Multi-robot path planning method and device for complex scene and medium

Abstract

The invention discloses a multi-robot path planning method and device for a complex scene and a medium, and relates to the technical field of robot path planning. The multi-robot path planning method for the complex scene comprises the following steps: S1, constructing an initial path of each robot according to a task starting point and a task ending point of each robot in a robot cluster, and generating a path set; s2, performing path conflict detection on the initial paths in the path set to generate a conflict mark set; s3, establishing a conflict bypassing scheme according to the conflict mark set, and updating the path set; s4, for conflicts which cannot be solved through the conflict bypassing scheme, re-planning the paths of the corresponding robots, and updating the path set; s5, iteratively executing the steps S2 to S4 to generate at least one path set; and S6, issuing the path set with the shortest total length to a robot cluster for execution. By adopting the method, the calculation complexity can be reduced, and the path planning efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of robot path planning, and particularly to a multi-robot path planning method, device and medium for complex scenarios. Background Art

[0002] Multi-Agent Path Finding (MAPF) is an NP-hard combinatorial optimization problem, which involves planning a conflict-free and efficient path for multiple robots to reach their destinations. This problem has wide applications in many fields, such as warehousing logistics, autonomous driving, robot swarms and other scenarios.

[0003] The methods for solving the multi-robot path planning problem are mainly divided into centralized planning algorithms and distributed execution algorithms. The centralized planning algorithm is to plan paths for all robots by a central controller, which needs to know the starting positions, target positions and obstacle positions of all robots. Such algorithms mainly include algorithms based on A* search, conflict search, cost-increasing tree and reduction. The distributed execution algorithm is mainly an algorithm based on reinforcement learning. Each robot only knows the information within its field of vision and interacts with the environment according to the current policy, with the goal of maximizing the cumulative reward to complete the path planning task.

[0004] In practical applications, the factors that need to be considered in multi-robot path planning include the speed, acceleration, turning angle of the robots and the constraints of various interferences. The algorithm usually discretizes the motion control into time steps. Focusing on the solution speed and quality of the multi-robot path planning problem is an important challenge in the field of artificial intelligence. Its main goal is to find a set of conflict-free paths for a group of robots on a given graph. In warehousing operations, the path planning of multiple handling robots also needs to consider dynamic obstacle avoidance and path optimization to improve the operation efficiency of the unmanned warehouse system. However, the existing centralized planning algorithms and distributed execution algorithms all have the problems of high computational complexity and low efficiency, especially when facing a large number of robots and complex environments.

[0005] Therefore, it is urgent to design a multi-robot path planning method, device and medium for complex scenarios to solve the above problems. Summary of the Invention

[0006] The technical problem to be solved by the present invention is to provide a multi-robot path planning method, device and medium for complex scenarios, which can reduce the computational complexity and improve the efficiency of path planning.

[0007] To solve the above technical problems, the present invention provides a multi-robot path planning method, device and medium for complex scenarios, including: S1, respectively constructing initial paths for each robot in the robot cluster according to the task start point and task end point of each robot, and generating a path set; S2, performing path conflict detection on the initial paths in the path set to generate a conflict marker set; S3, establishing a conflict bypass scheme according to the conflict marker set, and updating the path set; S4, for conflicts that cannot be resolved by the conflict bypass scheme, re-planning the path of the corresponding robot, and updating the path set; S5, iteratively executing steps S2 to S4 to generate at least one path set; S6, sending the path set with the shortest total length to the robot cluster for execution.

[0008] As an improvement of the above solution, the step of respectively constructing initial paths for each robot in the robot cluster according to the task start point and task end point of each robot includes: constructing a scene map of the robot working environment; respectively obtaining the task start point and task end point of each robot; according to the constraint conditions, task start point and task end point of the scene map, using the safe time interval algorithm to generate initial paths for each robot respectively.

[0009] As an improvement of the above solution, the step of performing path conflict detection on the initial paths in the path set to generate a conflict marker set includes: performing path conflict detection on the initial paths in the path set to identify conflict points; marking the conflict area and conflict time of the conflict points to generate conflict markers; integrating all the conflict markers into a conflict marker set.

[0010] As an improvement of the above solution, the step of establishing a conflict bypass scheme according to the conflict marker set includes: respectively generating alternative paths to bypass the corresponding conflicts according to the conflict markers in the conflict marker set; if the total length of the alternative path is less than or equal to the length of the corresponding initial path, then using the alternative path as the new initial path, if the total length of the alternative path is greater than the length of the corresponding initial path, then using a path optimization algorithm to optimize the alternative path; if the length of the optimized alternative path is less than the length of the corresponding initial path, then using the optimized alternative path as the new initial path, if the length of the optimized alternative path is greater than or equal to the length of the corresponding initial path, it means that the conflict cannot be resolved by the conflict bypass scheme.

