Joint planning and scheduling method and system for task paths of multiple pull type robots

Abstract

The invention provides a multi-pull type robot task path joint planning and scheduling method and system, and the method comprises the steps: determining an initial task distribution scheme according to a preset task execution constraint; determining a current task allocation scheme according to the initial task allocation scheme; according to the current task allocation scheme, path backtracking is carried out on each pull-type robot, and attitude information of each pull-type robot is obtained; according to the posture information of each pull-type robot, conflict detection is conducted on the pull-type robots, and anti-collision information of each pull-type robot is determined; and according to the current initial task allocation scheme and the anti-collision information of each pull-type robot, determining an optimal task allocation scheme and an optimal task execution path of the plurality of pull-type robots. Through the method and the device, joint decision-making and planning of task allocation, execution sequence and collision-free paths of multiple pull type robots are realized, the decision-making time is shortened, the decision-making efficiency is improved, and the execution effect is improved.

Description

Technical Field

[0001] The present application relates to the technical field of multi-robot planning and scheduling, and in particular, to a method and system for joint planning and scheduling of task paths of multiple trailer-type robots. Background Art

[0002] With the advancement of robotics technology, various types of mobile robots have become ubiquitous in all aspects of society. Compared to a single robot, multiple robots can collaborate and coordinate to complete more complex and large-scale tasks. In the past decade, multi-robot systems have gradually emerged in practical applications, such as Amazon's KIVA robots in warehousing and logistics, and wheeled mobile robots in smart factories and automated terminals. At the same time, multi-robot systems have also brought new scientific challenges. Among these multi-robot systems, multi-robot collaborative planning is a core scientific problem, encompassing both "upper-level" task planning and "lower-level" path planning.

[0003] The upper-level task planning assigns a set of tasks to multiple robots and calculates the order of these tasks. Many classic combinatorial optimization and scheduling problems are closely related to task planning. However, existing research often overlooks the requirement for robots to avoid collisions with each other during the actual motion of each task. However, introducing obstacle avoidance constraints into task planning changes the problem structure, increases the search space for task planning, and reduces the computational efficiency of existing algorithms, thus limiting the application of task planning-related technologies in real-world scenarios.

[0004] The underlying multi-robot path planning requires rapidly planning optimal, collision-free paths for the robots, collaboratively planning the movements of multiple robots in an obstacle environment to avoid collisions and guide the robots to their respective task targets. Although multi-robot path planning has been extensively studied in academia over the past decade, related algorithms rarely consider the execution process and the connection between tasks. When considering the execution process and the connection between tasks, many existing multi-robot path planning algorithms lose their theoretical properties, slow down their solution speed, or simply cannot handle these scenarios. Both academia and industry urgently need new approaches to address these challenges.

[0005] In many applications, the upper-level task planning and lower-level path planning are coupled and influence each other. Therefore, separating task planning and path planning and solving them separately is one of the difficulties that limits the optimization performance of multi-robots. Jointly modeling and solving the two will lead to a large-scale constrained optimization problem with a huge state space and difficult to solve, especially when there are many robots and tasks.

[0006] The multi-robot joint path planning problem requires that the robots visit a set of known intermediate target locations, known as task points, before reaching their final destination. In other words, the problem requires planning a collision-free path from start point to target point to final destination. This problem is common in various logistics robotics applications, such as when a trailer truck collects empty carts and returns them to a warehouse. Overcoming the computational bottleneck of this joint task and path planning problem is a key issue.

[0007] In the prior art, patent "CN116907490A, A Multi-Robot Path Planning Method Based on Conflict Search" jointly plans tasks and paths. It uses an ant colony algorithm to calculate the order completion sequence and an A* algorithm to plan the path of each robot, eliminating path conflicts between robots. This solution to the multi-robot combined path planning problem oversimplifies or ignores the process of robots performing tasks at intermediate task points. For example, when a trailer arrives at the task point, workers will hook the empty material cart to the trailer. This process requires a certain amount of work time. At the same time, the length of the trailer body also increases accordingly, causing the trailer to block the path of other robots for a certain period of time. At the same time, the trailer may also collide with the material cart it is towing during subsequent operation.

