Airport unmanned luggage tractor path planning method and system

Abstract

The invention discloses an airport unmanned luggage tractor path planning method and system, and belongs to the technical field of unmanned vehicle autonomous navigation, and the method comprises the steps: obtaining an initial path and a final path of an airport unmanned luggage tractor; matching the global electronic map of the airport and the path information of the final path with the actual environment information, and judging whether abnormity exists or not; a local environment in an airport is divided into a plurality of two-dimensional grid units, various obstacles are identified in the two-dimensional grid units, potential energy values are given to the grid units corresponding to the obstacles respectively, when the potential energy values are larger than a preset potential energy maximum value, it is judged that abnormity occurs, and a driving path of the unmanned luggage tractor is re-planned and serves as a final path. The operation of judging whether the abnormity exists or not is executed again, and when the abnormity does not exist, the driving path of the unmanned luggage tractor is generated. The method can avoid local path planning of sudden dynamic obstacles.

Description

Technical Field

[0001] The present invention relates to the technical field of autonomous navigation of unmanned vehicles, and more specifically to a path planning method and system for an airport unmanned luggage tractor. Background Art

[0002] The "Roadmap for the Application of Unmanned Driving Equipment at Airports (2021-2025)" issued by the Civil Aviation Administration of China emphasizes the importance of the application of unmanned driving equipment at airports, aiming to promote the construction of smart civil aviation and four-type airports. The roadmap proposes a development vision with the goal of "application-oriented, quality and efficiency improvement, operational innovation, comprehensive verification, and international leadership", and clarifies the main tasks and goals at different stages. Unmanned baggage tractors are usually equipped with advanced sensors and cameras that can continuously monitor the surrounding environment, identify potential dangers more quickly and accurately than human drivers, and take measures to avoid collisions. Moreover, unmanned baggage tractors can work 24 hours a day without rest, which can significantly improve the efficiency of cargo handling and the continuity of airport operations.

[0003] In complex environments such as airports, driverless baggage tractors require precise path planning to avoid collisions with aircraft, other vehicles, staff or obstacles. Path planning can ensure that baggage tractors operate within the specified safety area and reduce the risk of accidents. By optimizing path planning, driverless baggage tractors can complete designated tasks in the shortest time, reduce unnecessary driving distance and time, and thus improve the operational efficiency of the entire airport.

[0004] An airport unmanned baggage tractor is an automated vehicle used for baggage and cargo transportation in an airport environment. Unmanned baggage tractors are usually equipped with advanced sensors and navigation systems, and are able to drive autonomously in specific areas of the airport without the need for human drivers. Unmanned baggage tractors come in different sizes and configurations, depending on the type and weight of the load they are designed to handle. Airport unmanned baggage tractors are essential ground support equipment for the aviation industry, ensuring safe, efficient and timely cargo loading and unloading operations.

[0005] In recent years, the navigation and path planning technology of autonomous delivery vehicles has made significant progress.

[0006] The Chinese patent with the patent number CN11816857A discloses a method and system for navigation and path planning of an autonomous unmanned delivery vehicle, including: obtaining and preprocessing the three-dimensional environmental information around the vehicle, generating a map of the area where the vehicle is located by using a map matching and feature extraction method based on the preprocessed information, using an angle formula and an area judgment strategy to identify obstacles and prohibited areas in the area, the distribution center specifies the delivery task and target location for the unmanned delivery vehicle, and plans the optimal path using a path planning strategy. The system sends navigation instructions to the unmanned delivery vehicle according to the optimal path, and compares the real-time collected environmental information with the map data. If an error occurs, compensation correction and feedback are performed, and safety protection is set. If an emergency occurs, the system needs to brake urgently or avoid obstacles. The vehicle arrives at the target location, delivers the goods to the consignee, and returns to the distribution center or waits for a new task. However, the above invention patent does not consider the problem of loading behind the vehicle. In the actual airport environment, there will be goods behind the baggage tractor. Ignoring the interaction between the goods behind and between the goods and the baggage tractor will lead to uncoordinated relative movement between the carriages in actual operation.

[0007] The Chinese patent with the patent number CN118329063A invented a lateral reference position decision method and vehicle for automatic driving path planning, which includes obtaining the driving boundary and offset decision of the vehicle allowed to drive in the lane; constructing a frenet coordinate system based on the center line of the lane; discretizing the left and right boundaries of the driving boundary in the longitudinal direction to obtain boundary point pairs; traversing each boundary point pair longitudinally to obtain the lateral limit positions corresponding to different boundary point pairs; obtaining the initial lateral reference position based on the offset decision and lateral limit position analysis; setting the obstacle detection boundary and obstacle container of the vehicle, traversing the obstacles in the longitudinal direction and classifying them; matching the avoidance decision of obstacles at different positions in the longitudinal direction according to the obstacle classification results and calculating the lateral reference position; filtering the lateral reference positions at different positions to obtain the filtered lateral reference position. It uses the lateral reference position as a reference for path planning to ensure safe and stable driving of the vehicle. However, the above invention patent does not consider the load problem. The load distribution of multiple carriages will directly affect the center of gravity position and steering characteristics of the baggage tractor. If the rear carriage is heavily loaded, it may cause the center of gravity to shift during steering, affecting the stability and steering radius of the baggage tractor.

