CN113791619B - Airport automatic driving tractor dispatching navigation system and method - Google Patents

Abstract

The invention relates to the technical field of aircraft traction operation, in particular to an airport autopilot tractor dispatching navigation system and method.

Description

Airport automatic driving tractor dispatching navigation system and method

Technical Field

The present invention relates to the technical field of aircraft towing operations, and in particular to an airport automatic driving towing vehicle dispatching navigation system and method.

Background technique

The rapid growth of air transport volume has greatly increased the operational pressure on airports. In order to improve the safety and efficiency of tractor operations, some universities and institutions at home and abroad have also carried out a series of related technical research. They mainly focus on one direction, that is, by installing sensors on the tractor to obtain traffic situation information around the tractor, and provide the information that affects the safety of the operation to the tractor driver in a visual interface. However, this method still cannot solve the problems of strict requirements for tractor drivers, personal dangers to operators, the need for multi-person cooperation and low efficiency.

Summary of the invention

This patent is proposed based on the above-mentioned needs of the prior art. The technical problem to be solved by this patent is to provide an airport automatic driving tractor dispatching and navigation method and system to ensure the safety of tractor operators and improve operating efficiency.

In order to solve the above problems, the technical solutions provided by this patent include:

Provided is an airport autonomous driving tractor dispatching and navigation system, comprising: a cloud control center, the cloud control system comprising a tractor matching system and a real-time navigation map generating system; the tractor matching system retrieves an autonomous driving tractor corresponding to the model of an aircraft to be towed, and sends an assignment command and information about the aircraft to be towed to an idle tractor closest to the aircraft to be towed; the real-time navigation map generating system abstracts an airport scene map into a scene network diagram, and generates a real-time navigation map through data information transmitted by a smart roadside system; a smart roadside system, the smart roadside system comprising a roadside communication unit and a multi-element roadside perception unit; the multi-element roadside perception unit comprising a laser radar and a camera, collecting point cloud data through the laser radar and collecting video data through the camera; and an autonomous driving platform, the autonomous driving platform comprising The system comprises multiple sensors, a perception unit, a decision-making and planning unit and a control unit; the multiple sensors respectively collect the position, posture, video and point cloud data of the aircraft to be towed; the perception unit obtains the data information transmitted by the multiple sensors through distributed fusion; the decision-making and planning unit is divided into three levels: path planning, behavior decision and motion planning. The path planning layer generates a global path. After receiving the path, the behavior decision layer makes a behavior decision in combination with the information received from the perception unit, and the motion planning layer plans a characteristic trajectory according to the behavior decision. The trajectory is the final driving path planned for the autonomous driving traction, and the control unit controls the tractor to move along the obtained trajectory; the cloud control center, the smart roadside system and the autonomous driving platform transmit information through the 5G aviation airport mobile communication system.

Preferably, the path division layer uses the A* algorithm to generate a global path. The real-time navigation map at a certain moment can be regarded as a static road network. The road network is simplified into small squares. The combination of the small squares is the path finally found. The initial node and the target node are the starting point and the end point of the path respectively. The priority of each node is calculated by the heuristic function f(n)=g(n)+h(n), where f(n) is the comprehensive priority of node n. When selecting the next node to be traversed, the node with the highest comprehensive priority will always be selected. g(n) is the cost of node n from the starting point, and h(n) is the estimated cost of node n from the end point.

Preferably, during the operation of the A* algorithm, the node with the highest priority is selected from the priority queue each time as the next node to be traversed, and finally a shortest path is determined.

Preferably, the motion planning layer plans the desired motion trajectory of the vehicle in the local space and time at the decision point based on environmental information, upper-level decisions and real-time posture information of the vehicle body, and the motion trajectory includes trajectory, speed, direction and state.

Preferably, the cloud control center also includes an information processing system, which obtains all flight information of the day from the airport operation control center, sorts the information in chronological order from early to late according to the expected departure time of each aircraft, and arranges the flight information in the system to form a waiting queue for departing aircraft.

Preferably, the cloud control center includes a communication system, the smart roadside system includes a roadside communication unit, and the autonomous driving platform includes a communication unit. The roadside communication unit transmits the point cloud data and video data obtained by the multi-dimensional roadside perception unit to the communication system, and the cloud control center processes the received data to obtain an implementation navigation map, and transmits it to the communication unit through the communication system.