[0008] The existing application scenarios of trailer logistics robots have the following defects:

[0009] (1) Split solution method: Task planning and path planning are separated and solved, which makes it difficult to find a feasible solution or the solution quality is poor;

[0010] (2) Joint solution method: Joint planning of existing tasks and paths, ignoring the self-collision caused by the increase in the robot's own length, which leads to task planning failure;

[0011] (3) Joint solution method: Joint planning of existing tasks and paths ignores the fact that the robot occupies the workspace for a long time during the task execution, resulting in low planning and solution efficiency and poor solution quality. Summary of the Invention

[0012] In response to the deficiencies in the prior art, the purpose of this application is to provide a method and system for joint planning and scheduling of task paths for multiple trailer-type robots.

[0013] In a first aspect of the present application, a method for joint planning and scheduling of task paths of multiple trailer robots is provided, comprising:

[0014] According to the preset task execution constraints, the initial tasks are assigned to multiple towed robots to determine the initial task assignment plan;

[0015] Determine a current task allocation plan based on the initial task allocation plan;

[0016] Performing path backtracking on each of the towed robots according to the current task allocation plan to obtain posture information of each of the towed robots;

[0017] performing collision detection on the plurality of tractor-mounted robots according to the posture information of each tractor-mounted robot, and determining anti-collision information of each tractor-mounted robot;

[0018] An optimal task allocation scheme and an optimal task execution path for the plurality of tractor robots are determined according to the current task allocation scheme and the anti-collision information of each tractor robot.

[0019] Optionally, allocating the initial task to a plurality of towed robots according to preset task execution constraints and determining the initial task allocation scheme includes:

[0020] Constructing a task graph for the plurality of towed robots according to the preset task execution constraints;

[0021] According to the task graphs of the plurality of towed robots, a preset algorithm is used to allocate and sort the initial tasks to determine the initial task allocation scheme, which includes task allocation and task sorting.

[0022] Optionally, determining a current task allocation plan based on the initial task allocation plan includes:

[0023] If there are more than a preset number of path conflicts between the tractor robots in the initial task allocation plan and the total path completion time of the tractor robots after adding the anti-collision information is greater than a preset time threshold, a new task allocation plan is determined and the new task allocation plan is used as the current task allocation plan;

[0024] If there are no more than a preset number of path conflicts between the towed robots in the initial task allocation plan and the total time for completing the path of the towed robots after adding the anti-collision information is no more than a preset time threshold, the initial task allocation plan will be used as the current task allocation plan.

[0025] Optionally, performing collision detection on the plurality of tractor-mounted robots based on the posture information of each tractor-mounted robot to determine the anti-collision information of each tractor-mounted robot includes:

[0026] determining, based on the posture information of each of the tractor-mounted robots, collision information of the plurality of tractor-mounted robots;

[0027] According to the collision information of the plurality of tractor-type robots, anti-collision information of the tractor-type robots associated with the collision information is determined, wherein the anti-collision information includes anti-collision information during a non-task execution period and anti-collision information during a task execution period.

[0028] Optionally, determining the optimal task allocation scheme and the optimal task execution path for the plurality of tractor robots based on the current task allocation scheme and the anti-collision information of each tractor robot includes:

[0029] In the current task allocation plan, if there is no path conflict between the multiple towed robots, the current task allocation plan is used as the optimal task allocation plan, and the optimal task allocation plan includes optimal task allocation and optimal task sorting;

[0030] According to the optimal task allocation plan, the multiple towed robots are controlled to perform all tasks, the total path completion time of the multiple towed robots is minimized, and the path corresponding to the minimum total path completion time of the multiple towed robots is used as the optimal task execution path of the multiple towed robots.

[0031] Optionally, determining the optimal task allocation scheme and optimal task execution path for the plurality of tractor robots based on the current task allocation scheme and the anti-collision information of each tractor robot further includes:

[0032] If there is a path conflict between the multiple tractor robots in the current task allocation plan, establishing a path search space for the multiple tractor robots according to the current task allocation plan and the anti-collision information of each of the tractor robots;

[0033] In the path search space of the multiple towed robots, the multiple towed robots are controlled to perform all tasks, the total time for completing the collision-free paths of the multiple towed robots is minimized, the path corresponding to the minimum total time for completing the collision-free paths of the multiple towed robots is used as the optimal task execution path of the multiple towed robots, and the current task allocation plan is used as the optimal task allocation plan.

[0034] Optionally, determining the optimal task allocation scheme and optimal task execution path for the plurality of tractor robots based on the current task allocation scheme and the anti-collision information of each tractor robot further includes:

[0035] If there is a path conflict between the multiple towed robots in the current task allocation plan, and there is no collision-free path for the multiple towed robots to perform all tasks, then return to the step of determining the current task allocation plan based on the initial task allocation plan.

