Path planning method and device and electronic equipment

Through continuous iterative optimization and cross-learning of path populations, combined with dynamic programming, the problem that existing algorithms are prone to fall into local optimality is solved, and the efficiency and global optimization capabilities of autonomous mobile machines are improved.

CN119085655APending Publication Date: 2024-12-06UNIV OF ELECTRONICS SCI & TECH OF CHINA ZHONGSHAN INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411257642.X
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Priority Date
2024-02-01
Filing Date
2024-09-09
Publication Date
2024-12-06

AI Technical Summary

Technical Problem

Existing path planning algorithms such as dynamic window method and greedy search algorithms are prone to fall into local optimal solutions, resulting in autonomous mobile machines circling longer distances and being less efficient.

Method used

By generating path populations, continuously iterative optimization is performed according to driving direction parameters, combining heuristic design and multi-stage update ideas, path planning is optimized, cross-learning and dynamic planning are added, and global optimization is ensured.

Benefits of technology

It significantly improves the efficiency of autonomous mobile machine path planning, avoids local optimal traps, and ensures that the optimal path is found in complex environments.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN119085655A_ABST
    Figure CN119085655A_ABST
Patent Text Reader

Abstract

The invention provides a path planning method, a path planning device and electronic equipment, which are used for improving the problem of low efficiency of path planning of an autonomous mobile machine. The method comprises the following steps: generating a path population according to an environment map and path parameters of the autonomous mobile machine, wherein the path population comprises a plurality of candidate paths; calculating a driving direction parameter of the autonomous mobile machine according to the path parameter of the candidate path; updating the path population according to the driving direction parameters to obtain an updated path population; determining a fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and an environment map, and determining an optimal path according to the fitness values of the candidate paths; and iteratively updating the path population according to the path parameters of the optimal path until the fitness value of the optimal path is smaller than a preset threshold value, and obtaining the optimal path of the autonomous mobile machine in the environment map.
Need to check novelty before this filing date? Find Prior Art

Description

[0001] This application claims the priority of application number 202410146999.4, invention name “A path planning method, device and electronic device”, and application date February 1, 2024. Technical Field

[0002] The present application relates to the technical fields of robots, autonomous driving and path planning, and more specifically, to a path planning method, device and electronic equipment. Background Art

[0003] At present, most of the methods for path planning for autonomous mobile machines use the Dynamic Window Approach (DWA) and real-time obstacle avoidance algorithms based on greedy search to plan the paths of autonomous mobile machines. Autonomous mobile machines here include robots, self-driving cars, drones, and autonomous mobile engineering vehicles. In the specific practice process, it is found that most of these algorithms tend to focus on local-to-local paths, which makes it easy to fall into the local optimal solution, causing the autonomous mobile machine to detour a long distance and take too long to reach the target location. Therefore, the efficiency of using the current path planning algorithm to plan the path of autonomous mobile machines is low. Summary of the invention

[0004] The purpose of the embodiments of the present application is to provide a path planning method, device and electronic device for improving the problem of low efficiency in path planning for an autonomous mobile machine.

[0005] An embodiment of the present application provides a path planning method, including: generating a path population according to an environmental map and path parameters of an autonomous mobile machine, the path population including a plurality of candidate paths; calculating a driving direction parameter of the autonomous mobile machine according to the path parameters of the candidate paths; updating the path population according to the driving direction parameters to obtain an updated path population; determining a fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environmental map, and determining an optimal path according to the fitness values ​​of the candidate paths; iteratively updating the path population according to the path parameters of the optimal path until the fitness value of the optimal path is less than a preset threshold, thereby obtaining the optimal path of the autonomous mobile machine in the environmental map. In the implementation process of the above scheme, a path population is generated through the environmental map and path parameters of the autonomous mobile machine, and the path population is updated according to the driving direction parameters calculated according to the path parameters of the candidate paths in the path population, which effectively transforms the multi-constrained path planning problem from the starting point to the target point into a continuous iterative optimization problem of the driving direction parameters. It combines the idea of ​​heuristic design and multi-stage continuous iterative population update to take into account the real-time of local optimization and the overall optimality of global planning, greatly improving the efficiency of path planning for autonomous mobile machines.

[0006] Optionally, in an embodiment of the present application, before updating the path population according to the driving direction parameter, it also includes: selecting the optimal path from multiple candidate paths; and cross-learning the optimal path and each candidate path in the multiple candidate paths. In the implementation process of the above scheme, by cross-learning the optimal path and each candidate path in the multiple candidate paths, the candidate paths are effectively cross-learned from the optimal path, and partial features of two or more excellent solutions are combined to generate a new solution, which significantly enhances the diversity of the population, thereby helping the search algorithm to jump out of the local optimum and effectively improving the probability of exploring the global optimal solution.

[0007] Optionally, in an embodiment of the present application, the driving direction parameters include: steering angle and heading angle; updating the path population according to the driving direction parameters includes: determining the latest road section according to the steering angle and heading angle; and updating the path population using the latest road section. In the implementation of the above scheme, by determining the latest road section according to the steering angle and heading angle, and updating the path population using the latest road section, the originally independently updated path is deeply intervened, which helps to overcome the premature convergence problem caused by relying solely on individual extreme values ​​and global extreme values, so that the algorithm can jump out of the local optimal trap to a greater extent, and improve the possibility of finding the global optimal path.

[0008] Optionally, in an embodiment of the present application, the path population is updated using the latest road segment, including: for each candidate path in the path population, determining whether the number of road segments of the candidate path is greater than a preset threshold; if so, replacing the terminal road segment in the candidate path with the latest road segment, otherwise, adding the latest road segment after the terminal road segment of the candidate path. In the implementation of the above scheme, by replacing or adding the latest road segment according to the number of road segments in the candidate path, a road segment in the candidate path is replaced with a known optimal road segment to reach the target point, ensuring that the final complete path can minimize unnecessary distance or cost, thereby combining the idea of ​​dynamic programming optimization and the improvement of global optimization performance, greatly improving the efficiency of path planning for autonomous mobile machines.

