidate group related to the key frame matching the predetermined consistency range from the second candidate group using the consistency check algorithm. The global map estimator rechecks the mis-matched candidates. For example, a maximum clique or a single cluster graph partitioning technique may be applied as the consistency check algorithm. Also of a third candidate node extracted in _{61 (p m-1, p} m, p m + 1) is shown.

In step S2440, the global map estimator detects a loop closure edge between nodes that are not consecutive with respect to the key frame belonging to the third candidate group. The global map estimator forms a spatial relationship between a recent node and a meaningful node. In Fig. 12, loop closure edges (e _{n, m} ) between non-contiguous nodes (p _n , p _m ) are shown.

In step S2450, the global map estimator corrects the error of the odometry edge and the error of the loop closure edge using the graph optimization technique. The global map estimator finds a node value that minimally satisfies the constraints or minimizes the constraints and optimizes the position of the node by correcting the errors at the positions of the nodes. That is, the nodes converge to the optimal node value based on the constraint. For example, the error can be corrected in such a manner as to calculate the minimum value of the weighted squared sum of the error vectors by a plurality of odometry edges and a plurality of loop closure edges.

The global map estimator may output a global map including a plurality of area maps.

FIG. 64 is a flowchart illustrating a method of recognizing and generating a position according to another embodiment of the present invention. The location recognition and mapping method of the mobile robot can be performed by the map generator of the mobile robot.

In step S2510, the map generator acquires scan information of a space where the mobile robot is located. In step S2510 of acquiring scan information, distance information of all directions or a partial area may be obtained using a Lydia sensor. The step of acquiring the scan information (S2510) may further acquire the odometry information of the mobile robot together with the distance information.

In step S2520, the map generator generates a key frame related to the node using the scan information, calculates an odd metric edge between consecutive nodes, and updates the key frame to estimate the area map.

In step S2520, the scan map information is rotated or moved to scan-match or the odometry information of the mobile robot is measured to calculate the odometry edge. The step of estimating the local map (S2520) corrects the error between the predicted odd edge and the measured odd edge based on the estimated value of the node through scan matching.

The step of estimating the local map (S2520) filters the key frame if the first condition regarding the moving distance of the mobile robot is satisfied. The step of estimating the local map (S2520) comprises the steps of discriminating the moving obstacle and the fixed obstacle from the plurality of scan information acquired in the predetermined time interval, comparing the newly obtained scanning information, And updates the area map.

In step S2530, the map generator detects a loop closure edge between nodes that are not consecutive with respect to the set of updated key frames, and estimates the global map by correcting the position of the node based on the edge map and the loop closure edge.

The step of estimating the global map (S2530) estimates the global map of the set of updated key frames when the moving distance of the mobile robot or the second condition regarding the change of the surrounding environment is satisfied.

The step of estimating the global map (S2530) includes correcting the error of each key frame using the uncertainty propagation based on the specified key frame, and determining a key frame located within a predetermined distance using the Mahalanobis distance 1 candidate group is extracted.

The global map estimation step S2530 compares the key frame of the first candidate group and the key frame acquired in the predetermined time interval to extract a second candidate group related to the key frame matching the predetermined probability range.

In step S2530 of estimating the global map, a third candidate group related to the key frame matching the predetermined consistency range is extracted from the second candidate group using the consistency check algorithm. The map generator generates a loop closure edge for at least one of the nodes belonging to the first to third candidate groups.

The step of estimating the global map (S2530) corrects the error of the odometry edge and the error of the loop closure edge using the graph optimization technique.

<Path control of the Lidar device>

65-75 are views of a path controller of a laddering device according to embodiments of the present invention.

Even if the cleaning robot fuses the information obtained by these sensors, it is possible to distinguish the material of the floor such as a carpet or to detect the wall, and to distinguish the height of the floor such as the inclination, There is a problem that it is impossible to detect low obstacles such as excrement, threshold, and the like.

Hereinafter, a cleaning robot according to an embodiment of the present invention will be described with reference to a point cloud based on a bottom surface state, in particular, an elevation state of a bottom surface or an algorithm for detecting a low obstacle.

The path controller 30 may further include a floor condition determination unit 810, an obstacle detection unit 820, a slip determination unit 830, or a combination thereof.

66 to 68 are conceptual diagrams showing point cloud data obtained in different forms according to the state of the floor surface.

