DE102023105755A1 - DRIVERLESS TRANSPORT VEHICLE - Google Patents

Abstract

The invention relates to a driverless transport vehicle, in particular for material transport in industrial processes, with a safe environmental sensor, a control device and a drive, wherein the control device is designed to control the drive in order to control the transport vehicle with regard to direction and speed along a travel route, wherein the safe environmental sensor and the control device are designed as separate units, wherein the safe environmental sensor is designed to transmit environmental data to the control device, wherein the safe environmental sensor is designed to record measurement data of the environment of the transport vehicle, to carry out object recognition based on the measurement data and to communicate recognized objects to the control device in the form of object data.

Description

The present invention relates to a driverless, i.e. in particular autonomously driving, transport vehicle, which can be used, for example, for material transport in industrial processes.

Such transport vehicles are generally known. Such transport vehicles, also called AGVs (Automated Guided Vehicle), usually have an environmental sensor that detects obstacles or objects in the environment. The transport vehicle can be programmed to follow a predetermined route, with rigid, virtual protective fields being defined around the transport vehicle, which are monitored by the environmental sensor. If an object (for example a person) enters the protective field, the speed of the transport vehicle is initially reduced. If the protective field is permanently violated, this often leads to an emergency stop of the transport vehicle in order to avoid a collision. Such a slowing down of the speed and/or an emergency stop disrupts and slows down the transport process, which is undesirable.

It is therefore an object underlying the invention to provide a transport vehicle which reduces productivity losses due to shutdowns or frequent speed reductions.

This object is achieved by a driverless transport vehicle according to claim 1.

The driverless transport vehicle of the invention is in particular an autonomously driving transport vehicle, preferably for transporting materials in industrial processes. The transport vehicle comprises a safe environmental sensor, a control device and a drive, wherein the control device is designed to control the drive in order to control the transport vehicle in terms of direction and speed along a route. The safe environmental sensor and the control device are designed as separate units. The safe environmental sensor is also designed to transmit environmental data to the control device. The safe environmental sensor is also designed to record measurement data of the environment of the transport vehicle, to carry out object recognition based on the measurement data and to communicate recognized objects to the control device in the form of object data, wherein the object data is in particular part of the environmental data.

Due to the fact that the environmental sensor is a safe environmental sensor, a great deal of trust can be placed in the measurement data generated by the safe environmental sensor. In particular, it can be assumed, for example, that a person who may be at risk from the transport vehicle is detected. Furthermore, according to the invention, object detection is already carried out in the environmental sensor, which is based on the safe measurement data. The data on the objects determined in this way therefore also have a high level of security.

Unlike conventional solutions, the environmental sensor transmits detected objects to the control device in the form of object data, whereby the control device receives secure data which, due to its high reliability, can be used, for example, to avoid a detected object (as will be explained in more detail later).

According to the invention, this results in the advantage that rigid protective fields can be dispensed with and that objects can often simply be avoided, which means that the driverless transport vehicle can be prevented from slowing down or even stopping. Productivity losses due to constant braking or stopping can thus be reduced or eliminated entirely according to the invention.

Static objects that would traditionally have led to a shutdown or a stop of the transport vehicle can thus be avoided with the help of the invention, whereby, for example, no or only a slight reduction in speed can occur. In comparison to previous complete stops of the transport vehicle, avoiding them, possibly also with a reduction in speed, is significantly more productive.

According to the invention, it is particularly advantageous that through the use of the secure environmental sensor and the object recognition in the secure environmental sensor, secure data is made available to the control device, whereby the security level of the environmental sensor is "transferred" to the control device, so to speak, without the control device itself necessarily having to meet a certain security level. In addition, the task of object recognition is eliminated for the control device, so that a simpler control device with lower computing power can be used. Such control devices with lower computing power are also available as secure control devices, so that both both a non-safe and a safe control device can be used.

Further details of the invention are explained below.

The driverless transport vehicle is in particular a floor-based conveyor with its own drive that is automatically controlled. The transport vehicle is used in particular for transporting materials, for example pulling and/or carrying conveyed goods. The transport vehicle can comprise active and/or passive load-handling devices. For example, the load-handling device can be a tray or surface for the material to be transported. Alternatively or additionally, the load-handling device can be, for example, the fork of a forklift truck, which means that the transport vehicle can be an autonomous forklift truck, for example. The transport vehicle can also have a small width, for example a maximum of 1.0 m or 1.5 m, in order to be able to be used flexibly in industrial processes.