A method for dispatching and navigating an airport autonomous tractor is also provided, comprising: S1, a cloud control center obtains flight information, arranges all flight information in chronological order to form an aircraft waiting queue, and the aircraft to be towed issues a push-out application before its expected departure, and sends its own information to the cloud control center at the same time, the information including the aircraft model, position, attitude and target position; S2, a tractor matching system of the cloud control center retrieves an autonomous tractor corresponding to the model of the aircraft to be towed according to the aircraft to be towed that has issued the application, and searches for the tractor closest to the aircraft to be towed among the same type of idle autonomous tractors, and sends the assignment command and the aircraft to be towed information to the tractor; S3, a communication unit on an idle autonomous tractor The unit receives the assignment command and sends a wake-up signal to the autonomous driving platform on the autonomous driving tractor, which starts each unit and enters a standby state; S4, the cloud control center generates a real-time navigation map based on the airport scene map by processing the data collected by the multi-roadside perception unit in the intelligent roadside system, and sends it to the tractor. The multi-roadside perception unit includes a lidar and a camera, and the data includes point cloud data collected by the lidar and video data collected by the camera; S5, the autonomous driving platform obtains the starting point of the towing operation according to the information of the aircraft to be towed, and plans the best route to the starting point according to the real-time navigation map; the autonomous driving tractor drives to the starting point according to the best route.

Preferably, the communication unit of the autonomous driving platform remains online for a long time, while other units of the autonomous driving platform are in standby state for a long time. When the communication unit receives the assignment command sent by the cloud control center, it sends a wake-up signal to other units that have been in standby state for a long time, so that the automatic driving tractor enters a ready state.

Compared with the prior art, the present invention can realize fully automatic docking between the self-driving tractor and the aircraft to be towed, thereby ensuring the safety of operators and reducing manual labor while improving operating efficiency and accuracy.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

FIG1 is a flowchart of the steps of a method for dispatching and navigating an automatic driving tractor for an airport according to the present invention;

FIG2 is an architecture diagram of an airport autonomous driving tractor dispatching and navigation system according to the present invention.

Detailed ways

In order to make the purpose, technical solution and advantages of the embodiments of the present application clearer, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

To facilitate understanding of the embodiments of the present application, further explanation will be given below with reference to specific embodiments in conjunction with the accompanying drawings. The embodiments do not constitute a limitation on the embodiments of the present application.

Example 1

This embodiment provides an airport automatic driving tractor dispatching and navigation system, refer to FIG. 2 .

The airport autonomous driving tractor dispatching and navigation system includes a cloud control center, an intelligent roadside system and an autonomous driving platform. The cloud control center, the intelligent roadside system and the autonomous driving platform transmit information through 5G AeroMACS (Aeronautical Mobile Airport Communications System).

The cloud control center includes a communication system, an information processing system, a tractor matching system and a real-time navigation map generation system.

The communication system communicates with the autonomous driving platform and the intelligent roadside system via 5G AeroMACS.

The information processing system obtains all flight information of the day from the airport operation control center, sorts the information in chronological order from early to late according to the estimated departure time of each aircraft, and arranges the flight information in the system to form a waiting queue for departing aircraft.

The tractor matching system includes tractors of various models corresponding to various aircraft models. The tractor matching system will retrieve idle tractors corresponding to the aircraft models for which an application has been issued, and match them in real time.

The real-time navigation map generation system first abstracts the airport scene map into a scene network diagram, and the scene network diagram includes nodes and paths.

The communication system transmits the data received by the multiple roadside perception units in the intelligent roadside system to the real-time map generation system, and generates a real-time navigation map by processing the point cloud data and video data collected by the multiple roadside perception units, marking obstacles and prompt information on the basis of the scene network diagram.

A smart roadside system, the smart roadside system includes a roadside communication unit, a multi-element roadside perception unit and marking lines.

The roadside communication unit communicates with the cloud control center and the autonomous driving platform via 5G AeroMACS.

The multi-element roadside perception unit includes a laser radar and a camera, and the laser radar and the camera are installed on the airport surface.

The laser radar uses a 32-line laser radar with a horizontal field of view of 360° and a vertical field of view of 40° respectively. The laser radar is used to collect point cloud data of the location, and the point cloud data is used to assist in generating a real-time navigation map.

The camera uses a CMOS camera, which has the advantages of flexible image capture, high sensitivity, wide dynamic range and high resolution, and is used to collect airport scene video images.