[0008] In summary, in an airport environment, baggage tractors and cargo are heavy and have additional restrictions on straight-line speed and turning speed, so they cannot adapt to all types of environments. In addition, the airport environment is affected by factors such as weather and special obstacles such as special vehicles, resulting in poor adaptability of the final planned route. Summary of the invention

[0009] In view of the problems existing in the above-mentioned fields, the present invention proposes a path planning method and system for an airport unmanned baggage tractor, which divides the local environment in the airport into multiple two-dimensional grid units, identifies various obstacles in the two-dimensional grid units and assigns potential energy values ​​to the grid units corresponding to the obstacles, compares the potential energy value of the grid unit corresponding to each obstacle with a preset maximum potential energy value, and determines that it is abnormal when it is greater than the preset maximum potential energy value. The driving path of the unmanned baggage tractor is replanned and used as the final path, and the operation of determining whether there is an abnormality is performed again. When there is no abnormality, the driving path of the unmanned baggage tractor is generated. This method can avoid local path planning of sudden dynamic obstacles.

[0010] In order to solve the above technical problems, the present invention discloses a path planning method for an airport unmanned luggage tractor, comprising the following steps:

[0011] Get the initial path of the airport's driverless baggage tractor;

[0012] Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and the initial path is fitted using a high-order polynomial. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points.

[0013] Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is an abnormality; divide the local environment within the airport into multiple two-dimensional grid cells, identify various obstacles in the two-dimensional grid cells, and assign potential energy values ​​to the grid cells corresponding to the obstacles;

[0014] When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0015] Preferably, the identifying various obstacles in the two-dimensional grid cells and assigning potential energy values ​​to the grid cells corresponding to the obstacles respectively include:

[0016] Perform unmanned navigation tasks on unmanned baggage tractors through local path planning algorithms;

[0017] The local path planning algorithm includes the following steps:

[0018] The local environment in the airport is divided into multiple two-dimensional grid cells, each of which has a cell size of Δx×Δy. The resolution of the two-dimensional grid cell depends on the size and accuracy requirements of the unmanned baggage tractor.

[0019] Identify various obstacles in the two-dimensional grid cells, including other vehicles, buildings, and fences, assign potential energy values ​​P(x, y) to the two-dimensional grid cells of the obstacles, and set a maximum value P obstacle ;

[0020] The final position of the driverless baggage tractor is obtained as pgoal = (x goal ,y goal ), the potential energy value of the end position is U(x goal ,y goal ) takes the value of zero;

[0021] From this starting point, the minimum time or cost to each two-dimensional grid cell is calculated step by step using the Eikonal equation to generate the potential energy field. The Eikonal equation format is:

[0022]

[0023] in, is the gradient of the potential energy field, and F(p) is the velocity function;

[0024] From the current position of the unmanned baggage tractor pstart = (x start ,y start ) and follow the negative gradient direction of the potential energy field Move to generate the initial path p(t) to the end point:

[0025]

[0026] The generated initial path is smoothed, and the optimization goal is to minimize the cost function of the path length L and curvature k:

[0027]

[0028] in, represents the path speed, λ is the weight parameter, and k(p(t)) is the path curvature.

[0029] Preferably, the step of taking the replanned driving path of the unmanned baggage tractor as the final path comprises the following steps:

[0030] The baggage tractor is equipped with sensors to collect surrounding environment data in real time, including radar, camera and ultrasonic sensor;

[0031] Input the real-time collected surrounding environment data into the environment model, build a real-time two-dimensional grid map, and mark the location of obstacles on the map;

[0032] While planning the path, the potential energy value of each grid cell is calculated and the potential energy value of each grid cell is compared with the maximum value P obstacle Compare, when it is detected that the potential energy value of the grid unit is greater than the maximum value P obstacle When it is detected, it will be considered abnormal;

[0033] When an abnormality is detected, the obstacle avoidance distance is increased and the expected speed of the vehicle is reduced; the obstacle distribution in the environment is re-evaluated, and the optimal path is recalculated through the dynamic window method or the path planning strategy based on model prediction to continuously monitor the execution of the new path, and the path is adjusted in real time according to environmental changes, and the driving path of the unmanned baggage tractor is replanned; the re-planned driving path of the unmanned baggage tractor is used as the final path;

[0034] According to the final path, it is again determined whether there is any abnormal operation. When there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0035] Preferably, the step of obtaining the final path of the airport unmanned baggage tractor comprises the following steps:

[0036] The initial path of the airport unmanned baggage tractor is taken as the input condition, x(t) and y(t) are used to represent the points on the path, t is the path length, and are the first-order derivatives of the path in the x and y directions, respectively, and are the second-order derivatives of the path in the x and y directions respectively;

[0037] Define the path curvature as:

[0038]

[0039] Set the path curvature to a polynomial function:

[0040] k(t)=k0+at+bt 2

[0041] Among them, k0 is the initial curvature, and a and b are coefficients to be determined.