[0009] Optionally, in an embodiment of the present application, the autonomous mobile machine is an engineering vehicle, and the path parameters include: the starting position point, starting direction, wheelbase, vehicle length, target position point and target parking direction of the engineering vehicle; the fitness value of each candidate path in the updated path population is determined according to the path parameters of the candidate path and the environmental map, including: for each candidate path in the path population, the fitness value of the candidate path is calculated according to the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction and environmental map. In the implementation process of the above scheme, in the real-time path planning scenario, the fitness value can reflect the changes in the current environmental information, so that the algorithm can quickly respond to factors such as obstacle movement and traffic conditions, and continuously search for the optimal path in a changing environment, ensuring that the algorithm can efficiently find a satisfactory optimal path in a complex environment, greatly improving the efficiency of path planning for autonomous mobile machines.

[0010] Optionally, in an embodiment of the present application, determining the optimal path according to the fitness value of the candidate path includes: if the fitness value of the candidate path is less than the fitness value of the optimal path, determining the candidate path as the optimal path.

[0011] The embodiment of the present application also provides a path planning device, including: a path population generation module, which is used to generate a path population according to the environmental map and path parameters of the autonomous mobile machine, and the path population includes multiple candidate paths; a direction parameter calculation module, which is used to calculate the driving direction parameters of the autonomous mobile machine according to the path parameters of the candidate paths; a path population update module, which is used to update the path population according to the driving direction parameters to obtain an updated path population; a fitness value determination module, which is used to determine the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environmental map, and determine the optimal path according to the fitness values ​​of the candidate paths; an optimal path acquisition module, which is used to iteratively update the path population according to the path parameters of the optimal path, until the fitness value of the optimal path is less than a preset threshold, and the optimal path of the autonomous mobile machine in the environmental map is obtained.

[0012] Optionally, in an embodiment of the present application, the path planning device further includes: an optimal path screening module, used to screen out an optimal path from multiple candidate paths; and a path cross-learning module, used to perform cross-learning on the optimal path and each candidate path in the multiple candidate paths.

[0013] Optionally, in an embodiment of the present application, the driving direction parameters include: a turning angle and a heading angle; a path population update module includes: a latest road section determination submodule, used to determine the latest road section based on the steering angle and the heading angle; a path population update submodule, used to update the path population using the latest road section.

[0014] Optionally, in an embodiment of the present application, the path population update submodule includes: a segment number judgment unit, which is used to judge whether the number of segments of each candidate path in the path population is greater than a preset threshold; a segment replacement and addition unit, which is used to replace the terminal segment in the candidate path with the latest segment if the number of segments of the candidate path is greater than the preset threshold; if the number of segments of the candidate path is less than or equal to the preset threshold, add the latest segment after the terminal segment of the candidate path.

[0015] Optionally, in an embodiment of the present application, the autonomous mobile machine is an engineering vehicle, and the path parameters include: the starting position point, starting direction, wheelbase, vehicle length, target position point and target parking direction of the engineering vehicle; the fitness value determination module includes: a fitness value calculation submodule, which is used to calculate the fitness value of each candidate path in the path population according to the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction and environmental map.

[0016] Optionally, in an embodiment of the present application, the fitness value determination module includes: an optimal path determination submodule, which is used to determine the candidate path as the optimal path if the fitness value of the candidate path is less than the fitness value of the optimal path.

[0017] An embodiment of the present application further provides an electronic device, including: a processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions execute the method described above when executed by the processor.

[0018] An embodiment of the present application further provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the method described above is executed.

[0019] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings required for use in the embodiments of the present application will be briefly introduced below. It should be understood that the following drawings only show certain embodiments of the embodiments of the present application and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0021] Figure 1 A schematic diagram of a process flow of a path planning method provided by an embodiment of the present application is shown;

[0022] Figure 2 A schematic diagram of an environment map of an autonomous mobile machine in simulation software provided by an embodiment of the present application is shown;

[0023] Figure 3 A schematic diagram of an arc section provided in an embodiment of the present application is shown;

[0024] Figure 4 The latest road section schematic diagram in the rectangular coordinate system provided by the embodiment of the present application is shown;

[0025] Figure 5 A schematic diagram of the structure of a path planning device provided in an embodiment of the present application is shown;

[0026] Figure 6 A schematic diagram of the structure of an electronic device provided in an embodiment of the present application is shown. DETAILED DESCRIPTION

[0027] To make the purpose, technical scheme and advantages of the embodiment of the present application clearer, the technical scheme in the embodiment of the present application will be clearly and completely described below in conjunction with the drawings in the embodiment of the present application. It should be understood that the drawings in the embodiment of the present application only serve the purpose of explanation and description, and are not used to limit the protection scope of the embodiment of the present application. In addition, it should be understood that the schematic drawings are not drawn in real proportion. The flowchart used in the embodiment of the present application shows the operation implemented according to some embodiments of the embodiment of the present application. It should be understood that the operation of the flowchart can be implemented out of order, and the steps without logical context can be reversed in order or implemented simultaneously. In addition, those skilled in the art can add one or more other operations to the flowchart under the guidance of the content of the embodiment of the present application, or remove one or more operations from the flowchart.

[0028] In addition, the described embodiments are only a part of the embodiments of the present application, rather than all embodiments. The components of the embodiments of the present application described and shown in the drawings generally here can be arranged and designed in various configurations. Therefore, the following detailed description of the embodiments of the present application provided in the drawings is not intended to limit the scope of the embodiments of the present application claimed, but merely represents the selected embodiments of the embodiments of the present application.