The moving object 1 can acquire information on the surrounding environment as three-dimensional information by using a sensor capable of acquiring three-dimensional information of an object such as a LiDAR sensor and a TOF (Time Of Flight) sensor. When each point on the bottom surface is defined as a point 1110, the moving object 1 can acquire information about each point 1110 as three-dimensional information using a Lidar sensor, a TOF sensor, or the like.

<Determination of floor condition>

When the information about each point 1110 is obtained as three-dimensional information, it becomes possible to distinguish the state of the floor surface, for example, the height of the floor surface, from the information about the point cloud that is a set of these points 1110 . In the present invention, the moving object 1 measures the state of the floor, for example, the height of the floor on the basis of the point cloud data with respect to the floor.

As shown in Figs. 66 to 68, the portions where the elevation is formed on the floor surface in a home environment of a general home are a threshold, a ramp, a cliff, and the like. Fig. 66 is an example of a threshold, Fig. 67 is an example of a ramp, and Fig. 68 is an example of a cliff.

In order to improve the safe traveling and cleaning performance of the moving body 1, the moving body 1 must be able to distinguish the portions on the bottom face where the height is formed. However, since the information used to distinguish the height of the floor is very limited, it is not easy to distinguish the height of the floor when the situation is encountered.

For example, when the robot goes up a threshold or a ramp, the pitch of the robot is generated as the robot is lifted forward. This makes it difficult to distinguish whether the robot is climbing the threshold or climbing the ramp.

Also, when the front of the robot is heard when the robot climbs the threshold, the bottom sensor mounted on the robot is exposed. This may cause the robot to misunderstand the condition of the floor as a cliff.

In the present invention, the moving object 1 obtains information about the floor surface located in front of the point cloud data using a Lidar sensor, a TOF sensor, or the like. The point group data 1120, 1130 and 1140 will be different according to the state of the floor surface as shown in FIGS. 66 to 68. Accordingly, the moving object 1 calculates the point group data 1120, 1130 and 1140 based on the point cloud data 1120, State can be detected.

Also, the moving object 1 can detect a change in height of the floor surface based on the point cloud data 120, 130, and 140, and perform an appropriate control function according to the state of the floor surface.

<Obstacle detection>

When the TOF sensor is used as the sensor for acquiring the three-dimensional information of the object, the moving object 1 can acquire the three-dimensional information of the object using the TOF-based 360-degree distance measuring device.

When the moving object 1 uses the TOF-based 360-degree distance measuring device, not only the state of the floor but also the carpet detection, obstacle detection, and position estimation can be realized.

As described above, by using the Lada sensor, TOF sensor, or the like, the moving object 1 can reconstruct the surroundings such as the floor surface in three dimensions based on the point cloud data. The moving body 1 separates the information about the floor surface from the information about the reconstructed surrounding environment so that the floor surface material (tiles, floors, carpets, etc.) in addition to the height of the floor surface The presence of low obstacles (clothes, animal feces, blocks, etc.) scattered on the floor can be detected.

The lidar apparatus 10 can obtain information on the first points and information on the second points with information on each point in the room. The radar apparatus 10 may obtain information about the first points using at least one channel signal associated with the low beam. The lidar device 10 may obtain information about the second points using at least one channel signal associated with the high beam.

The obstacle detecting unit 820 detects an obstacle located in the room based on the information about the first points or / and the information about the second points related to the indoor space. The obstacle detecting unit 820 can easily detect low obstacles scattered on the floor based on the information about the bottom surface obtained from the point cloud data 1150, thereby avoiding traveling and cleaning.

The cleaning unit provided in the mobile unit performs the cleaning function for the room based on the condition of the floor and the information about the obstacle.

According to the present invention, if the mobile object 1 can perform an appropriate countermeasure against a low obstacle, the cleaning success rate and coverage performance of the mobile object 1 can be improved.

<Slip detection>

Next, a slip detection method of the mobile unit 1 will be described. In the following description, the mobile object 1 will be described as an object to which the slip detection method is applied. However, the slip detection method proposed in the present invention is not limited to this, and may be applied to any robot that drives using wheels.

The mobile unit 1 can detect a slip using a gyro sensor and a passive encoder, which is a front wheel recognition sensor. In this case, the moving object 1 detects the slip by comparing its rotational speed estimated from the encoder of both wheels with the gyro information, detects its own movement using a manual encoder, and detects its own speed To check whether or not the user actually moves.

However, there is a problem that the rotation of the passive encoder is different according to the state of the floor surface, and inaccurate information is provided in an environment in which there is a difference in the height of the running surface, such as an uneven region, a threshold, or a carpet.