[0042] Define the steering angle θ(t) as:

[0043]

[0044] Get the initial position and initial direction, smooth the initial path, and generate the coordinates of the final path:

[0045]

[0046]

[0047] Preferably, the step of obtaining the initial path of the airport unmanned baggage tractor comprises the following steps:

[0048] According to the known global electronic map of the airport, the starting point and target point of the unmanned baggage tractor are obtained, and the nodes and obstacle sets of the search space are determined;

[0049] The initial path is generated through bidirectional greedy search. An adaptive step size is used to dynamically adjust the step size according to the distance between the node and the obstacle closest to the node. Lazy sampling is adopted. In the path optimization stage, invalid nodes that have no effect on the optimization result are deleted to form a collision-free optimal path from the starting point to the end point as the initial path.

[0050] Preferably, the actual environmental information of the airport unmanned baggage tractor is obtained by collecting dynamic information of the airport unmanned baggage tractor through an inertial navigation system INS, an odometer and a global satellite navigation system GNSS device, including the speed, position and posture of the vehicle.

[0051] Preferably, the method further includes sending speed and angular velocity information to the unmanned baggage tractor according to the generated driving path of the unmanned baggage tractor, and determining whether the unmanned baggage tractor has reached the destination. When the unmanned baggage tractor reaches the destination, the delivery is completed.

[0052] Preferably, it also includes a path planning system for an airport unmanned baggage tractor, comprising:

[0053] An initial path generation module is used to obtain the initial path of the airport's driverless baggage tractor;

[0054] The final path generation module is used to fit the initial path based on the initial path, with the minimum turning radius and path curvature of the unmanned baggage tractor as conditions, using a high-order polynomial to fit the initial path, and by adding speed and angular velocity information to the path points to obtain the final path of the airport unmanned baggage tractor;

[0055] Environmental perception module, used to obtain the actual environmental information of the airport's driverless baggage tractor;

[0056] The path planning module is used to match the path information of the global electronic map of the airport and the final path with the actual environment information to determine whether there is an abnormality; divide the local environment in the airport into multiple two-dimensional grid units, identify various obstacles in the two-dimensional grid units and assign potential energy values ​​to the grid units corresponding to the obstacles; when it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset potential energy maximum value, it is determined to be an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0057] Preferably, a computer device is also included, the computer device comprising a memory and a processor, the memory storing a computer program, and when the computer program is executed by the processor, the processor executes the following steps:

[0058] Get the initial path of the airport's driverless baggage tractor;

[0059] Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and the initial path is fitted using a high-order polynomial. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points.

[0060] Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is an abnormality; divide the local environment within the airport into multiple two-dimensional grid cells, identify various obstacles in the two-dimensional grid cells, and assign potential energy values ​​to the grid cells corresponding to the obstacles;

[0061] When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0062] Preferably, it further comprises a computer-readable storage medium, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor executes the following steps:

[0063] Get the initial path of the airport's driverless baggage tractor;

[0064] Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and the initial path is fitted using a high-order polynomial. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points.

[0065] Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is an abnormality; divide the local environment within the airport into multiple two-dimensional grid cells, identify various obstacles in the two-dimensional grid cells, and assign potential energy values ​​to the grid cells corresponding to the obstacles;

[0066] When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] The present invention proposes a path planning method for an airport unmanned baggage tractor. The method is based on the generated initial path, and is conditioned by the minimum turning radius and path curvature of the unmanned baggage tractor. In order to ensure the continuity of the path curvature, a high-order polynomial is used for fitting to achieve smoothing of the path. According to the actual environmental information of the airport unmanned baggage tractor, the global electronic map of the airport and the path information of the final path are matched with the actual environmental information to determine whether there is an abnormality; when there is an abnormality, the unmanned baggage tractor determines whether the path planning strategy needs to be adjusted according to the current environment and vehicle status to respond to the emergency. The local environment in the airport is divided into multiple two-dimensional grid units, various obstacles are identified in the two-dimensional grid units, and potential energy values ​​are assigned to the grid units corresponding to the obstacles; when the potential energy value of the grid unit corresponding to the obstacle is detected to be greater than the preset maximum potential energy value, it is judged as abnormal, and the driving path of the unmanned baggage tractor is replanned and used as the final path; according to the final path, the operation of judging whether there is an abnormality is performed again. When there is no abnormality, the driving path of the unmanned baggage tractor is generated, and the execution of the new path is continuously monitored by recalculating the optimal path, and the path is adjusted in real time according to environmental changes until the vehicle safely passes through the abnormal area. This method takes the global path as a condition, and can realize local path planning that dynamically avoids sudden dynamic obstacles in real time, and the generated driving path of the unmanned baggage tractor has better adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0069] Figure 1 This is a flow chart of the path planning method for the airport unmanned luggage tractor proposed by the present invention;