[0029] It is understandable that the "first" and "second" in the embodiments of the present application are used to distinguish similar objects. Those skilled in the art will understand that the words "first" and "second" do not limit the quantity and execution order, and the words "first" and "second" do not necessarily limit the difference. In the description of the embodiments of the present application, the term "and / or" is merely a description of the association relationship of associated objects, indicating that there may be three relationships, such as A and / or B, which can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, the character " / " in this article generally indicates that the related objects before and after are in an "or" relationship. The term "multiple" refers to more than two (including two), and similarly, "multiple groups" refers to more than two groups (including two groups).

[0030] It should be noted that the path planning method provided in the embodiment of the present application can be executed by an electronic device, where the electronic device refers to a device terminal or server with the function of executing a computer program, such as a smart phone, a personal computer, a tablet computer, a personal digital assistant, or a mobile Internet device. A server refers to a device that provides computing services through a network, such as an x86 server and a non-x86 server, and a non-x86 server includes a mainframe, a minicomputer, and a UNIX server.

[0031] Among some related technologies, some use the Dynamic Window Approach (DWA) for path planning. The basic working principle of the DWA algorithm is to sample in the robot's velocity space, then evaluate the future motion trajectory corresponding to each sampling point based on the current environmental information, and select the trajectory with the highest score as the next action to be executed. In a complex and dynamically changing environmental map, the initial path selection may be inappropriate, resulting in a long detour, and the entire path planning process from the starting point to the target point deviates from the ideal result. Since this DWA algorithm has limited adaptability to changes and only performs local searches in a limited search range such as a dynamic window, it is easy to fall into a local optimal solution when using the DWA algorithm for path planning.

[0032] In some related technologies, some real-time obstacle avoidance algorithms based on greedy search are used for path planning. The basic characteristic of greedy algorithms is that each step takes the currently optimal option without considering the impact that may have on the global optimal solution in the future. In the real-time obstacle avoidance problem, the greedy strategy usually makes an immediate decision based on the environmental information and obstacle positions detected by the current sensor, that is, chooses the path that seems to be the closest or the lowest cost at the moment. However, the greedy algorithm is not suitable for complex and dynamically changing environmental maps, especially in scenes with dynamic changes or multiple obstacles and complex spatial structures. Relying only on current local information for decision-making may cause the algorithm to be unable to foresee the long-term situation, thereby selecting a suboptimal branch on the overall optimal path that seems to be the best but is actually the best. The greedy algorithm only focuses on how to pass through the current narrow channel as quickly as possible and ignores the better path in the subsequent large-scale open space. Therefore, it is also easy to fall into the local optimal solution when using the greedy algorithm for path planning.

[0033] In response to the above problems, in an embodiment of the present application, the path planning of the autonomous mobile machine is modeled by the driving direction parameters, that is, the environmental map and path parameters of the autonomous mobile machine are used to generate a path population, and the driving direction parameters calculated according to the path parameters of the candidate paths in the path population are used to update the path population, thereby effectively converting the multi-constraint path planning problem from the starting point to the target point into a continuous iterative optimization problem of the driving direction parameters, thereby combining the idea of ​​heuristic design and multi-stage continuous iterative update of the population to take into account the real-time nature of local optimization and the overall optimality of global planning, greatly improving the efficiency of path planning for autonomous mobile machines.

[0034] See also Figure 1The schematic diagram of the flow chart of the path planning method provided by the embodiment of the present application is shown; the main idea of ​​the path planning method is to transform the multi-constraint path planning problem from the starting point to the target point into a continuous iterative optimization problem of the driving direction parameters, thereby combining the idea of ​​heuristic design and multi-stage continuous iterative population update to simultaneously take into account the real-time nature of local optimization and the overall optimality of global planning. The implementation methods of the above path planning method may include:

[0035] Step S110: Generate a path population according to the environment map and path parameters of the autonomous mobile machine, wherein the path population includes a plurality of candidate paths.

[0036] Autonomous mobile machines refer to machines that require path planning when moving autonomously, including but not limited to: robots, autonomous vehicles (such as driverless cars), drones, and engineering vehicles. The environment map is a map of the environment where the autonomous mobile machine is located. The environment map here can be an environment map in simulation software or an environment map in a real scene.

[0037] See also Figure 2 The schematic diagram of the environment map of the autonomous mobile machine provided by the embodiment of the present application in the simulation software is shown; the path parameters of the autonomous mobile machine refer to the path-related parameters used to generate the path population, such as the starting position point, starting direction, target position point and target parking direction of the autonomous mobile machine. In the problem field of path planning, the path population can include multiple candidate paths (which can also be understood as individual particles), and each candidate path (individual particle) represents a path that may become the optimal path.

[0038] Step S120: Calculate the driving direction parameters of the autonomous mobile machine according to the path parameters of the candidate paths.

[0039] It is understandable that, in the process of path planning of the autonomous mobile machine, the autonomous mobile machine itself or the path planning task has a constraint space of the driving direction, and the constraint space of the driving direction can be represented by the driving direction parameter. For example, assuming that the autonomous mobile machine is a vehicle-type machine such as an unmanned vehicle or an engineering vehicle, the constraint space of the steering angle of the vehicle-type machine itself can be represented as (S min ,S max ), for another example: assuming that the starting position and the target position of the autonomous mobile machine are determined, the heading angle can be set to be within a preset deviation angle range from the starting position and the target position during driving. Therefore, the constraint space of the heading angle of the autonomous mobile machine can be expressed as (H min ,H max ).

[0040] For ease of understanding and illustration, in the embodiments of the present application, an engineering vehicle is taken as an example for description. The engineering vehicle can be an articulated vehicle, that is, the steering of the engineering vehicle can be achieved by the rotation of a hinge. Of course, in the specific practice process, the selection of the autonomous mobile machine should not be construed as a limitation on the embodiments of the present application. For the engineering vehicle, a constraint space for the steering angle and a constraint space for the heading angle of the engineering vehicle on the environmental map can be set simultaneously. For example, each candidate path of the engineering vehicle can be modeled as a path composed of multiple road segments, and each road segment can be represented by a driving direction parameter as (s, h), where the steering angle s is within the range of (S min , S max ), and the heading angle h is within the range of (H min , H max ).