[0036] A second aspect of the present application provides a multi-trailer robot task path joint planning and scheduling system, comprising:

[0037] An initial task allocation module is used to allocate initial tasks to multiple towed robots according to preset task execution constraints and determine an initial task allocation plan;

[0038] A task allocation updating module is used to determine a current task allocation plan based on the initial task allocation plan;

[0039] a posture information acquisition module, configured to perform path retracement on each of the towed robots according to the current task allocation plan, and acquire posture information of each of the towed robots;

[0040] an anti-collision information determination module, configured to perform collision detection on the plurality of tractor-mounted robots based on the posture information of each tractor-mounted robot, and determine anti-collision information of each tractor-mounted robot;

[0041] The task path joint planning module is used to determine the optimal task allocation plan and the optimal task execution path of the multiple tractor robots based on the current task allocation plan and the anti-collision information of each tractor robot.

[0042] The third aspect of the present application provides a non-temporary computer-readable storage medium having a computer program stored thereon, characterized in that when the program is executed by a processor, the steps of any one of the methods provided in the first aspect of the present application are implemented.

[0043] A fourth aspect of the present application provides an electronic device, comprising:

[0044] a memory having a computer program stored thereon;

[0045] A processor is used to execute the computer program in the memory to implement the steps of any one of the methods provided in the first aspect of the present application.

[0046] Compared with the prior art, the embodiments of the present application have at least one of the following beneficial effects:

[0047] The method for joint planning and scheduling of task paths for multiple tug robots provided in the present application adopts the technical means of joint planning of tasks and paths by using the current task allocation schemes of multiple tug robots and the anti-collision information of each tug robot, optimizes the paths of the tug robots in executing tasks, can obtain high-quality collision-free task execution paths that pass through multiple task points, improve the solution quality of task and path planning, prevent multiple tug robots from colliding with each other or colliding with themselves, prevent multiple tug robots from causing deadlock, and improve the efficiency of multiple tug robots in executing tasks.

[0048] Other technical effects brought about by the additional features will be further explained in the corresponding embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0049] Other features, objects and advantages of the present application will become more apparent upon reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0050] Figure 1 The present invention is a flowchart of a method for joint planning and scheduling of task paths of multiple trailer robots according to an exemplary embodiment.

[0051] Figure 2 The figure is an overall flow chart of a method for joint planning and scheduling of task paths of multiple trailer robots according to an exemplary embodiment.

[0052] Figure 3 The figure is a task diagram of multiple towed robots according to an exemplary embodiment.

[0053] Figure 4 The figure is a schematic diagram of joint planning of task paths for multiple trailer-type robots according to an exemplary embodiment.

[0054] Figure 5 The figure is a schematic diagram showing conflict detection and resolution of multiple trailer-type robots according to an exemplary embodiment.

[0055] Figure 6 The figure is a schematic diagram of path planning for a towed robot according to an exemplary embodiment.

[0056] Figure 7 The figure is a structural diagram of a multi-trailer robot task path joint planning and scheduling system according to an exemplary embodiment. DETAILED DESCRIPTION

[0057] The present application is described in detail below with reference to specific embodiments. The following embodiments will help those skilled in the art to further understand the present application, but are not intended to limit the present application in any form. It should be noted that those skilled in the art may make several variations and improvements without departing from the scope of the present application. These all fall within the scope of protection of the present application.

[0058] Existing methods for solving multi-robot combined path planning problems oversimplify or ignore the process of robots performing tasks at intermediate task points. In addition, existing technologies separate the task and the path for solution, resulting in poor solution quality or difficulty in finding a feasible solution. Existing methods for jointly solving tasks and paths ignore the time the workspace is occupied during task execution, resulting in low planning and solution efficiency and poor solution quality. They also ignore the possibility of self-collision due to increased robot length, which reduces task execution efficiency. Based on the above problems, the present application provides a method for jointly planning and scheduling task paths for multiple trailer robots, focusing on "multi-robot combined path planning", paying attention to the impact of task execution time and trailer robot size changes on multi-trailer robot path planning and scheduling, in order to solve the above problems.

[0059] Figure 1 The present invention is a flowchart of a method for joint planning and scheduling of task paths of multiple trailer robots according to an exemplary embodiment. Figure 2 The figure is an overall flow chart of a method for joint planning and scheduling of task paths of multiple trailer robots according to an exemplary embodiment.