It is also possible for the moving object 1 to detect the slip based on the optical flow of the infrared signal (or laser signal). In this case, the moving object 1 detects its own motion based on the flow of signals between the light emitting element and the light receiving element attached to the bottom surface, compares the motion information with the encoder information of the wheel, Or not.

However, since the signal is very sensitive to the reflectance of the bottom surface, there is a problem that slip detection is difficult when the bottom surface is uneven.

In order to solve such a problem, a method of detecting a slip of the moving object 1 based on a laser odometry will be described.

The laser odometry estimates the amount of movement of the moving object 1 by using the distance from the short-channel / multi-channel 360 degree laser distance measuring device to the surrounding obstacles, thereby calculating the movement amount of the moving object 1 estimated from the initial position And the current position is estimated. Here, the speed of the moving object 1 can be estimated by dividing the difference value between the surrounding obstacle information measured at the current point of time and the surrounding obstacle information obtained at the previous point of time by the time difference.

71 is a reference diagram for explaining a method of detecting a slip situation of a mobile body.

When the moving object slides from the point A to the point B, it is necessary to use the obstacles 1310 and 1320 located in the periphery of the moving object to find out the slip of the moving object. In the present invention, information on the obstacles 1310 and 1320 located in the periphery of the moving object is obtained by using a multi-channel (or short channel) laser distance measuring device.

72 and 73 are reference views showing distance information from the moving object to the obstacles 310 and 320 obtained at two different points. 72 shows distance information 1331, 1332, 1333, 1334 between the moving object and the obstacles 1310, 1320 measured at point A in FIG. 71, and FIG. 73 shows distance information 1331, 1332, 1333, 1334 between the moving object and obstacles 1300 measured at point B in FIG. (1341, 1342, 1343, 1344) between the mobile stations (1310, 1320).

The moving object can calculate its movement amount at two points by moving or rotating the information 1351, 1352, 1353, and 1354 about the obstacles 1310 and 1320 obtained at two different points.

In this embodiment, the moving object can match the information 1351, 1352, 1353, and 1354 about obstacles 1310 and 1320 obtained at two different points by applying a particle filter.

74 and 75 are reference views for explaining a method of matching obstacle information obtained at two different points.

74 shows a process of matching the obstacle information obtained at the point B with respect to the point A as a reference. The random particles 1360 are located within the particle boundary 1370 determined based on the point A and have information about randomly generated candidates such as position and attitude.

Assuming that the information on the moving object at point B is information possessed by a specific random particle, and matching the obstacle information obtained at two points based on the information of each random particle, You can find random particles that have similar information. Then, the information held by the random particle can be estimated as information on the moving object at point B.

75 shows a result of matching the obstacle information at points A and B based on the information held by the selected random particle 1380. [ 7, when the obstacle information is matched and the obstacles 1310 and 1320 are arranged at the same position, it is possible to know the relative position difference at each point of the moving object, and to detect the degree of sliding of the moving object It becomes possible.

Also, the current speed of the moving object can be calculated by dividing the detected information by the time difference (the difference value between the time at which the moving object is located at the point A and the time when the moving object is located at the point B).

Further, if the number of random particles 1360 is increased and a stochastic error with respect to the position of the moving object is taken into consideration, more accurate information can be predicted.

<Projection map generation>

Figures 76-83 are diagrams of a projection map generator of a laddering device in accordance with embodiments of the present invention.

Referring to Fig. 76, the moving body includes a ladder device 10, a moving device 20, a path controller 30, and a projection map generator 60. [ FIG. 77 is a flowchart of a method for detecting an obstacle in a multi-channel Lada-based mobile terminal according to an embodiment of the present invention.

The lidar apparatus 10 includes at least two optical transceivers whose emission angles are different from each other in order to detect three-dimensional environment information including obstacle information of a moving object.

The projection map generator 60 generates the three-dimensional environment information for defining obstacles in three dimensions based on the reception information of the at least two optical transceivers, and outputs the projection map including the three- . The 3D environment information includes information on the obstacle, and information on the obstacle can include information on the position, size, and shape of the obstacle. Also, the two-dimensional space map can be formed as a two-dimensional grid map in one form.

The path controller 30 is for generating an optimal path, determines a position on the space of the obstacle based on the projection map, and generates a traveling path for avoiding the obstacle. The path controller 30 can determine the size and shape of the obstacle when determining the position on the space of the obstacle, and can generate a traveling route for avoiding the obstacle by approaching the minimum distance without collision with the obstacle .