The marking lines are used to provide information about the environment and define lanes, and to assist the autonomous tractor to move to the target location on the airport surface. The marking lines are recognized by the camera on the autonomous driving platform. For example, the camera recognizes lane lines to assist the tractor in lane keeping; dynamic traffic signs similar to traffic lights are set at the intersection.

The autonomous driving platform includes multiple sensors, a communication unit, a perception unit, a decision-making planning unit and a control unit, and is the executor of the method in the embodiment of the present invention.

The multi-sensor includes a high-precision positioning module, a camera and a lidar.

The high-precision positioning module is composed of a satellite antenna, an inertial/satellite combined navigation host and a host computer software, which are respectively used to receive satellite signals, calculate and provide multi-parameter navigation information, assist in positioning and analyze data, etc.

The camera uses a GigE camera to transmit image data at high speed, and is used to collect video images in front of the tractor and identify marking lines.

The laser radar uses a 32-line laser radar with a horizontal field of view of 360° and a vertical field of view of 40° respectively to detect obstacles around the tractor and send environmental information to the decision-making planning unit.

A communication unit, wherein the communication unit communicates with the cloud control center and the intelligent roadside system via 5G AeroMACS.

The perception unit is used to receive information obtained by multiple sensors and perform distributed fusion on it, that is, first locally process the raw data obtained by each independent sensor, and then send the result to the information fusion center for intelligent optimization and combination to obtain the final result.

The decision-making planning unit is divided into three levels: path planning, behavior decision and motion planning. First, the path planning layer generates a global path. After receiving the global path, the behavior decision layer combines the information received from the perception unit to make specific behavior decisions. Finally, the motion planning layer plans and generates a trajectory that meets specific constraints based on the specific behavior decisions. This trajectory is used as the input of the control unit to determine the final driving path of the vehicle.

The path planning layer is to plan the global path, also known as navigation planning. Here, the A* algorithm is used for path planning. The A* algorithm is the most effective direct search method for solving the shortest path in a static road network. The real-time navigation map at a certain moment can be regarded as a static road network. First, the area to be searched is simplified into small squares. The path finally found is a combination of some small squares. The initial node and the target node represent the starting point and the end point of the path respectively. The A* algorithm calculates the priority of each node through the heuristic function f(n) = g(n) + h(n). f(n) is the comprehensive priority of node n. When selecting the next node to be traversed, the node with the highest comprehensive priority (the smallest value) will always be selected. g(n) is the cost of node n from the starting point. h(n) is the estimated cost of node n from the end point. During the operation of the A* algorithm, each time the node with the smallest f(n) value (the highest priority) is selected from the priority queue as the next node to be traversed, and finally a shortest path is determined.

Behavioral decision-making, also known as behavioral planning, plans reasonable driving behavior based on the global planning route information, the current traffic scene and environmental perception information, and the current driving status of the vehicle, under the constraints of traffic rules. A hierarchical finite state machine is used here. A finite number of states are constructed in the finite state machine, and external input can only make the state machine switch between them. The hierarchical finite state machine contains the following parts: 1. Input set: also called stimulus set, containing all inputs that the state machine may receive; 2. Output set: that is, the set of responses that the state machine can make; 3. Use a directed graph to describe the internal state and transition logic of the state machine; 4. The state machine has a fixed initial state; 5. End state set; 6. Transition logic: that is, the conditions for the state machine to transfer from one state to another.

Motion planning is to plan and determine the desired motion trajectory of the vehicle in local space and time, including driving trajectory, speed, direction and state, based on local environmental information, upper-level decision-making tasks and real-time posture information of the vehicle body, while satisfying certain kinematic constraints. The expected vehicle speed and drivable trajectory output by the plan are fed into the control unit, which can ultimately generate a series of specific control signals for the vehicle to enable the vehicle to drive according to the planned objectives.

The control unit adopts PID control to generate control commands for the throttle, brake, steering wheel and gear lever of the tractor according to the planned driving trajectory and speed as well as the current position, posture and speed, so that the tractor can travel along the target trajectory at the target speed and acceleration.

The cloud control center, the smart roadside system and the autonomous driving platform establish a communication network through 5G AeroMACS by the communication units included in their respective systems to exchange data.