[0041] Step S130: Update the path population according to the driving direction parameters to obtain the updated path population.

[0042] Step S140: Determine the fitness value of each candidate path in the updated path population according to the path parameters of the candidate path and the environmental map, and determine the optimal path according to the fitness value of the candidate path.

[0043] Step S150: Iteratively update the path population according to the path parameters of the optimal path until the fitness value of the optimal path is less than a preset threshold, and obtain the optimal path of the autonomous mobile machine in the environmental map.

[0044] For example, the implementation manner of the above step S150 is as follows: Return to iteratively execute the driving direction parameters of the autonomous mobile machine calculated according to the path parameters of the candidate path in step S120 until the fitness value of the optimal path is less than the preset threshold, and obtain the optimal path of the autonomous mobile machine in the environmental map. Specifically, it can be expressed by the formula as min Cost(u, v) < stop_fitness, where min Cost(u, v) represents the fitness value of the optimal path, and stop_fitness represents the preset threshold of the fitness value for stopping iterative update.

[0045] In the implementation process of the above solution, a path population is generated through the environmental map and path parameters of the autonomous mobile machine, and the path population is updated according to the driving direction parameters calculated from the path parameters of the candidate paths in the path population, effectively transforming the multi-constraint path planning problem from the starting position point to the target position point into a continuous iterative optimization problem of the driving direction parameters, thereby combining the ideas of heuristic design and multi-stage continuous iterative update of the population to simultaneously take into account the real-time performance of local optimization and the overall optimality of global planning, and greatly improving the efficiency of path planning for the autonomous mobile machine.

[0046] As an optional implementation of the above-mentioned path planning method, in order to facilitate the conversion between coordinate systems, a conversion mapping relationship between coordinate systems can also be established. The process of establishing the conversion mapping relationship between the image pixel coordinate system and the rectangular coordinate system is, for example: the rectangular coordinate system can be mapped to the image pixel coordinate system using the formula I′=flip_Y(I), so that the upper left corner of the converted image is used as the origin of the rectangular coordinate system for easy understanding and application calculation; wherein, I′ represents the environment map in the rectangular coordinate system, flip_Y(·) is a flip function along the y-axis, and I represents the environment map in the image pixel coordinate system. It can be understood that the conversion mapping relationship between the real-world physical coordinate system and the image pixel coordinate system is also based on a similar principle, so the process of establishing the conversion mapping relationship will not be repeated.

[0047] See also Figure 3 The schematic diagram of the arc section provided in the embodiment of the present application is shown; assuming that the vehicle length of the engineering vehicle is L, the vehicle width is W, the steering angle of the engineering vehicle is denoted as s, and the wheelbase is denoted as C wheelbase , the radius of the arc section is represented by r; it is easy to see that the steering angle s is related to the radius r of the arc section and the wheelbase C wheelbase Constitute a trigonometric function relationship, which can be expressed as It is easy to see that the center angle corresponding to the above-mentioned arc section is the heading angle of the engineering vehicle, which can be expressed as h. Therefore, when the steering angle s and heading angle h of the engineering vehicle in a circular section are determined, the trajectory of the engineering vehicle in the latest section of the circular section is also determined.

[0048] As an optional implementation of the above step S110, for example, the path population can be initialized according to the environmental map and path parameters of the engineering vehicle. Assume that the number of path populations is represented by M, the number of candidate paths (particle individuals) in the path population is represented by N, and the constraint space of the steering angle is represented by (S min ,S max ), the heading angle constraint space is expressed as (H min ,H max ). If each segment in the candidate path is modeled as a segment in the form of an arc constructed by a turning angle (expressed as s) and a heading angle (expressed as h), and each segment is expressed as (s, h), then a candidate path can be selected from the path population and expressed as X 0 ={(s 1 ,h 1 ),(s 2 ,h 2 ),(s 3 ,h 3 ),...,(s d ,hd )},s j ∈(S min ,S max ),h j ∈H min ,H max ), where d is the number of initialized road segments, j is the jth candidate path in the path population, and 1≤j≤d.

[0049] As an optional implementation of the above path planning method, before updating the path population according to the driving direction parameter, cross learning may also be performed. This implementation may include:

[0050] Step S121: Filter out the optimal path from multiple candidate paths.

[0051] The implementation method of the above step S121 is, for example, after initializing the path population according to the environmental map and path parameters of the engineering vehicle, a candidate path can be randomly selected from multiple candidate paths as the initial optimal path. However, in the following process of iteratively updating the path population according to the path parameters of the optimal path, it is necessary to screen out the optimal path from multiple candidate paths according to the fitness value of the candidate path (the calculation process of the fitness value will be described below).

[0052] Step S122: cross-learning is performed on the optimal path in the path population and each candidate path in the plurality of candidate paths to obtain candidate paths after cross-learning.

[0053] The implementation method of the above step S122 is as follows: Assume that the candidate path in the path population is represented by X m,i (1≤m≤M,1≤i≤N), then according to the formula lr=(i / N)*0.95 and the formula X′ m,i =X best *lr+X m,i *(1-lr) Cross-learning is performed on the optimal path in the path population and each candidate path in multiple candidate paths; where lr represents the learning rate, N represents the number of candidate paths (particle individuals) in the path population, i represents the i-th candidate path in the path population, m represents the m-th path population in multiple path populations, and X m,i represents the i-th candidate path in the m-th path population, X best represents the optimal path in the path population, X′ m,i represents the i-th candidate path in the m-th path population after cross-learning.