The projection map generator 60 analyzes the received signal of the ladder device 10 to generate shape information of an obstacle including a threshold, a slope, and a cliff, and expresses the three-dimensional environment information using a cubic shape of n ³ have.

The projection map generator 60 can determine the marking grid position and the grid marking value using each of the horizontal positional information and the vertical positional information of the obstacle obtained in the ladder device 10, A binary projection map obtained by changing the projection map to binary information of 1 or 0 can be Huff transformed to extract a linear region.

 The projection map generator 60 can increase the priority of the area that has not been cleaned compared with the cleaning map and increase the priority of the straight line near the distance.

The path controller 30 can estimate the presence or absence of an obstacle on the basis of the vertical distance from the at least two optical transceivers 101 and 102 to the bottom surface and the horizontal downward launch angle of the transmission signal of the optical transceiver.

The lidar apparatus 10 applied to the moving object 1 rotates a plurality of (two or more) laser foot / light receiving sets 101 and 102 within 360 degrees or within a certain range to acquire the obstacle information in the three dimensions (S2610). In the present embodiment, the three-dimensional environment information means three-dimensional information about the periphery of the moving object, and the three-dimensional environment information also includes information about the obstacle.

The mobile can distinguish low obstacles (small toys, clothing, etc.), vertical environments (threshold, slope, cliff), and plan a collision avoidance path based on 3D area map.

In general, low obstacles (for example, toy blocks, socks, clothes, etc.) may be interfered with the intake or the wheels of the cleaning robot, which must be considered for safe running of the cleaning robot. Nevertheless, existing infrared or ultrasonic based obstacle detection cleaning robots were unable to detect obstacles that were lower than the sensors' position.

In particular, the cleaning robot based on the short channel laser scan sensor, which has the best map generation and position estimation performance, can not detect low obstacles because the sensor position is set at the uppermost position of the robot to improve the SLAM performance.

In contrast to conventional cleaning robots, a multichannel LiDAR-based moving object 1 to which an embodiment of the present invention is applied can measure the bottom of the cleaning robot as well as horizontal obstacle detection, It is possible to detect low obstacles regardless of the height of the cleaning robot and the position of the sensor, as shown in FIG. 4, while maintaining excellent SLAM performance as in the cleaning robot.

The vertical environment is an important environment that greatly affects the running and positioning performance of the cleaning robot. Cleaning robots that measure obstacles horizontally (such as bumper-based, infrared, or ultrasound-based, single-channel laser scan-sensor based cleaning robots) can not ascertain the vertical environment in advance, And over. Climbing the threshold or climbing up a slope (such as a bar-chair fixture) will affect the location estimation performance of the cleaning robot and degrade cleaning performance. In addition, in case of a cliff, since the operation of the cleaning robot is no longer possible when the cleaning robot is removed, a floor measuring sensor is installed to prevent the cleaning robot from falling into the cliff. Even in the case of the floor measuring sensor, since it is mounted in the cleaning robot and only the situation immediately before the wheel is detected, it is impossible to improve the stability of the cleaning robot in advance when the cleaning robot rotates or moves quickly.

FIG. 78 shows lattice information when 3-dimensional environment information is represented by 1 byte (8 bits) according to an embodiment of the present invention. FIG. 79 and FIG. 80 show the height information FIG. 81 and FIG. 82 show a 1-byte projection map display result of the three-dimensional obstacle according to the embodiment of the present invention.

When the mobile unit 1 uses the multi-channel ladder unit 10, it is possible to acquire the obstacle information shown in Figs. 66 to 68 with respect to the vertical environment, and to plan the traveling path of the cleaning robot in advance have. It is therefore possible to detect and avoid the cleaning robot before it is confronted with a vertical environment.

Among the conventional cleaning robots, the cleaning robot based on the laser scanning sensor can detect the obstacles in a state in which the obstacle information of the surroundings is kept at a distance (non-confronted state). Therefore, . However, as mentioned earlier, there is still a risk of collision with obstacles, since low obstacles can not be detected and vertical environments can not be distinguished.

Since the mobile body 1 to which the multi-channel lidar apparatus 10 is applied provides three-dimensional information about the surroundings, it is possible to avoid collision between low obstacles and obstacles affecting the safe running of the robot including vertical environments It is possible to plan the route before facing obstacles.