[0011] As an improvement of the above solution, the step of re-planning the path of the corresponding robot for conflicts that cannot be resolved by the conflict bypass scheme includes: adding time constraints and space constraints to the robots with unresolved conflicts; using the safe time interval algorithm to re-plan the paths of the robots with added time constraints and space constraints one by one; using the re-planned paths as the new initial paths.

[0012] As an improvement to the above solution, the step of re-planning the path of the corresponding robot for the conflicts that cannot be resolved by the conflict bypass solution further includes: for the robot whose path re-planning fails, adjusting its working time window to bypass the conflict.

[0013] As an improvement to the above solution, the robot collaborates with lidar to build a scene map of the working environment of the robot cluster.

[0014] Correspondingly, the present invention also discloses a computer device, including a memory and a processor, the memory stores a computer program, and when the processor executes the computer program, the steps of the above multi-robot path planning method for complex scenarios are implemented.

[0015] The present invention also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above multi-robot path planning method for complex scenarios are implemented.

[0016] The beneficial effects of implementing the present invention are as follows:

[0017] The multi-robot path planning method for complex scenarios of the present invention reduces the number of nodes for robot path optimization by establishing a conflict bypass solution, reduces the computational complexity, and improves the efficiency of path planning. By establishing an initial path, conflict detection, establishing a conflict bypass solution, and re-planning the path, it covers the complete process from the initial stage to the final execution of robot path planning, and can systematically solve the complex problems in multi-robot path planning. Description of the Drawings

[0018] Figure 1 It is a flowchart of an embodiment of the multi-robot path planning method for complex scenarios of the present invention;

[0019] Figure 2 It is a flowchart of an embodiment of establishing a conflict bypass solution according to the conflict marker set in the multi-robot path planning method for complex scenarios of the present invention;

[0020] Figure 3 It is a constraint tree diagram of vertex conflicts of three robots in the multi-robot path planning method for complex scenarios of the present invention. Detailed Embodiment

[0021] To make the objectives, technical solutions, and advantages of the present invention clearer, the present invention will be further described in detail below with reference to the accompanying drawings. It is hereby stated that the azimuth terms such as above, below, left, right, front, rear, inner, and outer that appear or will appear in the text of the present invention are only based on the accompanying drawings of the present invention, and they do not specifically limit the present invention.

[0022] See Figure 1 , Figure 1 which shows a flowchart of an embodiment of the multi-robot path planning method for complex scenarios according to the present invention, including:

[0023] S1. Construct initial paths for each robot in the robot cluster according to the task start point and task end point of each robot respectively, and generate a path set;

[0024] Specifically, the step of constructing initial paths for each robot in the robot cluster according to the task start point and task end point of each robot respectively includes:

[0025] (1) Construct a scene map of the robot working environment;

[0026] Preferably, the robot collaborates with a lidar to establish a scene map of the robot cluster working environment.

[0027] (2) Obtain the task start point and task end point of each robot respectively;

[0028] (3) According to the constraint conditions, task start point and task end point of the scene map, use the safety time interval algorithm to generate initial paths for each robot respectively.

[0029] It should be noted that the initialization of the initial paths of the robot cluster needs to ensure the following aspects:

[0030] (1) The lidar of the robot can scan and obtain complete environmental data in the task scene and generate a Scalable Vector Graphics (SVG) file as the map for scheduling execution. All passable paths and nodes, obstacle positions, and map edges need to be clearly marked in the map to ensure the accuracy of the planning;

[0031] (2) Multiple robots {r1, r2,..., r n} are connected to the scheduling platform through a WiFi network. To avoid data loss during upload and download, it is necessary to ensure that the WiFi signal in the task scene can cover without dead spots to ensure the stability and reliability of data transmission;

[0032] (3) The scheduling platform sends the complete map data to all robots in the scene to ensure that each robot completes the task according to the specified path;

[0033] (4) The scheduling platform can automatically create or manually create tasks {t1, t2,..., t m} according to the working set, and generate all steps {s i 1, s i 2,..., s ik}, mainly including the point coordinates of each step and the operations to be performed by the robot, etc., that is, s i j = [(x i j , y i j ), Operation], and assign it to the corresponding robot to ensure the integrity and effectiveness of the task;

[0034] (5) According to the starting coordinates and task end coordinates of all robots performing tasks in the scene, the dispatching platform plans the initial paths of all robots, that is, a set of coordinate sequences.

[0035] Therefore, through step S1, an initial path based on real-world constraints can be established for each robot to ensure that the robot can obtain a conflict-free path before the task starts.