[0070] Figure 2 This is the architecture diagram of the path planning system for the airport unmanned baggage tractor proposed by the present invention;

[0071] Figure 3 This is a detailed diagram of the path planning system architecture of the airport unmanned baggage tractor proposed by the present invention. DETAILED DESCRIPTION

[0072] The following will be combined with the attached embodiment of the present invention Figure 1-Figure 3 , the technical solutions in the embodiments of the present invention are clearly and completely described. It should be understood that the terms described in the present invention are only used to describe specific implementation methods and are not used to limit the present invention.

[0073] The present invention proposes a path planning method for an airport unmanned luggage tractor, comprising the following steps:

[0074] Get the initial path of the airport's driverless baggage tractor;

[0075] Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and the initial path is fitted using a high-order polynomial. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points.

[0076] Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is an abnormality; divide the local environment within the airport into multiple two-dimensional grid cells, identify various obstacles in the two-dimensional grid cells, and assign potential energy values ​​to the grid cells corresponding to the obstacles;

[0077] When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0078] Based on this method, the present invention also proposes a path planning system for an airport unmanned baggage tractor, comprising:

[0079] An initial path generation module is used to obtain the initial path of the airport's driverless baggage tractor;

[0080] The final path generation module is used to fit the initial path based on the initial path, with the minimum turning radius and path curvature of the unmanned baggage tractor as conditions, using a high-order polynomial to fit the initial path, and by adding speed and angular velocity information to the path points to obtain the final path of the airport unmanned baggage tractor;

[0081] Environmental perception module, used to obtain the actual environmental information of the airport's driverless baggage tractor;

[0082] The path planning module is used to match the path information of the global electronic map of the airport and the final path with the actual environment information to determine whether there is an abnormality; divide the local environment in the airport into multiple two-dimensional grid units, identify various obstacles in the two-dimensional grid units and assign potential energy values ​​to the grid units corresponding to the obstacles; when it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset potential energy maximum value, it is determined to be an abnormality, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the re-planned driving path of the unmanned baggage tractor is used as the final path; according to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

[0083] This method can avoid local path planning of sudden dynamic obstacles, and the generated driving path of the unmanned baggage tractor is more adaptable.

[0084] Example

[0085] like Figure 1 As shown, an embodiment of the present invention provides a method for path planning of an airport unmanned luggage tractor, comprising the following steps:

[0086] Step S1: According to the task requirements, determine the starting point and the end point in the global electronic map of the airport, i.e. the initial position and target location inertial coordinate system.

[0087] Based on the reference distance between the vehicle and the target point, the globally optimal collision-free path planning is performed.

[0088] In this example, we initialize the parameters, define the starting point and target point, determine the search space and obstacle set, set the ellipse parameters, construct the transformation matrix L and rotation matrix C, and initialize two random trees.

[0089] The initial path is generated through a bidirectional greedy search, and an adaptive step size is used to dynamically adjust the step size according to the distance between the node and the nearest obstacle to avoid inefficiency or frequent collision detection due to too large or too small step size.

[0090] By adopting lazy sampling, invalid nodes that have no effect on the optimization results are deleted during the path optimization stage, thereby reducing the algorithm operation pressure and improving the convergence speed, and finally forming a collision-free optimal path from the initial point to the end point as the initial path of the airport unmanned baggage tractor.

[0091] Step S2: Considering that the airport unmanned tractor is not an ideal vehicle, path planning needs to meet specific constraints. Therefore, based on the minimum turning radius and path curvature of the unmanned luggage tractor, a high-order polynomial is used for fitting to achieve path smoothing in order to ensure the continuity of the path curvature.

[0092] After the path planning is completed, the vehicle speed and the angular velocity of the front wheel steering are discretized, and the planned path is fitted using a polynomial equation. The specific implementation steps are as follows:

[0093] (1) Take the initial path in step S1 as the input condition, use x(t) and y(t) to represent the points on the path, and t is the path length. and are the first-order derivatives of the path in the x and y directions, respectively, and are the second-order derivatives of the path in the x and y directions, respectively.