The present invention proposes a projective map for displaying the surrounding three-dimensional environment information of the cleaning robot obtained through the multi-channel Lathar device 10 without loss of information in a small storage space. The cleaning robot can not guarantee a large memory size and high computing capacity because it has to maintain a low manufacturing cost. In other words, there is a limit to apply the point cloud which requires large storage capacity and good computing performance, which have been used in existing mobile researches. The present invention proposes a projective map for storing obstacle information and map information around a cleaning robot in a small storage space without loss of information.

Grid maps have been mainly used in general cleaning robots because they have small storage capacity and can be used for small computation performance. However, since the point cloud that represents the environment in three dimensions requires more storage space because the spatial order is larger than that of the two dimensional grid map, the cleaning robot is limited in the computing capacity and storage space. Accordingly, a projection map including three-dimensional information is generated on the grid map in the following manner (S2620)

1) A three-dimensional space is expressed using a cubic n3 (S2621)

2) Determine the grid position to be marked using the horizontal position information of the obstacle obtained in LiDAR (S2622)

3) The value to be marked on the grid is determined using the vertical position information of the obstacle obtained in LiDAR (S2623)

For example, if the three-dimensional surrounding space is expressed using a cubic body of 33 (cm 3), the three-dimensional surroundings are expressed by cubic lattices of 3 cm x 3 cm x 3 cm.

Here, the case where each lattice is represented by 1 byte (8 bits) can be determined as shown in FIG. 78.

As shown in FIG. 78, three-dimensional information can be displayed using a two-dimensional grid map without loss of information by allocating and displaying each bit using a height information grid size. The projection map generation method of FIG. 78 will be described in more detail using the following example.

For example, if the grid size is 1.5 cm (n = 1.5) and the vertical position of the obstacle is 4.5 cm from the ground, Expressed using hexadecimal numbers, the value to be entered into the lattice is 0x70.

   0: 1: 1: 1: 0: 0: 0: 0

As shown in FIGS. 79 and 80, if a grid map is expressed by 2 bytes (16 bits), it can be expressed more finely if it has the same measurement range as a 1 byte grid, and if it has the same grid size as a 1 byte grid, The range can be expressed.

As shown in FIGS. 78 to 80, a method of generating a projection map without loss of position information using a two-dimensional grid map without a three-dimensional array or a three-dimensional point cloud is proposed. A method of representing three-dimensional information by the cleaning robot using the maps of FIGS. 78 to 80 can be shown in FIGS. 81 and 82. FIG.

As shown in FIGS. 81 and 82, it can be seen that the three-dimensional obstacle measured by the cleaning robot can be expressed without loss of information in the two-dimensional projection map. 81, since the obstacle at the (2, 1) position of the area map is within the height of 1n, M (2,1) = 00010000 (where M represents the area map) 8, M (2, 1) = 0x10.

In cleaning robots, distinguishing walls is one of the most important skills. The ability to clean by attaching to the edge or corner of the wall and floor where the dust is gathered is an important measure of the cleaning performance of the cleaning robot which must collect the dust. Therefore, wall following function is very important because it separates the wall from surrounding information and moves close to the wall without collision.

Unlike the conventional cleaning robot that uses one or two infrared or distance measurement sensors to move the robot in a straight line along the wall according to the error of the sensor, it moves around in a zigzag manner. Cleaning scanners based on laser scanning sensors can use more environmental information to accurately identify the surrounding walls and provide more information (eg, the direction of the wall).

Generally, it is difficult to use the conventional 2D laser scan sensor based wall extraction algorithms because the cleaning robot based on 3D multi-channel Lada uses 3D environment information.

However, in spite of acquiring 3D environment information, a multi-channel RDA based moving object to which an embodiment of the present invention is applied can easily integrate all information using a projection map, And can be used as it is.

1) Replace the projection map containing height information with binary information expressed as 1 or 0.

2) Extract the linear region by Hough transform the binary projection map.

3) Increase the priority for areas that have not been cleaned compared to the cleaning map, and increase the priority of the straight line near the distance.

More specifically, in order to extract information about a wall, a projection map containing three-dimensional information is compressed into two-dimensional information through a binarization process represented by 1 and 0. The process of binarizing the projection map can be represented as Algorithm 1) (S2630)

Algorithm 1)

if M (n, m) > 0

B (n, m) = 1;

else

B (n, m) = 0;

Here, B is a map in which binary maps are used to display all the area maps using only 1 and 0.

As shown in Algorithm 1), the above process of the projection map M is transformed into the binary map B and converted to the two-dimensional information (S2631). The information in FIG. 82 is binarized as shown in FIG.