The drive of the transport vehicle can, for example, comprise one or more driven wheels and a steering system. The control device can control the wheels and the steering system accordingly in order to determine the route.

In particular, the control device can have a closed control loop, wherein the control device itself calculates the optimal route and/or vehicle speed in the closed control loop, wherein the calculation is carried out in such a way that dangerous collisions do not occur.

In particular, the route can also be determined in such a way that, for example, static (non-moving) objects are avoided at a smaller distance than moving objects.

The control device can preferably also be designed to carry out complex, application-specific and more complex computing operations, which are based in particular on the object data of the environmental sensor.

According to the invention, the safe environmental sensor and the control device are designed as separate units, which means that the environmental sensor and the control device do not share a computing device, for example. The environmental sensor and the control device can, however, be coupled to one another via a data connection, via which the safe environmental sensor transmits environmental data to the control device.

In general, the environmental data can contain information about the environment of the transport vehicle. In particular, according to the invention, the environmental data includes object data that contains information about objects detected by the environmental sensor. Based on the object data, the control device can then preferably adapt the route (and thus the direction and/or speed of the transport vehicle) in order to avoid the objects, for example.

The environmental sensor comprises in particular an optoelectronic sensor and a computing device to determine the measurement data. According to the invention, this computing device can also be used to carry out object detection. It is understood that in the following explanations, a safe environmental sensor is always meant, even if only an environmental sensor is mentioned, unless something else is explicitly stated.

Advantageous embodiments and further developments of the invention can be taken from the description, the figures and the dependent claims.

According to a first embodiment, the measurement data includes several separate measurement points for different locations in the environment. In addition, the environmental sensor can be designed to summarize several of the measurement points and communicate them to the control device as a recognized object. Several points can therefore be summarized and thus recognized and/or defined as an object. The advantage here is that individual points in the environmental data do not have to be transmitted to the control device, but only the (more abstract) information about the object.

For the sake of clarity, it should be explained that when the environment is detected by the environmental sensor, measurement data is generated, object recognition is carried out on the basis of the measurement data and the measurement data processed in this way are then transmitted to the control device as environmental data.

According to a further embodiment, the environmental sensor is designed to recognize several, in particular directly adjacent, measuring points whose distance from the environmental sensor differs by less than a predetermined threshold value as one object. The object recognition can therefore take place in such a way that measuring points that have a similar distance from the environmental sensor are defined as one object. For example, a distance of 50 cm can be used as a predetermined threshold value, resulting in a distance range of 50 cm width, so that, for example, measuring points with a distance between 1.5 m and 2 m are grouped together and viewed as one object. Other predetermined threshold values of eg 20 cm or 100 cm are of course also possible.

In particular, only directly adjacent measuring points, i.e. directly adjacent measuring points as seen from the environmental sensor, are combined to form the same object. In this respect, several objects from measuring points with the same distance to the environmental sensor can also be detected if there is another object with measuring points with a different distance (or with a distance that differs by more than the predetermined threshold value) between these objects.

This type of simple object detection enables object detection even with the limited computing resources of the environmental sensor. Of course, other types of object detection are also possible, for example based on particle filtering or object classification.

In particular, the environmental sensor and/or the control device can also be designed to intentionally enlarge the detected objects, in particular by a predetermined value, for example by 10 cm, 30 cm or 50 cm in each spatial direction. In this way, a buffer zone can be created in order to avoid detected objects, for example at a safe distance.

According to a further embodiment, the environmental sensor is designed not to take into account measuring points whose distance from the environmental sensor exceeds a predetermined threshold value when detecting objects. For example, measuring points with a distance of more than 10 m, 20 m or 50 m can be disregarded. Due to this procedure, the environmental sensor is relieved because a large number of measured values or measuring points can be disregarded.

According to a further embodiment, the object data, in particular of at least one object, includes the distance and/or the speed and/or the extent of the detected object. The distance can preferably be the radial distance to the environmental sensor. The speed can in particular be the radial speed relative to the environmental sensor. The extent can in particular be the angular extent, which indicates, for example, in which angular range the environmental sensor "sees" or has detected the object. In the case of three-dimensional objects, for example, an orientation and/or the depth of the object can additionally be specified in the object data.