[0054] Optionally, after obtaining the candidate paths after cross-learning, the formula X″ can also be used m,i =Clamp(X′ m,i) spatially constrains the steering angle and heading angle of the candidate path after cross-learning, where X′ m,i represents the i-th candidate path in the m-th path population after cross-learning, X″ m,i represents the i-th candidate path in the m-th path population after spatial constraints, and Clamp represents the spatial constraint. The spatial constraint here can be described using the following pseudo code.

[0055]

[0056] Referring to the pseudo code above, the spatial constraint can be to use a random function to constrain the upper and lower limits of the steering angle and the heading angle, where the upper and lower limits of the steering angle can be determined by the physical requirements of the engineering vehicle (for example, the maximum left and right swing angle is 37 degrees), and the heading angle is pre-set according to the specific task customization.

[0057] See also Figure 4 The latest road section schematic diagram in the rectangular coordinate system provided by the embodiment of the present application is shown; the origin of the rectangular coordinate system is O(x 0 ,y 0 ), the horizontal axis represents the east-west span of the engineering vehicle in the environment map, and the vertical axis represents the north-south span of the engineering vehicle in the environment map. Since the candidate path is modeled as a trajectory formed by the continuous splicing of multiple arc sections, when the steering angle and heading angle of the engineering vehicle in a circular section have been determined, the latest section can be determined based on the steering angle and heading angle. The following describes the process of calculating the steering angle and heading angle of the engineering vehicle in a circular section when the starting direction, starting position point and target position point of the circular section are known. Assume that the starting position point is represented by A(x 1 ,y 1 ), the starting direction vector is represented by u = (a, b), where the starting direction vector is the extended unit vector u of the engineering vehicle at the starting position, and the target position is represented by B (x 2 ,y 2 ), therefore, when calculating the arc section between the starting position point and the target position point, what is actually calculated is the steering angle and heading angle of the engineering vehicle in the latest section. The steering angle is calculated first, and then the heading angle of the engineering vehicle is calculated.

[0058] As an optional implementation of the above step S130, the above driving direction parameters may include: a steering angle and a heading angle; the above implementation of updating the path population according to the driving direction parameters may include:

[0059] Step S131: Determine the latest road section according to the steering angle and the heading angle.

[0060] The implementation method of the above step S131 is as follows: since the starting point A (x 1 ,y 1 ), starting direction vector u=(a,b), target position point B(x 2 ,y 2 ), we can draw a vertical line BD from point B, and extend the unit vector u of the starting direction vector to point D. The coordinates of point D can be expressed as (x d ,y d ), and then connect the line segment AB. Then, from the center of the arc segment O(x 0 ,y 0 ) Draw the perpendicular bisector OC of line segment AB, that is, the perpendicular bisector intersects AB at point C. Since line segment OA and line segment BD are both perpendicular to line segment AD, it is easy to prove that line segment OA is parallel to line segment BD, so angle OAC is equal to angle ABD. d =(a*b*x 2 +b 2 *y 2 +a 2 *y 1 -a*b*x 1 ) / (a 2 +b 2 ) and the formula x d =(a*y d -a*y 1 +b*x 1 ) calculate the coordinate value of point D, and according to the formula Calculate the radius of the arc section, and according to the formula x 0 =(2*a*x 1 *(y 2 -y 1 )+2*b*y 1 *(y 2 -y 1 )+b*x 1 2 -b*x 2 2 -b*y 2 2 +b*y 1 2 ) / (2*a*(y 2 -y 1 )+2*b*(x 1 -x 2 )) and formula y 0 =(x 1 -x 2 )*x 0 +(y 1 +y 2) / 2-(x 1 2 -x 2 2 ) / (2*(y 2 -y 1 )) Calculate the coordinates of the center of the arc section. Then, according to the above steering angle s, the radius r of the arc section and the wheelbase C wheelbase To form a trigonometric relationship, we can use the formula s = 2*tan -1 (C wheelbase / r)*180*π to calculate the steering angle; where s represents the steering angle of the engineering vehicle on the arc section, C wheelbase represents the wheelbase of the engineering vehicle, and r represents the radius of the circle corresponding to the arc section. Finally, according to the formula Calculate the heading angle of the engineering vehicle on the circular section; where h represents the heading angle of the engineering vehicle on the circular section, r represents the radius of the circle corresponding to the circular section, and x 1 and 1 They represent the horizontal and vertical coordinates of the starting point A of the engineering vehicle on the circular section, respectively. 2 and 2 They respectively represent the horizontal and vertical coordinates of the target position point B of the engineering vehicle on the circular arc section.

[0061] It is understandable that in the specific practice, the steering angle can be divided into clockwise and counterclockwise. When the steering angle is a positive value, it means that the steering angle is counterclockwise, and when the steering angle is a negative value, it means that the steering angle is clockwise. Specifically, according to the vector cross product formula To determine whether the steering angle is positive or negative, where s last Indicates that the latest road section can distinguish between clockwise and counterclockwise steering angles, s indicates the steering angle of the engineering vehicle on the arc section, and x 0 and 0 They represent the horizontal and vertical coordinates of the origin O of the arc section, respectively. 1 and 1 They respectively represent the horizontal and vertical coordinates of the starting point A of the engineering vehicle on the circular arc section, and (a, b) represents the starting direction vector of the engineering vehicle on the circular arc section.

[0062] Similarly, the heading angle can be divided into clockwise and counterclockwise according to whether the center of the arc section is the left or right center. For example, when the center of the arc section is the left center, the heading angle is positive, which means the heading angle is counterclockwise. When the center of the arc section is the right center, the heading angle is negative, which means the heading angle is clockwise. You can use the formula

[0063]

[0064] To determine whether the heading angle is positive or negative, h last Indicates that the latest road section can distinguish between clockwise and counterclockwise heading angles, h indicates the heading angle of the engineering vehicle on the arc section, x 0 and 0 They represent the horizontal and vertical coordinates of the origin O of the arc section, respectively. 1 and 1 They represent the horizontal and vertical coordinates of the starting point A of the engineering vehicle on the circular section, respectively. 2 and 2 They represent the horizontal and vertical coordinates of the target position point B of the engineering vehicle on the circular arc section, and abs(·) is the absolute value function.