Using the binary map shown in FIG. 83, we can easily apply the existing algorithms. In this paper, linear components are extracted through Hough transform, which is often used for existing image processing and obstacle map analysis (S2632). In this paper,

After the projection map for displaying the three-dimensional environment information is converted into a binary map through binarization and the straight line components are extracted through the Hough transform to estimate the wall, the estimated walls are given priority in step S2633. Priority of each wall is classified according to the following.

1) Increase the priority of cleaning and non-cleaning.

2) The higher the distance relation and distance, the higher the priority.

Priority is given to all estimated walls based on the above two points, and the wall of the highest priority among them is followed.

The low obstacles that are not detected by the conventional cleaning robots must be detected and avoided because they have a lot of influence on the safe running and cleaning performance of robots, such as clothes, toy blocks, or animal excrements that can be caught in the robots. Since the moving object 1 equipped with the multi-channel LiDAR to which the embodiment of the present invention is applied can acquire information about obstacles lower than the position where the multi-channel LiDAR is mounted, it can prevent collision with a low obstacle and various problems Can be avoided. For this purpose, it is very important to distinguish obstacles from the floor in 3D map information.

69, the distance from the sensor to the floor in the downward direction information of the cleaning robot measured from the multichannel LiDAR mounted on the cleaning robot is dn = h / cos (? N) (where n = 1, 2, 3 ), It can be estimated as the bottom. Otherwise, it can be displayed as an obstacle rather than a floor. If this is expressed using the projection map, it can be expressed as follows (S2640)

For example, referring to FIG. 81, when the position of the multi-channel LiDAR sensor is 3n, low obstacles are displayed with a number smaller than 0x80 (10000000). Also, all non-low obstacles, such as normal cleaning robots, will be displayed on the projection map with a value greater than or equal to 0x80. By combining these two pieces of information, the obstacle information can be obtained, and the cleaning robot can plan the collision avoidance path in advance before facing the obstacle as in the next chapter (S2650)

&Lt;

84 to 88 are views of a cleaning part of a moving object according to embodiments of the present invention.

The moving body (cleaning robot) provided with the cleaning part 90 can detect the inflow amount of the dust 1430 by using a flow sensor 1410 mounted on one side of the suction port 1420 as shown in FIG. 85 have.

For this purpose, the moving body 1 can use an ultrasonic flow sensor, a TOF-based flow detector, or the like as a flow rate sensor. When the moving object 1 detects the inflow amount of dust by using such a flow sensor, the moving object 1 may be less sensitive to the dust adhering to the window than when using the optical device.

The cleaning unit 90 may further include a foreign matter inflow amount detecting unit. The foreign matter inflow amount detecting unit detects a foreign matter inflow amount by using a flow rate sensor attached to an intake port of a device performing a cleaning function. The cleaning robot can control the cleaning function for the room based on the state of the floor surface and the inflow amount of foreign matter.

86 and 87, the cleaning unit 90 includes a side brush and / or a brush wheel. Referring to FIG. 88, a cylindrical brush wheel attached to or formed at the lower end of the front end of the mobile robot acts as a wheel so that the threshold or the carpet can be raised relatively easily.

&Lt; Moving Device of Moving Object &

FIGS. 89 to 96 are views for explaining an operation of moving a moving object according to the embodiments of the present invention.

The mobile device 20 includes a forward drive portion 910, a steering portion 920, and a reverse drive portion 930. [ The moving body 1 can perform the cleaning motion using the forward movement or the backward movement. The moving object 1 determines a moving direction and a path using a TOF-based 360-degree distance measuring device.

The moving object (1) equipped with the TOF-based 360-degree distance measuring device can freely move backward regardless of the mounting position of the sensors such as the front and rear sensors, so that the following various motions can be applied Do.

First, when the mobile body 1 is turned 180 degrees, the mobile body 1 does not rotate as shown in FIG. 90 but directly moves backward as shown in FIG. 91, thereby reducing the entire cleaning time.

Second, when the mobile unit 1 is concentratedly cleaned in dusty places, the mobile unit 1 can be cleaned while moving back and forth as shown in FIG. 92, thereby sufficiently removing dust.

In this case, since the mobile body 1 does not rotate and cleans the area, it is possible to maintain the position estimation performance and to reduce the slip that often occurs in the rotation. In addition, since the moving body 1 uses both the brush and the mop in both directions, it solves the problem that the brush and the mop are used only in one direction, It is possible to maintain the performance and to reduce the slip that often occurs in rotation.