The distance to an object can correspond to the shortest determined distance of a measuring point within the object. The object speed, in particular the radial object speed, can correspond in particular to the largest (e.g. negative) value of the radial distance change between two measurements by the environmental sensor. The angular object extent can correspond in particular to the largest connected angular range of an object and is particularly dependent on the radial object distance.

In addition to the object data, the environmental sensor can also transmit a proven measurement uncertainty and/or confidence for at least some (or all) of the values of distances of the object and/or speeds of the object and/or dimensions of the object to the control unit. Such measurement uncertainties are usually only known for secure sensors.

According to a further embodiment, the object data represent recognized objects as a two-dimensional or three-dimensional geometric shape, in particular as an ellipse, ellipsoid, rectangle and/or cuboid. Preferably, all recognized objects can be represented as the same geometric shape. Alternatively, different geometric shapes are also possible.

By choosing simple geometric shapes, the amount of data to be transmitted from the environmental sensor to the control device can be reduced, since for a rectangle, for example, only the diagonal intersection and the side lengths need to be transmitted. For an ellipse, for example, the positions of the focal points and the lengths of the main vertices can be transmitted.

According to a further embodiment, the control device is designed to adapt the route and/or the speed of the transport vehicle based on an object communicated to it, ie in particular based on the object data, in order to avoid a collision of the transport vehicle with the object, for example. The route can be determined in the control device in particular with the help of a travel path. The travel path can indicate the area in which the transport vehicle can move without the risk of a collision with an object. Based on detected objects, the control device can Change the travel path and then adjust the travel route to the changed travel path. In particular, the travel route is laid out around the detected objects, and the speed of the transport vehicle can be reduced if, for example, objects come very close to the transport vehicle, which can be determined, for example, based on the radial object distance (which can be part of the secure object data).

It is also possible that movements of a detected object and/or movements of the transport vehicle are extrapolated into the future in order to predict future collisions. If such a future collision is detected, the route can be changed and/or the speed increased/reduced in order to avoid the collision.

Preferably, the transport vehicle is designed to determine the travel route based on the environmental data of only one environmental sensor. In particular, the transport vehicle only comprises exactly one secure environmental sensor. By using only one secure environmental sensor, the need to fuse the data of several non-secure environmental sensors in order to obtain secure environmental data can be eliminated.

Alternatively, an embodiment with several secure environmental sensors is also possible. The use of several secure environmental sensors has the advantage that secure object data is generated from different perspectives, which then combined enables better confidence in the object data and/or increases the availability of the system if, for example, the front window of an environmental sensor is dirty.

According to a further embodiment, the control device is designed to communicate the route to the environmental sensor, wherein the environmental sensor is designed to adapt the acquisition of the environmental data to the route based on the route. There can therefore be a return channel between the control device and the environmental sensor, which uses, for example, the same data connection via which the environmental data is transmitted from the environmental sensor to the control device. It is then possible for the environmental sensor to scan the planned route more precisely and/or more frequently and/or, for example, to try to "look into" the curve as far as possible when cornering.

According to a further embodiment, the safe environmental sensor is designed to detect contamination of its optics and/or to diagnose aging of its components and a resulting loss of sensitivity and/or the environmental sensor is designed according to Safety Integrity Level (SIL) 2, 3 or 4. The above features are examples of how a "safe" environmental sensor can be achieved. Alternatively or additionally, the environmental sensor can also be designed to detect remissions of less than 5%, preferably less than 2%, particularly preferably less than 1.8%. Also alternatively or additionally, it is possible for the environmental sensor to have a redundant computing device in order to meet the increased safety requirements. Finally, the environmental sensor can also have a minimum object resolution, which allows the detection of objects with a size (e.g. a width) of at least 150 mm, preferably at least 70 mm, particularly preferably at least 50 mm. These values correspond, for example, to the size of a human leg or a torso.