[0094] The curvature of the path is set to a polynomial function in order to better control the motion trajectory of the moving object. The path curvature is defined as:

[0095]

[0096] (2) In order to achieve continuous curvature, it is usually necessary to ensure the continuity of the curvature k(t), that is, there should be no sudden changes in curvature between path segments. To ensure continuity, the curvature is usually designed as a polynomial function:

[0097] k(t)=k0+at+bt 2

[0098] Among them, k0 is the initial curvature, and a and b are coefficients to be determined.

[0099] (3) Calculate the steering angle θ(t)

[0100]

[0101] Wherein, represents the steering angle of the vehicle at time t=0.

[0102] Angular velocity information is added to ensure that the moving object can move smoothly and accurately along the intended path. Angular velocity is directly related to the curvature of the path, because the greater the curvature, the faster the angular velocity the vehicle needs to turn in order to maintain the path.

[0103] (4) Given the initial position and initial direction, the smoothed path is generated by the following formula, that is, the coordinates of the final path of the airport unmanned baggage tractor are:

[0104]

[0105] Step S3: Acquire the actual environmental information of the airport's driverless baggage tractor, including the driverless baggage tractor collecting surrounding environment and self-status data through sensors such as lidar, inertial navigation, IMU, and odometer.

[0106] The collected data is preprocessed, including noise removal, filtering, coordinate conversion, etc.

[0107] Use LiDAR data to build high-precision three-dimensional maps and store map information. Use inertial navigation, IMU, odometer and other data combined with map information for precise positioning.

[0108] Set up an exception handling mechanism, use sensors such as LiDAR to detect obstacles in the surrounding environment, and determine the impact of obstacles on the path. According to the abnormal situation, choose the appropriate adjustment path planning strategy, including detour, obstacle avoidance, re-planning the path, etc.

[0109] Continuously monitor vehicle status and environmental changes, and update maps and path information in a timely manner to ensure that the vehicle continues to drive safely when adjusting the path, avoiding accidents such as collisions.

[0110] Based on real-time data, the path planning strategy is continuously optimized to improve the efficiency and robustness of path planning.

[0111] Step S4: Based on the global path, the environment is perceived by the laser radar to achieve dynamic real-time avoidance of sudden dynamic obstacles and perform local path planning.

[0112] Among them, the specific steps of local path planning are as follows:

[0113] (1) The local environment of the airport is divided into two-dimensional grid cells. The cell size of each grid is Δx×Δy. The resolution of the grid depends on the size and accuracy requirements of the baggage tractor.

[0114] Identify various obstacles in the grid, including other vehicles, buildings, fences, etc.

[0115] Give the obstacle grid a high potential energy value P(x, y) and set a maximum value P obstacle , which is used to compare with the calculated potential energy value of each grid cell to determine whether it is abnormal.

[0116] (2) Set the target point: define the target position of the baggage tractor pgoal = (x goal ,y goal ), and the potential energy value U(x goal ,y goal ) is set to zero.

[0117] (3) Starting from the target point, the Eikonal equation is used to gradually calculate the minimum time or cost to each grid point to generate a potential energy field. This process is similar to the diffusion of electromagnetic waves from the target point outward.

[0118] The format of the Eikonal equation is:

[0119]

[0120] in, is the gradient of the potential energy field, and F(p) is the velocity function.

[0121] (4) From the current position of the baggage tractor, pstart = (x start ,y start ) and follow the negative gradient direction of the potential energy field Move and generate the initial path p(t) to the target point.

[0122]

[0123] Among them, p(t+1) represents the new position of the baggage tractor at the next time step t+1.

[0124] (5) The generated path is further smoothed to reduce sharp turns and unnecessary path fluctuations to improve the stability and safety of the baggage tractor. The generated path p(t) is smoothed, and the optimization objective is to minimize the cost function of the path length L and the curvature k:

[0125]

[0126] in, represents the path speed, λ is the weight parameter, and k(p(t)) is the path curvature.

[0127] Determining whether there is an abnormality includes the following steps:

[0128] The baggage tractor is equipped with sensors to collect surrounding environment data in real time, including radar, camera and ultrasonic sensor;

[0129] Input the real-time collected surrounding environment data into the environment model, build a real-time two-dimensional grid map, and mark the location of obstacles on the map;

[0130] While planning the path, the potential energy value of each grid cell is calculated and the potential energy value of each grid cell is compared with the maximum value P obstacle Compare, when it is detected that the potential energy value of the grid unit is greater than the maximum value P obstacle When it is detected, it will be considered abnormal;

[0131] When an anomaly is detected, the avoidance distance to the obstacle will be increased and the expected speed of the vehicle will be reduced. The obstacle distribution in the environment will be re-evaluated, and the path planning strategy will be adjusted using the dynamic window method or model prediction. By recalculating the optimal path, the execution of the new path will be continuously monitored, and the path will be adjusted in real time according to environmental changes until the vehicle safely passes through the abnormal area.