[0060] Reference Figure 1 、 Figure 2 As shown, in one embodiment of the present application, a method for joint planning and scheduling of task paths of multiple trailer robots includes S11 to S15, which realizes joint planning and scheduling of tasks and paths of multiple trailer robots.

[0061] S11, assigning initial tasks to multiple towed robots according to preset task execution constraints, and determining an initial task assignment plan.

[0062] Specifically, the initial task allocation plan includes task allocation and task sorting.

[0063] S12: Determine the current task allocation plan based on the initial task allocation plan.

[0064] Specifically, the task allocation plan is updated based on the total path completion time of multiple towed robots after the initial task allocation plan is added with conflict detection and anti-collision information, and the total path completion time of multiple towed robots after the other task allocation plans are added with conflict detection and anti-collision information.

[0065] S13, according to the current task allocation plan, performing path backtracking for each towed robot to obtain the posture information of each towed robot.

[0066] Specifically, the posture information of the towed robot includes the length of the towed robot and the position information of the towed robot, and the position information of the towed robot includes the positions of the trailer and the mounted truck.

[0067] S14 , performing collision detection on the multiple tractor-mounted robots based on the posture information of each tractor-mounted robot, and determining the anti-collision information of each tractor-mounted robot.

[0068] Specifically, the collision detection includes collision detection of any parts between multiple tractor-type robots and collision detection between parts of each tractor-type robot itself.

[0069] The anti-collision information indicates constraint information for preventing collisions, and includes anti-collision information during non-mission execution and anti-collision information during mission execution.

[0070] S15 , determining an optimal task allocation plan and an optimal task execution path for the plurality of tractor robots based on the current task allocation plan and the anti-collision information of each tractor robot.

[0071] Specifically, the optimal task allocation solution includes optimal task allocation and optimal task sequencing.

[0072] The method for joint planning and scheduling of task paths of multiple tug robots provided in the present application obtains the posture information of each tug robot by backtracing the path of the current task allocation plan, and performs conflict detection on multiple tug robots based on the posture information to determine the anti-collision information of each tug robot, effectively avoiding collisions between tug robots, reducing collision conflicts in the joint planning of multiple tug robots, and preventing deadlock. It can also effectively prevent the tug robot from colliding with itself after the body length increases when performing tasks; jointly plan tasks and paths based on the current task allocation plan of multiple tug robots and the anti-collision information of each tug robot, optimize the path of the tug robot to perform tasks, and can obtain high-quality collision-free task execution paths that pass through multiple task points, improve the solution quality of tasks and path planning, and improve the efficiency of multiple tug robots in performing tasks.

[0073] In order to determine the initial task allocation plan, in some specific embodiments of the present application, S11, according to the preset task execution constraints, the initial task is allocated to multiple towed robots, and the initial task allocation plan is determined, which can be achieved by using S111 to S112.

[0074] S111, constructing a task graph for multiple towed robots according to preset task execution constraints.

[0075] Specifically, the task execution constraint indicates that a specific task can only be completed by one or more specific robots.

[0076] Figure 3 The figure is a task diagram of multiple towed robots according to an exemplary embodiment.

[0077] Reference Figure 3 As shown, the task graph of multiple towed robots includes the starting point of each device, the corresponding point for executing the task, and the end point.

[0078] S112 , allocating and sorting initial tasks according to the task graphs of the multiple towed robots using a preset algorithm to determine an initial task allocation plan.

[0079] Specifically, the preset algorithm may be a Branch and Cut, an equivalent transformation method, a heuristic algorithm, or the like.

[0080] It should be noted that step S11 needs to be implemented while ignoring collisions.

[0081] The above-mentioned embodiments of the present application, while ignoring collisions, determine the task allocation and task sorting of each towed robot based on task execution constraints, which can optimize resource allocation, assign appropriate tasks to each towed robot, avoid resource waste and excessive load, reduce waiting or idle time, improve task response speed and enhance the stability of task execution.

[0082] To determine the current task allocation plan, in some specific implementations of the present application, S12, based on the initial task allocation plan, determines the current task allocation plan, which may be done by S121 or S122.

[0083] S121, if there are path conflicts between towed robots greater than a preset number in the initial task allocation plan and the total time for completing the path of the towed robots after adding anti-collision information is greater than a preset time threshold, determine a new task allocation plan and use the new task allocation plan as the current task allocation plan.