[0065] Step S132: Use the latest road segment to update the path population.

[0066] As an optional implementation of the above step S132, the implementation of updating the path population using the latest road segment may include:

[0067] Step S132a: for each candidate path in the path population, determine whether the number of road sections of the candidate path is greater than a preset threshold.

[0068] Step S132b: If the number of road segments of the candidate path is greater than a preset threshold, the terminal road segment in the candidate path is replaced by the latest road segment.

[0069] Step S132c: If the number of segments of the candidate path is less than or equal to a preset threshold, then a newest segment is added after the terminal segment of the candidate path.

[0070] The implementation method of the above steps S132a to S134c is as follows: for each candidate path in the path population, if the number of road sections of the candidate path is greater than a preset threshold, then the formula X″ m,i [d-1]=(s last ,h last ) replaces the terminal segment in the candidate path with the latest segment, where X″ m,i [d-1] represents the candidate path that replaces the end segment, s last Indicates that the latest road segment can distinguish between clockwise and counterclockwise turning angles, h last Indicates that the latest segment can distinguish between clockwise and counterclockwise heading angles. If the number of segments of the candidate path is less than or equal to the preset threshold, then according to the formula X″ m,i [d]=(s last ,h last) Add the latest segment after the terminal segment of the candidate path, where X″ m,i [d] indicates adding the candidate path of the latest segment after the terminal segment.

[0071] As an optional implementation of the above step S140, the above autonomous mobile machine may be an engineering vehicle, and the above path parameters may include: the starting position point, starting direction, wheelbase, vehicle length, target position point and target parking direction of the engineering vehicle; the above implementation of determining the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environment map may include:

[0072] Step S141: for each candidate path in the path population, obtain the starting position point, starting direction, wheelbase, vehicle length, target position point, and target parking direction of the candidate path.

[0073] An implementation example of the above-mentioned step S141 is as follows: theoretically, for each candidate path in the path population, the trajectory of the circular arc section on the environmental map can be drawn based on the relevant information of each circular arc section of the candidate path. The relevant information of the circular arc section here includes but is not limited to: the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction, etc. of the circular arc section. The following introduces the process of obtaining the relevant information of the circular arc section.

[0074] Specifically, it is assumed that the obtained candidate path related information includes: the center O j =(x 0 ,y 0 ), radius r j , start angle α j , end angle β j , end point P e,j and the end direction V e,j , 1≤j≤d, given the current point P j =(x,y) and current direction V j =(dx,dy), we can traverse X m,i ={(s 1 ,h 1 ),(s 2 ,h 2 ),(s 3 ,h 3 ),...,(s d ,h d )} The turning angle and heading angle of each arc section. First, according to the formula r j =C wheelbase / tan(abs(h j ) / 2) Calculate the radius of each arc segment, where rj represents the radius of the jth arc section of the candidate path, s represents the steering angle of the engineering vehicle on the arc section, C wheelbase Indicates the wheelbase of the engineering vehicle, h j Indicates the heading angle of the engineering vehicle on the circular section. Then, according to the given vehicle head direction V j In this case, the center of the circle is determined according to the following two formulas.

[0075]

[0076] Among them, O j,left Indicates that the center of the jth arc segment of the candidate path is the left center, O j,right The center of the j-th arc segment of the candidate path is the right center.

[0077] Optionally, the turning angle s of the j-th arc segment of the candidate path can be j The positive or negative value selects the left or right center of the circle and calculates the direction V′ j , for example: if the steering angle s j >0, select the center of the circle on the left and change the direction V j Rotate 90 degrees counterclockwise to get V' j ; If the steering angle s j <0, select the center of the circle on the right and change the direction V j Rotate 90 degrees clockwise to get V' j Specifically, the coordinate values ​​of the left and right center of the circle can be expressed as j,left =(x j,left ,y j,left ),O j,righ =(x j,righ ,y j,righ ), and use the formula Determine the center of the circle O j , then, we can use the formula

[0078] To calculate the direction V′ j Among them, O j,left Indicates that the center of the jth arc segment of the candidate path is the left center, O j,right Indicates that the center of the jth arc segment of the candidate path is the right center, s j represents the turning angle of the jth arc segment of the candidate path, x j,left and j,left They represent the horizontal and vertical coordinates of the left center of the jth arc segment of the candidate path, respectively. j,righ ,y j,righ They respectively represent the abscissa and ordinate of the right center of the j-th arc segment of the candidate path.

[0079] Optionally, by the formula To calculate the vehicle head direction V′ j = the angle between (dx′, dy′) and the positive direction of the x-axis (which can also be understood as the initial angle), that is, the angle between the vehicle head direction vector and the unit vector (0, 1) in the positive direction of the x-axis. Then, by rotating the vehicle head direction V′ j , respectively calculate the counterclockwise and clockwise rotation α j The vector V′ after j,1 and V′ j,2 If the vector V′ is obtained by rotating clockwise j,1 The vector V′ obtained by rotating counterclockwise j,2 Distance 0 =(0,1) vector is closer, indicating the initial angle α j It needs to be corrected and set to 360° minus the original angle, and the end angle β is calculated according to the following three formulas j , that is, the initial angle α j Add the driving angle, that is, the heading angle h j .

[0080] V′ j,1 =(dx*cosα j -dy*sinα j ,dx*sinα j +dy*cosα j );

[0081] V′ j,2 =(dx*cos-α j -dy*sin-α j ,dx*sin-α j +dy*cos-α j );

[0082]

[0083] Among them, V′ j,1 Indicates the direction of the vehicle head V′ j Rotate counterclockwise by angle α j The vector after, V′ j,2 Indicates the direction of the vehicle head V′ j Rotate clockwise by angle α j The vector after j It represents the angle between the vehicle head direction and the positive direction of the x-axis (i.e., the initial angle), and Distance(u,v) represents the vector distance function between vector u and vector v.