The laser radar and camera installed in the airport collect point cloud data and video data, and transmit the collected data to the communication system of the cloud control center through the roadside communication unit. The cloud control center processes the received data information, and the real-time navigation map generation system generates a real-time map based on the airport scene map, and sends the updated map to the communication unit on the automatic driving platform through the communication system. The decision-making planning unit combines the received updated map information with the information obtained by the perception unit to plan and generate a trajectory that meets specific constraints. This trajectory serves as the input of the control unit to determine the final driving path of the vehicle.

Example 2

This embodiment provides an airport automatic driving tractor dispatching and navigation method, refer to FIG. 1 .

S1, the cloud control center obtains flight information, arranges all flight information in chronological order to form an aircraft waiting queue, and the towed aircraft sends a push-out application before its scheduled departure, and sends its own information to the cloud control center at the same time, the information includes aircraft model, location, attitude and target location.

The cloud control center obtains all flight information for the day from the airport operation control center, sorts it in chronological order from early to late according to the expected departure time of each aircraft, and arranges the flight information in the system to form a waiting queue for departing aircraft.

The aircraft to be towed sends a push-out request T minutes before the departure time. T can be set according to the distance between the aircraft and the parking place of the towing vehicle. Preferably, in this embodiment, the value is [30,40]. At the same time, the aircraft sends its own information to the cloud control center. The aircraft model in the information is used to match the aircraft waiting queue, and the position, attitude and target position are used to provide the towing vehicle for navigation.

S2, the tractor matching system of the cloud control center retrieves the autonomous tractor corresponding to the model of the aircraft to be towed according to the aircraft to be towed that has applied for towing, and searches for the autonomous tractor closest to the aircraft to be towed among the idle autonomous tractors of the same type, and sends the assignment command and the aircraft information to the autonomous tractor.

In the tractor matching system of the cloud control center, aircraft and tractors are matched in real time.

The tractor matching system has multiple types of tractors, corresponding to different types of aircraft. According to the model of the aircraft to be towed that has been applied for, the corresponding automatic driving tractor is matched in real time. The automatic driving tractor has two states: working and idle.

The tractor matching system searches for the idle tractors that are closest to the aircraft to be towed, and sends the assignment command and the aircraft information to the tractor through the communication system.

S3, the communication unit on the idle autonomous driving tractor receives the assignment command and sends a wake-up signal to the autonomous driving platform on the autonomous driving tractor, and the autonomous driving platform starts each unit and enters a standby state.

The autonomous driving platform includes a communication unit, which remains online for a long time, and other units of the autonomous driving platform are in standby state for a long time.

After receiving the assignment command sent by the cloud control center, the communication unit sends a wake-up signal to the automatic driving platform, the automatic driving platform starts the perception unit, the decision-making planning unit and the control unit, and the automatic driving tractor enters a standby state.

S4, the cloud control center generates a real-time navigation map based on the airport scene map by processing the data collected by the multi-roadside perception unit in the intelligent roadside system, and sends it to the tractor. The multi-roadside perception unit includes a lidar and a camera, and the data includes point cloud data collected by the lidar and video data collected by the camera.

The laser radar and the camera are installed on the airport surface.

The real-time navigation map generation system of the cloud control center can provide navigation assistance for the tractor. First, the airport scene map is abstracted into a scene network diagram, which includes nodes and paths; then, by processing the point cloud data and video data collected by the multi-level roadside perception unit, obstacles and prompt information are marked on the basis of the scene network diagram, thereby generating a real-time navigation map and sending it to the tractor.

S5, the automatic driving platform obtains the starting point of the towing operation according to the information of the aircraft to be towed, and plans the best route to the starting point according to the real-time navigation map; the automatic driving towing vehicle drives to the starting point according to the best route.

The starting point of the towing operation is set at L m in front of the aircraft to be towed, where L is 10-15 m. From this position, the towing vehicle starts the subsequent docking operation with the aircraft and towing operation.

Based on the real-time navigation map provided by the cloud control center, the decision-making and planning unit of the autonomous driving platform plans the best route to the starting point of the towing operation.

The autonomous driving platform includes a decision-making planning unit, which uses the A* algorithm to perform path planning, constructs feasible paths from the starting point to the end point based on the known environment map and obstacle information in the map, and selects the one with the shortest driving time as the best route to the starting point of the traction operation.

It should be noted that there may be obstacles on the best route. Static obstacles require the tractor to avoid them autonomously, and dynamic obstacles will cause a waiting time. Regardless of the situation, the best route can always reach the starting point of the traction operation.