[0084] Specifically, in the initial task allocation plan, it is detected that too many robots have path conflicts with each other, that is, the path conflicts exceed the preset number, indicating that there are too many path conflicts. At this time, adding too many anti-collision constraints between robots will cause the overall plan efficiency to be too low. In this case, a new task allocation plan is determined and the new task allocation plan is used as the current task allocation plan.

[0085] S122: If there are no more than a preset number of path conflicts between the towed robots in the initial task allocation plan and the total path completion time of the towed robots after adding the anti-collision information is no more than a preset time threshold, the initial task allocation plan will be used as the current task allocation plan.

[0086] Specifically, if in the initial task allocation plan, the number of path conflicts is less than the preset number and the total path completion time of the towed robot after adding the anti-collision information is less than the preset time threshold, it means that the efficiency reduction caused by the added anti-collision constraint is within an acceptable range, and the initial task allocation plan is used as the current task allocation plan.

[0087] The above-mentioned embodiments of the present application update the current task allocation plan to ensure that the current task allocation plan is the task allocation plan with the highest task execution efficiency, improve the accuracy of task allocation, and improve the efficiency of multiple towed robots in performing tasks.

[0088] In the present application, steps S121 to S122 may be used to update the current task allocation plan to obtain a task allocation plan with the highest task execution efficiency.

[0089] To obtain the posture information of each tractor-mounted robot, in some specific embodiments of the present application, in S13, according to the current task allocation plan, a path backtracking is performed on each tractor-mounted robot to obtain the posture information of each tractor-mounted robot. The following methods may be used:

[0090] The entire path of each towed robot in the current task allocation plan is traced back, and the length information of each towed robot after collecting the mounted cargo box and during driving, as well as the position information of the trailer and the mounted cargo box, are obtained as the posture information of the towed robot.

[0091] Specifically, the posture information of the towed robot includes the length of the towed robot and the position information of the trailer and the mounted cargo box.

[0092] In the above-mentioned embodiment of the present application, the posture information of the towed robot is obtained by backtracing the entire path of the towed robot in the current task allocation plan, so as to consider the impact of the change of the towed robot's own posture on the collision factors and path planning, and prevent the self-collision of a single towed robot and the mutual collision of multiple towed robots.

[0093] In order to determine the anti-collision information of each towed robot, in some specific embodiments of the present application, S14, based on the posture information of each towed robot, collision detection is performed on multiple towed robots to determine the anti-collision information of each towed robot, which can be achieved by: S141 to S142.

[0094] S141 , determining collision information of multiple tractor-type robots based on the posture information of each tractor-type robot.

[0095] Specifically, the conflict information of the multiple tractor-type robots includes conflict information of any parts between the multiple tractor-type robots and conflict information between parts of a single tractor-type robot itself.

[0096] Exemplarily, the posture information of each towed robot at each moment is retrieved. When a towed robot has two different body parts occupying the same spatial position at a certain moment, conflict information between its own parts is detected; when two body parts of two towed robots occupy the same spatial position at a certain moment, or the trailers of two towed robots exchange spatial positions at the same moment, conflict information of any parts between multiple towed robots is detected.

[0097] S142: Determine, based on the collision information of the plurality of tractor-type robots, the collision avoidance information of the tractor-type robots associated with the collision information.

[0098] Specifically, the anti-collision information represents constraint information for preventing collisions, including anti-collision information during non-mission execution and anti-collision information during mission execution.

[0099] Exemplarily, according to the collision information of multiple tractor-type robots, corresponding anti-collision information is added to each tractor-type robot involved in the collision information.

[0100] The above-mentioned embodiment of the present application performs conflict detection on the towed robots and adds anti-collision information, which can effectively resolve path conflicts between multiple towed robots during path planning and prevent multiple towed robots from self-collision and mutual collision.

[0101] In order to determine the optimal task allocation plan and the optimal task execution path for multiple towed robots, in some specific embodiments of the present application, S15, based on the current task allocation plan and the anti-collision information of each towed robot, determines the optimal task allocation plan and the optimal task execution path for multiple towed robots, and S151 to S152 can be used.

[0102] S151: In the current task allocation plan, if there is no path conflict between the multiple towed robots, the current task allocation plan is used as the optimal task allocation plan.