[0084] Optionally, update α according to the initial angle and final angle after selecting and combining the steering direction j and β j , for example: According to the formula and formula Update the initial angle and the final angle, where α j represents the angle between the vehicle head direction and the positive direction of the x-axis (i.e. the initial angle), β j represents the end angle after the steering direction is selected. Then, by formula P e,j =((dx-x 0 )*cosh j -(dy-y 0 )*sinh j ,(dx-x 0 )*sinh j +(dy-y 0 )*cosh j ) and formula V e,j =(dx*cosh j -dy*sinh j ,dx*sinh j +dy*cosh j ) Calculate the end point P of this section of the trajectory e,j and the trajectory end direction V e,j , and rotate the current coordinate P j Around the center O j =(x 0 ,y 0 )Rotation angle h j The point P obtained later e,j , V e,j The current direction V j Rotate counterclockwise around the origin by angle h j The direction obtained after that. Among them, P e,j Indicates the end point of the trajectory of this section, V e,j Indicates the end direction of the trajectory of this section, x 0 and 0 They represent the horizontal and vertical coordinates of the origin O of the circular arc segment, respectively. j represents the heading angle of the engineering vehicle at the jth road section. Finally, traverse each track and draw the path according to the width W of the vehicle to obtain the trajectory I″ of the entire candidate path of the engineering vehicle in the environment map. Figure 2 The trajectories of the candidate paths are shown in .

[0085] Step S142: Calculate the fitness value of the candidate path according to the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction and environment map of the candidate path.

[0086] An implementation method of the above-mentioned step S142, for example, can use the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction and environmental map of the candidate path to determine the distance cost function between the current position point of the engineering vehicle and the target position point and the collision cost function of the obstacle, as well as the posture cost function between the starting direction of the current point and the target parking direction to determine the fitness value of the candidate path. Specifically, the fitness value can be calculated using the following formula.

[0087] Cost(X m,i )=Distance(P e,j ,P end )+Crash(I′,I″)+Direction(V e,j ,V end )#

[0088] Among them, Cost(X m,i ) represents the fitness value of the ith subsequent path in the mth path population, Distance(P e,j ,P end ) represents the distance cost function between the current position of the engineering vehicle and the target position, Crash(I′,I″) represents the collision cost function between the engineering vehicle and the obstacle (i.e., the cost of the intersection of the engineering vehicle and the obstacle), Direction(V e,j ,V end ) represents the posture cost function between the starting direction of the engineering vehicle at the current point and the target parking direction (that is, the cost of the final posture of the engineering vehicle and the target posture).

[0089] As an optional implementation of the above step S140, determining the optimal path according to the fitness value of the candidate path includes:

[0090] Step S143: If the fitness value of the candidate path is less than the fitness value of the optimal path, the candidate path is determined as the optimal path.

[0091] See also Figure 5 The schematic diagram of the structure of the path planning device provided in the embodiment of the present application is shown; the embodiment of the present application provides a path planning device 200, including:

[0092] The path population generation module 210 is used to generate a path population according to the environment map and path parameters of the autonomous mobile machine, and the path population includes multiple candidate paths.

[0093] The direction parameter calculation module 220 is used to calculate the driving direction parameters of the autonomous mobile machine according to the path parameters of the candidate paths.

[0094] The path population updating module 230 is used to update the path population according to the driving direction parameter to obtain an updated path population.

[0095] The fitness value determination module 240 is used to determine the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environment map, and determine the optimal path according to the fitness values ​​of the candidate paths.

[0096] The optimal path acquisition module 250 is used to iteratively update the path population according to the path parameters of the optimal path until the fitness value of the optimal path is less than a preset threshold, thereby obtaining the optimal path of the autonomous mobile machine in the environment map.

[0097] Optionally, in the embodiment of the present application, the path planning device further includes:

[0098] The optimal path screening module is used to screen out the optimal path from multiple candidate paths.

[0099] The path cross-learning module is used to perform cross-learning on the optimal path and each candidate path in multiple candidate paths.

[0100] Optionally, in the embodiment of the present application, the driving direction parameter includes: a turning angle and a heading angle; and the path population updating module includes:

[0101] The latest road section determination submodule is used to determine the latest road section according to the steering angle and the heading angle.

[0102] The path population update submodule is used to update the path population using the latest road segments.

[0103] Optionally, in the embodiment of the present application, the path population updating submodule includes:

[0104] The road section quantity determination unit is used to determine, for each candidate path in the path population, whether the number of road sections of the candidate path is greater than a preset threshold.

[0105] The segment replacement and adding unit is used to replace the terminal segment in the candidate path with the latest segment if the number of segments in the candidate path is greater than a preset threshold, and to add the latest segment after the terminal segment of the candidate path if the number of segments in the candidate path is less than or equal to the preset threshold.

[0106] Optionally, in an embodiment of the present application, the autonomous mobile machine is an engineering vehicle, and the path parameters include: a starting position point, a starting direction, a wheelbase, a vehicle length, a target position point, and a target parking direction of the engineering vehicle; the fitness value determination module includes:

[0107] The fitness value calculation submodule is used to calculate the fitness value of each candidate path in the path population according to the starting position point, starting direction, wheelbase, vehicle length, target position point, target parking direction and environment map.

[0108] Optionally, in the embodiment of the present application, the fitness value determination module includes:

[0109] The optimal path determination submodule is used to determine the candidate path as the optimal path if the fitness value of the candidate path is less than the fitness value of the optimal path.