According to a further embodiment, the safe environmental sensor and the control device are arranged at a distance on or in the transport vehicle, preferably in separate housings. As already mentioned, a data connection can be provided between the safe environmental sensor and the control device, which serves to transmit the environmental data. The separation of the environmental sensor and the control device has the advantage that the high requirements for the "safe" environmental sensor only have to be met for the environmental sensor. Lower requirements can then apply to the control device, which would not be possible with a common or combined unit. In particular, there are therefore at least two separate computing devices in the transport vehicle, one in the environmental sensor and another in the control device.

According to a further embodiment, the environmental sensor is a laser scanner or a LIDAR sensor. The environmental sensor can in particular be designed to measure distances, for example by means of a time-of-flight method or a signal-of-flight method. The environmental sensor can, for example, scan and/or detect an angular range of 180°, 270° or 360° around the transport vehicle. In particular, the environmental sensor can successively emit a large number of (light) signals in different spatial directions and determine the distance to objects in the environment from a signal reflected back from the environment. For example, the emitted signals can only be emitted in one plane, for example in a plane approximately 30 cm above the ground. Alternatively, the signals can also be emitted in several planes. to capture the three-dimensional shape of objects.

Instead of laser scanners or LIDAR sensors, other sensors are also possible, for example the environmental sensor can be a radar sensor, an ultrasonic sensor, a 3D camera and the like.

According to a further embodiment, the transport vehicle is an autonomous forklift or an autonomous industrial truck. The transport vehicle is in particular free-moving and is in particular not intended for the transport of people. For example, the transport vehicle can have a loading area for goods to be transported, whereby a pallet or a box can be transported on the loading area, for example.

The invention further relates to a method for operating a driverless transport vehicle, in particular for transporting materials in industrial processes, wherein a safe environmental sensor and a control device of the transport vehicle are designed as separate units. The safe environmental sensor transmits environmental data to the control device. The safe environmental sensor records measurement data of the environment of the transport vehicle, performs object recognition based on the measurement data and reports recognized objects to the control device in the form of object data.

The statements made about the transport vehicle according to the invention apply accordingly to the method according to the invention. This applies in particular with regard to advantages and preferred embodiments. Furthermore, it is understood that all of the features mentioned here can be combined with one another, unless explicitly stated otherwise.

• 1 ein Transportfahrzeug und ein Hindernis in seitlicher schematischer Darstellung;
• 2 eine schematische Darstellung von erkannten Objekten; und
• 3 eine schematische Darstellung der Objekte von 2 mit zusätzlichen Informationen zu den Objekten.

The invention is described below purely by way of example with reference to the drawings.

• 1 a transport vehicle and an obstacle in a schematic side view;
• 2 a schematic representation of detected objects; and
• 3 a schematic representation of the objects of 2 with additional information about the objects.

1 shows a driverless, autonomously driving transport vehicle in the form of an AGV 10. The AGV 10 moves along a route 12 in which an object 14 in the form of an obstacle is located.

The AGV 10 comprises an environmental sensor, which in the present example is designed as a laser scanner 16. The laser scanner 16 is connected to a separately implemented control device 20 via a data connection 18. The control device controls a drive of the AGV 10, of which wheels 22 in 1 are shown.

A cargo 24 is transported on the AGV 10.

During operation of the AGV 10, the laser scanner 16 repeatedly emits laser pulses 26 into the environment. The laser pulses 26 are reflected back to the laser scanner 16 by objects 14 in the environment as reflected pulses 28, whereby the laser scanner 16 can determine the distance to the objects 14 or individual measuring points based on the time of flight measurement.

2 shows schematically the measuring points recorded by the laser scanner 16, which originate from objects 14, in a plan view. 2 shows a first group of measuring points 30, which were recorded in a first scanning process of the laser scanner and a second group of measuring points 32, which were recorded in a later second scanning process of the laser scanner 16. A movement or change in the measuring points 30, 32 can be seen through the different scanning processes.

The laser scanner 16 comprises its own computing device (not shown in the figures) in which an object recognition or object separation takes place. An object 14 is recognized in particular by combining connected measuring points 30, 32 of the same group as one object 14 as seen from the laser scanner 16, which differ by less than a predetermined threshold value d _obj,min . In 2 It can be seen that four objects 14 were detected in this way, which the AGV 10 must avoid. The objects 14 were each abstracted as ellipses.

The objects 14 are communicated to the control device 20 via the data connection 18, whereupon the control device 20 defines a travel path 34 which runs around the objects 14. A travel route 12 for the AGV 10 is then defined within the travel path 34.