[0103] Specifically, the optimal task allocation solution includes optimal task allocation and optimal task sequencing.

[0104] Exemplarily, based on the posture information of each towed robot, conflict detection is performed on multiple towed robots. If there is no conflict between the multiple towed robots, that is, there is no self-collision and multi-machine collision between the multiple towed robots, indicating that there is no path conflict between the multiple towed robots, then the current task allocation plan is the optimal task allocation plan.

[0105] S152, according to the optimal task allocation plan, control multiple towed robots to perform all tasks, minimize the total path completion time of the multiple towed robots, and use the path corresponding to the minimum total path completion time of the multiple towed robots as the optimal task execution path of the multiple towed robots.

[0106] Specifically, the optimal task execution path is selected with the goal of minimizing the total time it takes to complete the paths of multiple towed robots. In the optimal execution path, the total time it takes for multiple towed robots to execute all tasks is the shortest, and there is no collision between the multiple towed robots, and each towed robot has no collision with itself.

[0107] In order to determine the optimal task allocation plan and the optimal execution path of multiple towed robots, in some specific embodiments of the present application, S15, based on the current task allocation plan and the collision information of each towed robot, determines the optimal task allocation plan and the optimal execution path of multiple towed robots, and can also use: S153 to S154.

[0108] S153: If there is a path conflict between the multiple tractor robots in the current task allocation plan, a path search space for the multiple tractor robots is established according to the current task allocation plan and the anti-collision information of each tractor robot.

[0109] Specifically, the path search space of the multiple tractor-mounted robots represents a set of all execution paths of the multiple tractor-mounted robots under the current task allocation scheme while avoiding collisions.

[0110] S154, in the path search space of multiple towed robots, based on the execution of all tasks by multiple towed robots, minimize the total time for completing the collision-free path of the multiple towed robots, use the path corresponding to the minimum total time for completing the collision-free path of the multiple towed robots as the optimal task execution path of the multiple towed robots, and use the current task allocation plan as the optimal task allocation plan.

[0111] Specifically, the optimal task execution path is selected with the goal of minimizing the total time for multiple towed robots to complete the path and avoiding collisions. In the optimal task execution path, the total time for multiple towed robots to perform all tasks is the shortest, there are no collisions between the multiple towed robots, and each towed robot has no collisions within itself.

[0112] If a collision-free optimal execution path for multiple tractor-mounted robots can be determined in the path search space of multiple tractor-mounted robots, it means that the current task allocation plan can be used as the optimal task allocation plan.

[0113] In order to determine the optimal task allocation plan and the optimal execution path of multiple towed robots, in some specific embodiments of the present application, S15, based on the current task allocation plan and the collision information of each towed robot, determines the optimal task allocation plan and the optimal execution path of multiple towed robots, and can also use: S155.

[0114] S155, if there is a path conflict between the multiple towed robots in the current task allocation plan, and there is no collision-free path for multiple towed robots to perform all tasks, then return to the step of determining the current task allocation plan based on the initial task allocation plan.

[0115] Specifically, if a collision-free path for multiple towed robots to perform all tasks is not found in the path search space of multiple towed robots, that is, there is self-collision or multi-machine collision during the path planning process, then return to step S12, redetermine the current task allocation plan, and loop through steps S12 to S15 until a collision-free path for multiple towed robots to perform all tasks is determined, and the optimal task allocation plan and the optimal task execution path are determined.

[0116] The above-mentioned embodiments of the present application allocate and sort the tasks of multiple tug robots by combining the task allocation scheme and the anti-collision information of the tug robots, and efficiently resolve the conflicts between the multiple tug robots, prevent the multiple tug robots from self-collision or multi-machine collision, realize the self-collision-free path planning of a single tug robot and the mutual-collision-free path planning of multiple tug robots, realize the joint planning and scheduling of tug robot tasks and paths, and improve the work efficiency of the tug robots.

[0117] The preferred features of the above embodiments can be used alone in any embodiment, or in any combination without conflict. In addition, parts not described in detail in the embodiments can be implemented using existing technologies.

[0118] The following specific embodiments are used to further illustrate in detail a multi-trailer robot task path joint planning and scheduling method provided in this application, so as to better understand the above technical solution of this application. It should be understood that the following are only some examples and are not used to limit this application.

[0119] Figure 4 The figure is a schematic diagram of joint planning of task paths for multiple trailer-type robots according to an exemplary embodiment.