[0036] S2, perform path conflict detection on the initial paths in the path set to generate a set of conflict markers;

[0037] Specifically, the step of performing path conflict detection on the initial paths in the path set to generate a set of conflict markers includes:

[0038] (1) Perform path conflict detection on the initial paths in the path set to identify conflict points;

[0039] (2) Mark the conflict area and conflict time of the conflict points to generate conflict markers;

[0040] (3) Integrate all the conflict markers into a set of conflict markers.

[0041] It should be noted that path conflict detection needs to ensure the following:

[0042] (1) The dispatching platform can find the coordinates of path conflicts in the initial path set and the timestamps when the conflicts occur, and represent them in the form of a triple {a i , a j , t, v} to indicate that robots a i and a j enter the vertex v of position v at the same time at time t;

[0043] (2) The dispatching platform can find all the conflicts in the initial path set so that the subsequent algorithm can solve them to ensure that the robot cluster in the current map does not collide or block each other when performing tasks.

[0044] Therefore, through step S2, the conflict areas and their occurrence times in the initial paths of all robots can be identified, providing markers and guidance for subsequent conflict resolution.

[0045] S3. Establish a conflict bypass scheme based on the conflict marker set and update the path set;

[0046] Such as Figure 2 shown, the steps of establishing a conflict bypass scheme according to the conflict marker set include:

[0047] S301. Generate alternative paths to bypass the corresponding conflicts respectively according to the conflict markers in the conflict marker set;

[0048] S302. If the total length of the alternative path is less than or equal to the length of the corresponding initial path, use the alternative path as the new initial path. If the total length of the alternative path is greater than the length of the corresponding initial path, use a path optimization algorithm to optimize the alternative path;

[0049] S303. If the length of the optimized alternative path is less than the length of the corresponding initial path, use the optimized alternative path as the new initial path. If the length of the optimized alternative path is greater than or equal to the length of the corresponding initial path, it means that the conflict cannot be resolved through the conflict bypass scheme.

[0050] It should be noted that when establishing a conflict bypass scheme according to the conflict marker set, the following aspects need to be ensured:

[0051] (1) Select the conflict to be resolved from the conflict triple {a i , a j , t, v}, try to perform path replanning on the relevant robots, and regenerate the coordinate sequence to represent the path that can bypass the conflict;

[0052] (2) Compare the total path cost after replanning to bypass the conflict and the original total path cost. If the total path cost after bypassing the conflict does not increase, directly use the path that bypasses the conflict to replace the original path to avoid node splitting for resolving the conflict.

[0053] Therefore, through step S3, an alternative path to bypass the marked conflict can be tried to be found, reducing the number of directly processed conflicts and lowering the algorithm complexity and running time.

[0054] S4. For the conflicts that cannot be resolved through the conflict bypass scheme, replan the paths of their corresponding robots and update the path set;

[0055] Specifically, the steps of replanning the paths of the robots corresponding to the conflicts that cannot be resolved through the conflict bypass scheme include:

[0056] (1) Add time constraints and space constraints to the robots with unresolved conflicts;

[0057] (2) Use the safety time interval algorithm to re-plan the paths of the robots with time constraints and space constraints added one by one;

[0058] (3) Take the re-planned path as the new initial path.

[0059] Furthermore, the step of re-planning the path of the corresponding robot for the conflicts that cannot be resolved by the conflict bypass scheme further includes: for the robots whose path re-planning fails, adjust their working time windows to bypass the conflicts.

[0060] Adding time constraints and space constraints and performing path re-planning need to ensure the following aspects:

[0061] (1) Split the child nodes. As Figure 3 shown, select a conflict from the conflict set, and by adding a constraint (a i , v, t) to the child node, prohibit robot a i from entering position v at time t, so that only one robot enters this position at the same time, thus resolving the conflict;

[0062] (2) In the child nodes, according to the constraints added by the parent node, re-plan the paths of the relevant robots, and re-generate a coordinate sequence that satisfies the constraint (a i , v, t) to represent a path that bypasses the conflict and ensure the resolution of the conflict.

[0063] Therefore, through step S4, time constraints and space constraints can be added to the robots with unresolved path planning, and the paths can be re-planned to ensure that each robot completes its task without spatio-temporal conflicts with other robots in the cluster.

[0064] S5. Iteratively execute steps S2 to S4 to generate at least one path set;

[0065] Therefore, through step S5, by iteratively performing high-level conflict detection and low-level path re-planning, the optimal solution can be gradually approached to ensure that all conflicts are finally resolved.

[0066] S6. Send the path set with the shortest total length to the robot cluster for execution.