[0110] It should be understood that the device corresponds to the above-mentioned path planning method embodiment and can execute the various steps involved in the above-mentioned method embodiment. The specific functions of the device can be found in the description above, and the detailed description is appropriately omitted here. The device includes at least one software function module that can be stored in a memory in the form of software or firmware or solidified in the operating system (OS) of the device.

[0111] See also Figure 6 The electronic device 300 provided in the embodiment of the present application includes: a processor 310 and a memory 320, wherein the memory 320 stores machine-readable instructions executable by the processor 310, and when the machine-readable instructions are executed by the processor 310, the above method is executed.

[0112] The embodiment of the present application also provides a computer-readable storage medium 330, on which a computer program is stored, and the computer program is executed by the processor 310 to execute the above method. The computer-readable storage medium 330 can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0113] The embodiment of the present application also provides a computer program product, including: a computer program or a computer instruction, and the computer program or the computer instruction executes the method described above when executed by a processor.

[0114] It should be noted that each embodiment in this specification is described in a progressive manner, and each embodiment focuses on the differences from other embodiments, and the same or similar parts between the embodiments can be referred to each other. For the device embodiment, since it is basically similar to the method embodiment, the description is relatively simple, and the relevant parts can be referred to the partial description of the method embodiment.

[0115] In several embodiments provided by the embodiments of the present application, it should be understood that the disclosed device and method can also be implemented in other ways. The device embodiments described above are only schematic, for example, the flowcharts and block diagrams in the accompanying drawings show the possible implementation architecture, functions and operations of the devices, methods and computer program products according to the multiple embodiments of the embodiments of the present application. In this regard, each box in the flowchart or block diagram can represent a part of a module, a program segment or a code, and a part of a module, a program segment or a code includes one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also be different from the order of occurrence marked in the accompanying drawings. For example, two consecutive boxes can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, which is mainly determined by the functions involved.

[0116] In addition, each functional module of each embodiment in the embodiment of the present application can be integrated together to form an independent part, or each module can exist separately, or two or more modules can be integrated to form an independent part. In addition, in the description of this specification, the description of reference terms "one embodiment", "some embodiments", "example", "specific example", "some examples", etc. means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the embodiment of the present application. In this specification, the schematic representation of the above terms does not necessarily target the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described can be combined in any one or more embodiments or examples in a suitable manner. In addition, those skilled in the art can combine and combine the different embodiments or examples described in this specification and the features of the different embodiments or examples without contradiction.

[0117] The above description is only an optional implementation manner of the embodiments of the present application, but the protection scope of the embodiments of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or replacements within the technical scope disclosed in the embodiments of the present application, which should be covered within the protection scope of the embodiments of the present application.

Claims

1. A path planning method, It is characterized in that include: generating a path population according to an environment map and path parameters of the autonomous mobile machine, wherein the path population includes a plurality of candidate paths; Calculating a driving direction parameter of the autonomous mobile machine according to the path parameters of the candidate path; Updating the path population according to the driving direction parameter to obtain an updated path population; Determining the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environment map, and determining the optimal path according to the fitness values ​​of the candidate paths; The path population is iteratively updated according to the path parameters of the optimal path until the fitness value of the optimal path is less than a preset threshold, thereby obtaining the optimal path of the autonomous mobile machine in the environment map.

2. The method according to claim 1, It is characterized in that Before updating the path population according to the driving direction parameter, the method further includes: Filter out an optimal path from the multiple candidate paths; Cross-learning is performed on the optimal path and each candidate path in the plurality of candidate paths.

3. The method according to claim 1, It is characterized in that The driving direction parameters include: a steering angle and a heading angle; and updating the path population according to the driving direction parameters includes: Determine the latest road section according to the steering angle and the heading angle; The path population is updated using the latest road segment.

4. The method according to claim 3, It is characterized in that The updating of the path population using the latest road segment includes: For each candidate path in the path population, determining whether the number of road sections of the candidate path is greater than a preset threshold; If so, the latest segment replaces the terminal segment in the candidate path; otherwise, the latest segment is added after the terminal segment of the candidate path.

5. The method according to claim 1, It is characterized in that The autonomous mobile machine is an engineering vehicle, and the path parameters include: a starting position point, a starting direction, a wheelbase, a vehicle length, a target position point, and a target parking direction of the engineering vehicle; determining the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environment map includes: For each candidate path in the path population, a fitness value of the candidate path is calculated according to the starting position point, the starting direction, the wheelbase, the vehicle length, the target position point, the target parking direction and the environment map.

6. The method according to claim 1, It is characterized in that The determining the optimal path according to the fitness value of the candidate path includes: If the fitness value of the candidate path is less than the fitness value of the optimal path, the candidate path is determined as the optimal path.

7. A path planning device, It is characterized in that include: A path population generation module, used to generate a path population according to an environment map and path parameters of the autonomous mobile machine, wherein the path population includes a plurality of candidate paths; a direction parameter calculation module, used to calculate the driving direction parameter of the autonomous mobile machine according to the path parameter of the candidate path; A path population updating module, used for updating the path population according to the driving direction parameter to obtain an updated path population; A fitness value determination module, used to determine the fitness value of each candidate path in the updated path population according to the path parameters of the candidate paths and the environment map, and determine the optimal path according to the fitness values ​​of the candidate paths; The optimal path acquisition module is used to iteratively update the path population according to the path parameters of the optimal path until the fitness value of the optimal path is less than a preset threshold, thereby obtaining the optimal path of the autonomous mobile machine in the environment map.

8. An electronic device, It is characterized in that include: A processor and a memory, wherein the memory stores machine-readable instructions executable by the processor, and the machine-readable instructions are executed by the processor to perform any method according to claims 1 to 6.

9. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the method according to any one of claims 1 to 6 is executed.

10. A computer program product, It is characterized in that include: A computer program or a computer instruction, wherein when the computer program or the computer instruction is executed by a processor, the method according to any one of claims 1 to 6 is executed.