As the tractor drives to the starting point of the traction operation, the multiple sensors of the autonomous driving platform continuously collect information about the surrounding environment and feed it back to the perception unit of the autonomous driving platform, which is then sent to the decision-making and planning unit after processing.

Among them, the high-precision positioning module realizes high-precision positioning, the camera collects video images in front of the tractor and identifies sign lines, and the lidar collects point cloud data around the tractor.

At the same time, when the tractor is driving to the starting point of the traction operation, the roadside communication unit of the intelligent roadside system continuously sends the local environmental information to the autonomous driving platform, including whether there are obstacles, obstacle information, and passable time.

The decision-making and planning unit of the autonomous driving platform analyzes and processes the information received from the perception unit and the roadside communication unit, makes a decision, and then the control unit controls the movement state of the tractor.

By implementing the method of the embodiment of the present invention, it is possible to realize the dispatching and navigation of the automatic driving tractor at the airport, reduce the burden on the staff, and improve the safety and efficiency of the tractor operation.

The specific implementation methods described above further illustrate the purpose, technical solutions and beneficial effects of the present application in detail. It should be understood that the above description is only the specific implementation method of the present application and is not intended to limit the scope of protection of the present application. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present application should be included in the scope of protection of the present application.

Claims (3)

1. An airport autopilot tractor dispatch navigation system comprising:

the cloud control system comprises a tractor matching system, a real-time navigation map generating system and an information processing system;

the tractor matching system retrieves an autopilot tractor corresponding to the type of the aircraft to be towed and sends an assignment command and information of the aircraft to be towed to a tractor which is idle and closest to the aircraft to be towed; the real-time navigation map generation system abstracts an airport scene map into a scene network map, and generates a real-time navigation map through data information transmitted by the intelligent road side system; the information processing system acquires all flight information of the same day from an airport operation control center, sorts the flight information from the early to the late according to the predicted departure time of each aircraft and arranges the flight information in the system to form a departure aircraft waiting queue;

the intelligent road side system comprises a road side communication unit and a multi-element road side sensing unit;

the multi-road side sensing unit comprises a laser radar and a camera, point cloud data are acquired through the laser radar, and video data are acquired through the camera;

and an autopilot platform on the autopilot tractor, the autopilot platform comprising a multi-sensor, a sensing unit, a decision planning unit and a control unit; the multi-sensor respectively collects the position, the gesture, the video and the point cloud data of the aircraft to be towed; the sensing unit is used for acquiring data information transmitted by the multiple sensors in a distributed fusion manner; the decision planning unit is divided into three layers of path planning, behavior decision and motion planning, the path planning layer generates a global path, the path planning layer adopts an A-algorithm to generate the global path, a real-time navigation map at a certain moment can be regarded as a static road network, the road network is simplified into small squares, the combination of the small squares is a finally found path, an initial node and a target node are respectively a starting point and an end point of the path, the priority of each node is calculated through heuristic functions f (n) =g (n) +h (n), wherein f (n) is the comprehensive priority of the node n, when the node to be traversed next is selected, the node with the highest comprehensive priority is always selected, g (n) is the cost of the node n from the starting point, and h (n) is the predicted cost of the node n from the end point; after the behavior decision layer receives the path, the behavior decision is made by combining the received information of the sensing unit, a characteristic track is planned by the motion planning layer according to the behavior decision, the track is a final driving path planned by automatic driving traction, and the motion planning layer plans the expected motion track of the vehicle in the local space and time according to the environment information, the upper layer decision and the real-time pose information of the vehicle body, wherein the motion track comprises a track, a speed, a direction and a state; the control unit controls the tractor to travel according to the obtained track;

the cloud control center, the intelligent road side system and the autopilot platform are used for information transmission through a 5G aviation airport mobile communication system.

2. An airport automatic driving tractor dispatching navigation system according to claim 1, wherein during the operation of the a-algorithm, each time a node with highest priority is selected from the priority queue as a next node to be traversed, a shortest path is finally determined.

3. The airport automatic driving tractor dispatching navigation system of claim 2, wherein the cloud control center comprises a communication system, the intelligent road side system comprises a road side communication unit, the automatic driving platform comprises a communication unit, the road side communication unit transmits the point cloud data and the video data obtained by the multi-road side sensing unit to the communication system, and the cloud control center processes the received data to obtain an implementation navigation map and transmits the implementation navigation map to the communication unit through the communication system.