[0067] In summary, the multi-robot path planning method for complex scenarios of the present invention reduces the number of nodes for robot path optimization by establishing a conflict bypass scheme, reduces the computational complexity, and improves the efficiency of path planning. By establishing an initial path, conflict detection, establishing a conflict bypass scheme, and re-planning the path, it covers the complete process from the initial stage to the final execution of robot path planning, and can systematically solve the complex problems in multi-robot path planning. At the same time, the multi-robot path planning method for complex scenarios of the present invention effectively reduces the search space of the high-level and low-level algorithms by introducing a conflict bypass scheme and a safety time interval algorithm into the multi-robot path planning algorithm based on conflict search, and improves the efficiency of path planning. When the conflict bypass strategy is used for high-level search, the number of node splits is reduced by bypassing conflict nodes, optimizing the high-level search process. When the safety time interval algorithm is used for low-level path planning, the path planning process is accelerated by considering the time interval and optimizing the search space.

[0068] Correspondingly, the present invention also discloses a computer device, including a memory and a processor, where the memory stores a computer program, and when the processor executes the computer program, the steps of the above-mentioned multi-robot path planning method for complex scenarios are implemented. At the same time, the present invention also discloses a computer-readable storage medium, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned multi-robot path planning method for complex scenarios are implemented.

[0069] The above is the preferred embodiment of the present invention. It should be noted that for those of ordinary skill in the art of the present technology, without departing from the principle of the present invention, several improvements and refinements can be made, and these improvements and refinements are also regarded as the protection scope of the present invention.

Claims

1. A multi-robot path planning method for complex scenarios, characterized in that, Including: S1. Respectively construct the initial paths of each robot according to the task start points and task end points of each robot in the robot cluster, and generate a path set; S2. Perform path conflict detection on the initial paths in the path set to generate a conflict marking set; S3. Establish a conflict bypass scheme according to the conflict marking set, and update the path set; S4. For conflicts that cannot be resolved by the conflict bypass scheme, re-plan the path of its corresponding robot, and update the path set; S5. Iteratively execute steps S2 to S4 to generate at least one path set; S6. Send the path set with the shortest total length to the robot cluster for execution.

2. The multi-robot path planning method for complex scenarios according to claim 1, wherein The step of respectively constructing the initial paths of each robot according to the task start points and task end points of each robot in the robot cluster includes: Construct a scene map of the robot working environment; Respectively obtain the task start point and task end point of each robot; According to the constraint conditions, task start points and task end points of the scene map, respectively generate the initial paths of each robot by using the safety time interval algorithm.

3. The multi-robot path planning method for complex scenarios according to claim 1, wherein The step of performing path conflict detection on the initial paths in the path set to generate a conflict marking set includes: Perform path conflict detection on the initial paths in the path set to identify conflict points; Mark the conflict area and conflict time of the conflict points to generate conflict markings; Integrate all the conflict markings into a conflict marking set.

4. The multi-robot path planning method for complex scenarios according to claim 1, wherein The step of establishing a conflict bypass scheme according to the conflict marking set includes: Respectively generate alternative paths to bypass the corresponding conflicts according to the conflict markings in the conflict marking set; If the total length of the alternative path is less than or equal to the length of the corresponding initial path, then use the alternative path as the new initial path; If the total length of the alternative path is greater than the length of the corresponding initial path, then use the path optimization algorithm to optimize the alternative path; If the length of the optimized alternative path is less than the length of the corresponding initial path, then use the optimized alternative path as the new initial path; If the length of the optimized alternative path is greater than or equal to the length of the corresponding initial path, it means that the conflict cannot be resolved by the conflict bypass scheme.

5. The multi-robot path planning method for complex scenarios according to claim 1, wherein The step of re-planning the path of its corresponding robot for conflicts that cannot be resolved by the conflict bypass scheme includes: Add time constraints and space constraints to the robots with unresolved conflicts; Use the safety time interval algorithm to re-plan the paths of the robots with added time constraints and space constraints one by one; Use the re-planned paths as the new initial paths.

6. The multi-robot path planning method for complex scenarios according to claim 5, wherein The step of re-planning the path of its corresponding robot for conflicts that cannot be resolved by the conflict bypass scheme further includes: For robots with failed path re-planning, adjust their working time windows to bypass the conflicts.

7. The multi-robot path planning method for complex scenarios according to claim 2, characterized in that The robots cooperate through lidar to construct a scene map of the robot cluster working environment.

8. A computer device, comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the multi-robot path planning method for complex scenarios according to any one of claims 1 to 7.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the multi-robot path planning method for complex scenarios according to any one of claims 1 to 7.

Images