[0120] Reference Figure 4 As shown in the figure, in the application scenario of assigning multiple tasks to multiple robots, first, the potential mutual collisions between the multiple robots are ignored, and the shortest path between any two task points is used as the task constraint condition to determine the initial task allocation scheme, which includes task allocation scheme 1; secondly, path planning and conflict resolution are performed for the multiple robots according to task allocation scheme 1. At this time, the potential mutual collisions between the multiple robots are considered.

[0121] Specifically, according to the task allocation scheme 1, path planning and conflict resolution are performed on multiple robots, including the operation process of conflict detection and adding anti-collision information.

[0122] If according to task assignment plan 1, conflicts between robot paths are frequent, that is, self-collisions or mutual collisions occur, resulting in an increase in path completion time, a new task assignment plan with higher execution efficiency is detected by solving the K-optimal multiple traveling salesman problem, that is, there is a new task assignment plan whose path completion time after conflict detection and adding anti-collision information is shorter than the path completion time of task assignment plan 1 after conflict detection and adding anti-collision information, the tasks are redistributed and sorted to determine the current task assignment plan, that is, task assignment plan 2, which is the second best task assignment plan after task assignment plan 1.

[0123] According to the task allocation scheme 2, path planning and conflict resolution are performed for multiple robots, taking into account the potential collisions between multiple robots.

[0124] Specifically, according to the task allocation scheme 2, path planning and conflict resolution are performed on multiple robots, including the operation process of conflict detection and adding anti-collision information.

[0125] In the application scenario of assigning multiple tasks to multiple robots, the dispatch center includes multiple task allocation schemes, each of which includes multiple task execution orders. According to the above method, the optimal task allocation scheme is searched and the optimal task execution path without collision is determined, fully considering the path collision requirements between robots to achieve joint planning and scheduling of tasks and paths.

[0126] Figure 5The figure is a schematic diagram showing conflict detection and resolution of multiple trailer-type robots according to an exemplary embodiment.

[0127] In the application scenario of conflict detection and resolution of multiple trailer robots, the trailer robot needs to wait in place for a preset time while performing the task. After the task is completed, the cargo box is mounted, and the length of the trailer robot increases. As a result, collisions are likely to occur between multiple trailer robots.

[0128] Reference Figure 5 As shown in (a), the body length of device J = 4, the body length of device I = 4, and collision detection is performed on device J and device I. Device J and device I have a part collision at time t = 5, and the collision parts include collision part = 0 of device J and collision part = 2 of device I.

[0129] Reference Figure 5 As shown in (b), the body length of device J = 4, the body length of device I = 4, and device I is executing a task, the task start time = 2, and the task end time = 10. A conflict detection is performed on device J and device I, and it is detected that device I and device J have a part conflict at time t = 5, and the collision parts include the collision part of device J = 0 and the collision part of device I = 2.

[0130] In multiple trailer robot conflict detection and resolution application scenarios, the path of each trailer robot is traced back to obtain the posture information of the trailer robot at each moment, that is, the length of the trailer robot and the position information of the trailer and the mounted cargo box. Based on the posture information of the trailer robot, the position conflict between any two trailer robots is detected, and the anti-collision information corresponding to the conflict is obtained as the anti-collision constraint information. Based on the effective anti-collision information, the robot path planning algorithm is used to plan a path where no collision will occur, thereby efficiently resolving the potential path conflicts between the trailer robots.

[0131] Figure 6 The figure is a schematic diagram of path planning for a towed robot according to an exemplary embodiment.

[0132] In the specific application scenario of trailer robot path planning, the trailer robot starts from the starting point, passes through all mission points, and arrives at the end point.

[0133] Reference Figure 6 As shown in the figure, when a towed robot starts from the starting point and moves to the task point and then reaches the destination, the existing method does not manage the posture of the entire process, resulting in the robot being unable to drive out after completing the task in a small space, and will inevitably collide with itself when reaching the destination.

[0134] A multi-trailer robot task path joint planning and scheduling method provided in this application is used. After determining that a self-collision occurs, the route is abandoned and a new path is detected to achieve a situation where no self-collision occurs when reaching the destination.

[0135] Based on A* and further expanding A*, a search is performed in the robot's path and task space, and the robot's complete path, that is, the path passing through multiple task points, is planned. Not only the optimality of the current target point path, but also the feasibility and optimality of subsequent paths are considered, and the optimal path of the robot without self-collision is determined. In combination with anti-collision information, the robot is prevented from colliding with other robots, thus realizing complete path planning for a single robot taking self-collision avoidance into consideration.