[0132] Step S5: According to the generated driving path of the unmanned baggage tractor, the speed and angular velocity information is sent to the unmanned baggage tractor, and the unmanned baggage tractor is judged whether it has reached the end point. When it reaches the end point, the delivery is completed. After completing the cargo transportation task, the unmanned baggage tractor returns to the airport workshop or waits for a new delivery task at the original location.

[0133] like Figure 2 As shown, the present invention also proposes a path planning system for an airport unmanned luggage tractor, including: an environment perception module, a path planning module, a motion sensing and posture control module, a decision module and a positioning navigation module, wherein:

[0134] The environmental perception module is used to obtain the surrounding information of the unmanned vehicle's current environment.

[0135] The path planning module is used to provide a reference path for the motion planning of the unmanned vehicle.

[0136] The motion sensing and attitude control module is used to control the vehicle body in real time to change the direction and speed of travel according to the current motion state of the unmanned vehicle, the surrounding environment conditions, and the shape factors of the reference path ahead, so that the vehicle can run along the planned ideal path at the required speed.

[0137] The decision-making module is used to infer the behavioral information such as the movement direction and speed of obstacles in the future based on the perception information of obstacles at the current and historical moments, combined with environmental information such as road topology and traffic rules.

[0138] The positioning and navigation module is used to determine the location of the unmanned vehicle. After given the vehicle's current location and destination information, it combines the detailed data of the electronic map and real-time traffic flow information to intelligently plan and guide the unmanned vehicle to travel along the global optimal path, thereby ensuring efficient and accurate navigation services.

[0139] Specifically, the perception module includes a data acquisition unit and a data processing unit:

[0140] The data acquisition unit is used to collect environmental data through various sensors, including the location and status information of road signs, vehicles, pedestrians and other obstacles. The present invention adopts laser radar, millimeter wave radar and surround view camera to achieve this.

[0141] like Figure 3 As shown in the figure, the LiDAR is connected to the switch via an 8-core network cable, and the collected environmental data is transmitted to the switch via the 8-core network cable, and then transmitted to the AGX module via the 8-core network cable for processing and analysis. The LiDAR calculates the distance to surrounding objects by emitting laser pulses and measuring their return time, and creates a high-precision 3D point cloud map to help identify the precise location of the vehicle and environmental features.

[0142] After the millimeter-wave radar obtains information such as the target's position and motion status, it transmits it to the CAN1 interface of the AGX module through the CAN bus, and is processed and analyzed by the AGX. The millimeter-wave radar uses high-frequency electromagnetic waves to detect the position, speed and other properties of the target, and can work effectively even in severe weather conditions, providing additional safety redundancy for the system.

[0143] Surround view cameras are installed around the vehicle to directly connect the surrounding high-resolution video surveillance surround view cameras to the USB Type-c Hub, and transmit the captured video signals to the AGX module through the USB Type-c Hub for processing. The inertial GPS module is used to collect the real-time position and attitude information of the vehicle. It is connected to the AGX module via USB to transmit the position and attitude information to the AGX module. The body emergency stop is directly connected to the chassis. When the body emergency stop unit detects a situation that requires an emergency stop, it triggers a stop signal to stop the chassis. The industrial remote control emergency stop is directly connected to the chassis. When the industrial remote control emergency stop receives an emergency stop command, it transmits it to the chassis, and then the chassis performs the corresponding emergency stop action.

[0144] The surround view camera captures video streams through multiple high-definition cameras installed around the vehicle body, which work together to form a 360-degree panoramic view, helping to detect a wider range of objects and scene changes. The data acquisition unit achieves a comprehensive perception of the vehicle's surrounding environment through the coordinated work of multiple sensors.

[0145] The data processing unit, including the edge computing unit and the industrial computer, is used to process and analyze the data obtained from the sensors, and identify road signs and obstacle types in the current scene, as well as predict the behavioral intentions of other vehicles and pedestrians.

[0146] The path planning module includes a global path planning unit and a local path planning unit:

[0147] The global path planning unit is responsible for designing a global driving path for the unmanned vehicle from a macro perspective based on the preset clear goals and overall mission planning. This process divides the grand global task into a series of more specific and manageable subtasks based on a detailed terrain database and the specific requirements of the task. Each subtask points to an important milestone of the global goal, ensuring that the vehicle can gradually achieve the final goal.

[0148] The local path planning unit focuses on the execution details of these phased subtasks. Based on the specific requirements of the current subtask, combined with the vehicle's own status information (such as position, speed, direction, etc.) provided in real time by the data processing module and the immediate road condition data in front of the vehicle, it dynamically and in real time plans a most suitable and feasible driving path in the current environment. The present invention is implemented using Jetson AGX.

[0149] The motion sensing and posture control module includes a motion sensing unit, a posture control execution unit and a computing unit.