<Creation of Collision Avoidance Path>

93 to 96, an optimal collision avoidance path generation algorithm capable of ensuring the safety of the cleaning robot and the maximum cleaning area will be described.

Since the cleaning robot according to the present embodiment expresses all the obstacle information including the low obstacle on the projection map, it is possible to plan the path that can avoid the collision before confronted with the obstacle. Unlike the conventional cleaning robots which generate a path for avoiding collision by connecting the furthest points from the obstacle, the cleaning robot according to the present embodiment closely approaches the obstacle and avoids collision with the obstacle, It can be cleaned.

93 shows the relationship between the cleaning robot and the obstacle when the cleaning robot performs cleaning.

As shown in Figure 93, all obstacles must be located outside the safe zone of the cleaning robot. However, it is possible to clean the large area as much as possible by creating a path such that the closest distance do from the cleaning robot to the obstacle is minimized. Based on FIG. 95, the optimal path is reduced to the optimization problem of equation (3).

The optimization problem of Eq. (3) must be determined and the movement path of the cleaning robot must be determined. For this, the optimization problem of Eq. (3) can be applied to the existing path planning algorithm of the mobile robot. Particularly, the present invention considers the RRT (rapid random tree) and the Potential field (similar to the wavefront method).

First, in the case of the RRT, a roadmap is created as a set of various paths through which the cleaning robot can move, and the optimization problem of Equation (3) is applied to select an optimum path among the paths.

94 shows a process of selecting an optimal path from the RRT to which the optimization problem of Equation (3) is applied.

In FIG. 94, the dotted line indicates all the paths that the cleaning robot can move, and the solid line is the movement path of the cleaning robot selected through the optimization problem of Equation (3) among all possible paths.

FIGS. 95 and 96 illustrate a potential field-based path generation process through the optimization problem of Equation (3).

When the potential field is applied, the potential as shown in FIG. 13 (a) is given according to the distance between the obstacle and the cleaning robot.

As shown in FIG. 95, when the optimization problem of Equation (3) is applied to the potential field, the potential is minimized when do is r + ds, so that the path shown in FIG. 96 can be obtained.

The clean running algorithm of a cleaning robot that can acquire 3D environment information by installing a multi - channel LiDAR has the following effects. First, by proposing a projection map, it is possible to apply the three-dimensional spatial information that requires a large storage space and computation capability to the existing two-dimensional grid map without information loss. In addition, wall information can be extracted by applying existing methods to 3D information easily using projection map, and obstacles can be distinguished. Also, it is possible to maximize the cleaning area based on the wall and obstacle information, and to plan the traveling route to avoid the collision with the obstacle.

<Display of route>

97 is a diagram exemplarily showing a moving path of a moving object according to embodiments of the present invention.

The moving body of the present invention may further include a display portion. At this time, a map indicating a traveling area of the moving object is displayed so that the user can visually confirm the area where the moving object has been cleaned through the display unit. For example, a map including both the area where the user performed the cleaning operation of the moving object and the area where the cleaning operation is to be performed may be displayed as the game character moves.

The moving body, and the Lidar device are shown separately in the respective drawings, the plurality of components may be mutually coupled to form at least one module. The components are connected to a communication path connecting a software module or a hardware module inside the device and operate organically with each other. These components communicate using one or more communication buses or signal lines.

The moving body and the Lidar device may be implemented in logic circuitry by hardware, firmware, software or a combination thereof, and may be implemented using a general purpose or special purpose computer. The device may be implemented using a hardwired device, a field programmable gate array (FPGA), an application specific integrated circuit (ASIC), or the like. Further, the device may be implemented as a System on Chip (SoC) including one or more processors and controllers.

The mobile device and the Lidar device may be implemented as software, hardware, or a combination thereof in a computing device having hardware components. The computing device includes a communication device such as a communication modem for performing communication with various devices or wired / wireless communication networks, a memory for storing data for executing a program, a microprocessor for executing and calculating a program, Device. &Lt; / RTI &gt;

It is to be understood that the present invention is not limited to the above-described exemplary embodiments, and that various changes and modifications may be made without departing from the scope of the present invention. Or may be variously modified and modified by executing one or more processes in parallel or by adding other processes.

The operations according to the present embodiments may be implemented in the form of program instructions that can be executed through various computer means and recorded in a computer-readable medium. A computer-readable medium represents any medium that participates in providing instructions to a processor for execution. The computer readable medium may include program instructions, data files, data structures, or a combination thereof. For example, there may be a magnetic medium, an optical recording medium, a memory, and the like. The computer program may be distributed and distributed on a networked computer system so that computer readable code may be stored and executed in a distributed manner. Functional programs, codes, and code segments for implementing the present embodiment may be easily deduced by programmers of the technical field to which the present embodiment belongs.