Various information about the objects 14 is transmitted from the laser scanner 16 to the control device 20. This additional information is in 3 In particular, an object distance 36 (and its direction), an object speed 38 and a Object extension 40 (for example as an angle) is determined and transmitted to the control device 20.

During operation of the AGV 10, the advantage arises that through the use of the safe laser scanner 16 and through the object detection in the laser scanner 16, reliable and thus safe data are transmitted to the control device 20, so that the route can be planned based on the safe object data.

List of reference symbols

10

AGV

12

Route

14

object

16

Laser scanner

18

Data connection

20

Control device

22

wheel

24

Transport goods

26

Laser pulse

28

reflected pulse

30

first group of measuring points

32

second group of measuring points

34

Driving tube

36

Object distance

38

Object speed

40

Object extent

Claims (13)

Driverless transport vehicle (10), in particular for material transport in industrial processes, with a safe environmental sensor (16), a control device (20) and a drive (22), wherein the control device (20) is designed to control the drive (22) in order to control the transport vehicle (10) with regard to direction and speed along a travel route (12), wherein the safe environmental sensor (16) and the control device (20) are designed as separate units, wherein the safe environmental sensor (16) is designed to transmit environmental data to the control device (20), wherein the safe environmental sensor (16) is designed to record measurement data of the environment of the transport vehicle (10), to carry out object recognition based on the measurement data and to communicate recognized objects (14) to the control device (20) in the form of object data. Transport vehicle (10) to Claim 1 , wherein the measurement data comprise a plurality of separate measurement points (30, 32) for different locations in the environment and the environmental sensor (16) is designed to combine a plurality of the measurement points (30, 32) and to communicate them to the control device (20) as a recognized object (14). Transport vehicle (10) to Claim 2 , wherein the environmental sensor (16) is designed to recognize a plurality of measuring points (30, 32) whose distance from the environmental sensor (16) differs by less than a predetermined threshold value as one object (14). Transport vehicle (10) according to one of the preceding claims, wherein the environmental sensor (16) is designed not to take into account measuring points (30, 32) whose distance from the environmental sensor (16) exceeds a predetermined threshold value during object detection. Transport vehicle (10) according to one of the preceding claims, wherein the object data comprise the distance (36) and/or the speed (38) and/or the extent (40) of the detected object (14). Transport vehicle (10) according to one of the preceding claims, wherein the object data represent recognized objects (14) as a two-dimensional or three-dimensional geometric shape, in particular as an ellipse, ellipsoid, rectangle and/or cuboid. Transport vehicle (10) according to one of the preceding claims, wherein the control device (20) is designed to adapt the travel route (12) and/or the speed of the transport vehicle (10) on the basis of an object (14) communicated to it, for example in order to avoid a collision of the transport vehicle (10) with the object (14). Transport vehicle (10) according to one of the preceding claims, wherein the control device (20) is designed to communicate the travel route (12) to the environmental sensor (16), wherein the environmental sensor (16) is designed to adapt the detection of the environmental data to the travel route (12). Transport vehicle (10) according to one of the preceding claims, wherein the safe environmental sensor (16) is designed to detect contamination of its optics and/or to diagnose aging of its components and a resulting loss of sensitivity and/or is designed according to Safety Integrity Level 2, 3 or 4. Transport vehicle (10) according to one of the preceding claims, wherein the secure environmental sensor (16) and the control device (20) are arranged at a distance on or in the transport vehicle (10), preferably in separate housings. Transport vehicle (10) according to one of the preceding claims, wherein the environmental sensor is a laser scanner (16) or LIDAR sensor. Transport vehicle (10) according to one of the preceding claims, wherein the transport vehicle (10) is an autonomous forklift truck or an autonomous industrial truck. Method for operating a driverless transport vehicle (10), in particular for material transport in industrial processes, wherein a safe environmental sensor (16) and a control device (20) of the transport vehicle (10) are designed as separate units, wherein the safe environmental sensor (16) transmits environmental data to the control device (20), wherein the safe environmental sensor (16) records measurement data of the environment of the transport vehicle (10), carries out object recognition based on the measurement data and communicates recognized objects in the form of object data to the control device (20).

Images