[0150] The motion sensing unit is responsible for collecting dynamic information of the smart vehicle, including speed, position and attitude, through devices such as the inertial navigation system (INS), odometer and global satellite navigation system (GNSS).

[0151] The attitude control execution unit is responsible for adjusting the vehicle's attitude to ensure that the vehicle can travel along a predetermined path. It consists of a set of reaction wheels, an offset wheel and a set of magnetic torque control devices.

[0152] The computing unit is responsible for processing the data collected by the attitude control system and performing necessary calculations.

[0153] The positioning and navigation module includes a positioning unit, an electronic map unit, a navigation control unit and a communication unit:

[0154] The positioning unit is used to receive data from GNSS, IMU, and wheel speed sensors to determine the precise position of the unmanned vehicle.

[0155] Electronic map unit, used to provide high-precision map data.

[0156] The navigation control unit is used to adjust the vehicle's speed and direction in real time so that the vehicle can travel along the planned route and handle emergencies.

[0157] The communication unit is responsible for data transmission with external systems, receiving dispatch instructions, sending vehicle status and location information, and strengthening remote control and monitoring.

[0158] The path planning method and system for the unmanned cargo and unmanned luggage tractor provided by the present invention can generate a driving path with enhanced adaptability and flexibility in the complex and changeable environment of an airport, thereby achieving efficient operation of the airport.

[0159] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

[0160] In addition, unless otherwise specified, all technical and scientific terms used in the present invention have the same meanings as commonly understood by those skilled in the art to which the present invention belongs. All documents mentioned in this specification are incorporated by reference to disclose and describe the methods related to the documents. In the event of any conflict with any incorporated document, the content of this specification shall prevail.

Claims

1. A method for path planning of an airport unmanned baggage tractor, characterized in that: The following steps are involved: Get the initial path of the airport's driverless baggage tractor; Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and a high-order polynomial is used to fit the initial path. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points. Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is any abnormality; The local environment of the airport is divided into a plurality of two-dimensional grid cells, various obstacles are identified in the two-dimensional grid cells, and potential energy values ​​are respectively assigned to the grid cells corresponding to the obstacles; When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as abnormal, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the replanned driving path of the unmanned baggage tractor is used as the final path; According to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

2. The method for path planning of an airport unmanned baggage tractor according to claim 1, characterized in that: The identifying various obstacles in the two-dimensional grid cells and assigning potential energy values ​​to the grid cells corresponding to the obstacles specifically includes: Perform unmanned navigation tasks on unmanned baggage tractors through local path planning algorithms; The local path planning algorithm includes the following steps: The local environment in the airport is divided into multiple two-dimensional grid cells, each of which has a cell size of Δx×Δy. The resolution of the two-dimensional grid cell depends on the size and accuracy requirements of the unmanned baggage tractor. Identify various obstacles in the two-dimensional grid cells, including other vehicles, buildings, and fences, assign potential energy values ​​P(x, y) to the two-dimensional grid cells of the obstacles, and set a maximum value P obstacle ; The final position of the driverless baggage tractor is obtained as pgoal = (x goal ,y goal ), the potential energy value of the end position is U(x goal ,y goal ) takes the value of zero; From this starting point, the minimum time or cost to each two-dimensional grid cell is calculated step by step using the Eikonal equation to generate the potential energy field. The Eikonal equation format is: in, is the gradient of the potential energy field, and F(p) is the velocity function; From the current position of the unmanned baggage tractor pstart = (x start ,y start ) and follow the negative gradient direction of the potential energy field Move to generate the initial path p(t) to the end point: The generated initial path is smoothed, and the optimization goal is to minimize the cost function of the path length L and curvature k: in, represents the path speed, λ is the weight parameter, and k(p(t)) is the path curvature.

3. The path planning method for an airport unmanned baggage tractor according to claim 2, characterized in that: The method of taking the replanned driving path of the unmanned baggage tractor as the final path includes the following steps: The baggage tractor is equipped with sensors to collect surrounding environment data in real time, including radar, camera and ultrasonic sensor; Input the real-time collected surrounding environment data into the environment model, build a real-time two-dimensional grid map, and mark the location of obstacles on the map; While planning the path, the potential energy value of each grid cell is calculated and the potential energy value of each grid cell is compared with the maximum value P obstacle Compare, when it is detected that the potential energy value of the grid unit is greater than the maximum value P obstacle When it is detected, it will be considered abnormal; When an abnormality is detected, the obstacle avoidance distance is increased and the expected speed of the vehicle is reduced; the obstacle distribution in the environment is re-evaluated, and the optimal path is recalculated through the dynamic window method or the path planning strategy based on model prediction to continuously monitor the execution of the new path, and the path is adjusted in real time according to environmental changes, and the driving path of the unmanned baggage tractor is replanned; the re-planned driving path of the unmanned baggage tractor is used as the final path; According to the final path, it is again determined whether there is any abnormal operation. When there is no abnormality, the driving path of the unmanned baggage tractor is generated.