The present embodiments are for explaining the technical idea of the present embodiment, and the scope of the technical idea of the present embodiment is not limited by these embodiments. The scope of protection of the present embodiment should be construed according to the following claims, and all technical ideas within the scope of equivalents thereof should be construed as being included in the scope of the present invention.

Claims (20)

A lidar device for measuring a distance from an object based on flight time;
A path controller for processing the point cloud data generated based on the measured distance to generate a movement path; And
And a moving unit configured to move the moving object based on the moving route,
. The method according to claim 1,
The Lidar apparatus includes:
An optical transceiver for transmitting and receiving light and outputting an electric signal;
A signal discriminator for analyzing the electrical signal to generate a control signal; And
A distance measuring unit for calculating a distance of the light based on the control signal,
And the moving object. 3. The method of claim 2,
Wherein the Lidar apparatus includes a plurality of optical transceivers, and the plurality of optical transceivers are installed at different angles to perform multi-channel scanning. 3. The method of claim 2,
The optical transceiver includes a first mirror including a hole through which the emitted light passes, and a second mirror moving in a predetermined cycle, wherein the first mirror has a focal point at which light rays reflected by a curved reflecting surface converge . 3. The method of claim 2,
Wherein the optical transceiver outputs an electrical signal during a detection period of a sampling period of the optical transceiver. 3. The method of claim 2,
Wherein the signal discriminator converts the electrical signal and detects a time point having a predetermined reference magnitude to generate the control signal. The method according to claim 6,
Wherein the distance measuring unit corrects the flight time using the pulse width of the control signal. 3. The method of claim 2,
Wherein the distance measuring unit calculates the flight time using a time digital converter having a plurality of oscillators whose clock widths are adjusted based on a position and a signal path of the controlled logic device. The method according to claim 1,
Wherein the moving body includes a clock generator for generating a reference clock,
The distance measuring device includes a reference clock counter for counting the reference clock received from the clock generator and counting an internal clock generated in an internal oscillator,
Wherein the distance measuring unit corrects the flight time using a ratio of the number of the reference clocks and the number of the internal clocks. The method according to claim 1,
The Lidar apparatus includes:
Further comprising an artificial landmark detector for determining whether the target object is an artificial landmark and analyzing an invisible bar code included in the artificial landmark. The method according to claim 1,
Analyzing the point cloud data to generate a key frame related to the node, calculating an odometry edge between consecutive nodes, updating the key frame to estimate a local map, Further comprising a map generator for detecting a loop closure edge between the nodes and estimating a global map by correcting the position of the node based on the odometry edge and the loop closure edge. The method according to claim 1,
Wherein the path controller comprises:
And a floor condition determination unit for analyzing the point cloud data to generate information related to a change in height of the floor surface. The method according to claim 1,
Further comprising a projection map generator for generating three-dimensional environment information for defining the obstacle on three dimensions by analyzing the point cloud data and generating a projection map including the three-dimensional environment information on a two-dimensional space map. An optical transceiver for transmitting and receiving light and outputting an electric signal;
A signal discriminator for analyzing the electrical signal to generate a control signal; And
A distance measuring unit for calculating a distance of the light based on the control signal,
Lt; / RTI &gt; 15. The method of claim 14,
The optical transceiver includes a first mirror including a hole through which the emitted light passes, and a second mirror moving in a predetermined cycle, wherein the first mirror has a focal point at which light rays reflected by a curved reflecting surface converge Lt; / RTI &gt; 15. The method of claim 14,
Wherein the optical transceiver outputs an electrical signal during a detection period of the sampling period of the optical transceiver. 15. The method of claim 14,
Wherein the signal discriminator converts the electrical signal and detects a time point having a predetermined reference size to generate the control signal. 18. The method of claim 17,
Wherein the distance measuring unit corrects the flight time using the pulse width of the control signal. 15. The method of claim 14,
Wherein the distance measuring device calculates the flight time using a time digital converter having a plurality of oscillators whose clock widths are adjusted based on the position and signal path of the controlled logic device. The method according to claim 1,
The distance measuring device includes a reference clock counter for counting a reference clock received from an external clock generator and counting an internal clock generated in an internal oscillator,
Wherein the distance measuring unit corrects the flight time using a ratio of the number of the reference clocks and the number of the internal clocks.

Images