4. The method for path planning of an airport unmanned baggage tractor according to claim 1, characterized in that: The method of obtaining the final path of the airport unmanned baggage tractor comprises the following steps: The initial path of the airport unmanned baggage tractor is taken as the input condition, x(t) and y(t) are used to represent the points on the path, t is the path length, and are the first-order derivatives of the path in the x and y directions, respectively, and are the second-order derivatives of the path in the x and y directions respectively; Define the path curvature as: Set the path curvature to a polynomial function: k(t)=k0+at+bt 2 Among them, k0 is the initial curvature, and a and b are coefficients to be determined. Define the steering angle θ(t) as: Get the initial position and initial direction, smooth the initial path, and generate the coordinates of the final path:

5. The method for path planning of an airport unmanned baggage tractor according to claim 1, characterized in that: The step of obtaining the initial path of the airport unmanned baggage tractor comprises the following steps: According to the known global electronic map of the airport, the starting point and target point of the unmanned baggage tractor are obtained, and the nodes and obstacle sets of the search space are determined; The initial path is generated through bidirectional greedy search. An adaptive step size is used to dynamically adjust the step size according to the distance between the node and the obstacle closest to the node. Lazy sampling is adopted. In the path optimization stage, invalid nodes that have no effect on the optimization result are deleted to form a collision-free optimal path from the starting point to the end point as the initial path.

6. The path planning method for an airport unmanned baggage tractor according to claim 1, characterized in that: The actual environmental information of the airport unmanned baggage tractor is obtained by collecting dynamic information of the airport unmanned baggage tractor through an inertial navigation system INS, an odometer, and a global satellite navigation system GNSS device, including the speed, position, and posture of the vehicle.

7. The method for path planning of an airport unmanned luggage tractor according to claim 1, characterized in that: It also includes sending speed and angular velocity information to the unmanned luggage tractor according to the generated driving path of the unmanned luggage tractor, judging whether the unmanned luggage tractor has reached the end point, and completing the delivery when the end point is reached.

8. A path planning system for an airport unmanned baggage tractor, characterized in that: include: An initial path generation module is used to obtain the initial path of the airport's driverless baggage tractor; The final path generation module is used to fit the initial path based on the initial path, with the minimum turning radius and path curvature of the unmanned baggage tractor as conditions, using a high-order polynomial to fit the initial path, and by adding speed and angular velocity information to the path points to obtain the final path of the airport unmanned baggage tractor; Environmental perception module, used to obtain the actual environmental information of the airport's driverless baggage tractor; The path planning module is used to match the global electronic map of the airport and the path information of the final path with the actual environment information to determine whether there are any anomalies; The local environment in the airport is divided into multiple two-dimensional grid units, various obstacles are identified in the two-dimensional grid units, and potential energy values ​​are assigned to the grid units corresponding to the obstacles; when it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as abnormal, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the replanned driving path of the unmanned baggage tractor is used as the final path; According to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

9. A computer device, characterized in that: The computer device includes a memory and a processor, wherein the memory stores a computer program, and when the computer program is executed by the processor, the processor performs the following steps: Get the initial path of the airport's driverless baggage tractor; Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and a high-order polynomial is used to fit the initial path. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points. Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is any abnormality; The local environment of the airport is divided into a plurality of two-dimensional grid cells, various obstacles are identified in the two-dimensional grid cells, and potential energy values ​​are respectively assigned to the grid cells corresponding to the obstacles; When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as abnormal, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the replanned driving path of the unmanned baggage tractor is used as the final path; According to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

10. A computer-readable storage medium, characterized in that: The computer readable storage medium stores a computer program, and when the computer program is executed by a processor, the processor performs the following steps: Get the initial path of the airport's driverless baggage tractor; Based on the initial path, the minimum turning radius and path curvature of the unmanned baggage tractor are used as conditions, and a high-order polynomial is used to fit the initial path. The final path of the airport unmanned baggage tractor is obtained by adding speed and angular velocity information to the path points. Obtain the actual environment information of the airport's driverless baggage tractor, match the airport's global electronic map and the path information of the final path with the actual environment information, and determine whether there is any abnormality; The local environment of the airport is divided into a plurality of two-dimensional grid cells, various obstacles are identified in the two-dimensional grid cells, and potential energy values ​​are respectively assigned to the grid cells corresponding to the obstacles; When it is detected that the potential energy value of the grid unit corresponding to the obstacle is greater than the preset maximum potential energy value, it is judged as abnormal, and the driving path of the unmanned baggage tractor is replanned by adjusting the path planning strategy, and the replanned driving path of the unmanned baggage tractor is used as the final path; According to the final path, the operation of determining whether there is an abnormality is performed again, and when there is no abnormality, the driving path of the unmanned baggage tractor is generated.

Images