[0136] The present application provides a method for joint planning and scheduling of task paths for multiple towed robots. A joint planning method for tasks and paths of multiple towed robots is adopted. When allocating and sorting tasks, the impact of collisions between towed robots is comprehensively considered to determine a high-quality path passing through multiple task points, thereby preventing the situation where the task and path planning are separated and the solution quality is low; conflict detection and conflict resolution are performed between multiple towed robots to prevent mutual collisions and deadlocks between the towed robots; based on the path planning of the towed robots, the optimal path planning of the towed robots without self-collision is achieved, thereby preventing the robots from colliding with themselves due to the increase in their own length.

[0137] The present application provides a method for joint planning and scheduling of task paths for multiple trailer robots, which realizes the joint decision-making and planning of task allocation, execution sequence, and collision-free paths for multiple trailer robots, overcoming the problems in the prior art of being unable to find feasible solutions for multi-trailer robot scenarios, the long decision-making process, and the poor execution effect.

[0138] Figure 7 The figure is a structural diagram of a multi-trailer robot task path joint planning and scheduling system according to an exemplary embodiment.

[0139] Reference Figure 7 As shown, the present application also provides a multi-trailer robot task path joint planning and scheduling system 100, including: an initial task allocation module 110, a task allocation update module 120, a posture information acquisition module 130, an anti-collision information determination module 140, and a task path joint planning module 150.

[0140] An initial task allocation module 110 is configured to allocate initial tasks to a plurality of towed robots according to preset task execution constraints and determine an initial task allocation plan;

[0141] The task allocation updating module 120 is used to determine the current task allocation plan based on the initial task allocation plan;

[0142] The posture information acquisition module 130 is used to perform path backtracking on each tug-type robot according to the current task allocation plan to obtain the posture information of each tug-type robot;

[0143] an anti-collision information determination module 140 for performing collision detection on a plurality of tractor-mounted robots based on the posture information of each tractor-mounted robot and determining anti-collision information of each tractor-mounted robot;

[0144] The task path joint planning module 150 is used to determine the optimal task allocation plan and the optimal task execution path for multiple tractor robots based on the current task allocation plan and the anti-collision information of each tractor robot.

[0145] The method for joint planning and scheduling of task paths of multiple tug robots provided in the present application obtains the posture information of each tug robot by backtracing the path of the current task allocation plan, and performs conflict detection on multiple tug robots based on the posture information to determine the anti-collision information of each tug robot, effectively avoiding collisions between tug robots, reducing collision conflicts in the joint planning of multiple tug robots, and preventing deadlock. It can also effectively prevent the tug robot from colliding with itself after the body length increases when performing tasks; jointly plan tasks and paths based on the current task allocation plan of multiple tug robots and the anti-collision information of each tug robot, optimize the path of the tug robot to perform tasks, and can obtain high-quality collision-free task execution paths that pass through multiple task points, improve the solution quality of tasks and path planning, and improve the efficiency of multiple tug robots in performing tasks.

[0146] Regarding the embodiment of the above system, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0147] Optionally, an electronic device includes a memory and a processor. The memory is used to store programs; the memory may include volatile memory (English: volatile memory), such as random-access memory (English: random-access memory, abbreviated: RAM), such as static random-access memory (English: static random-access memory, abbreviated: SRAM), double data rate synchronous dynamic random access memory (English: Double Data Rate Synchronous Dynamic Random Access Memory, abbreviated: DDR SDRAM), etc.; the memory may also include non-volatile memory (English: non-volatile memory), such as flash memory (English: flash memory). The memory is used to store computer programs (such as applications, functional modules, etc. that implement the above-mentioned methods), computer instructions, etc. The above-mentioned computer programs, computer instructions, etc. can be partitioned and stored in one or more memories. And the above-mentioned computer programs, computer instructions, data, etc. can be called by the processor.

[0148] The aforementioned computer programs, computer instructions, etc. may be partitioned and stored in one or more memories, and the aforementioned computer programs, computer instructions, data, etc. may be called by a processor.

[0149] The processor is configured to execute the computer program stored in the memory to implement the various steps of the method involved in the above embodiment. For details, please refer to the relevant description in the above method embodiment.

[0150] The processor and memory can be independent structures or integrated structures. When the processor and memory are independent structures, the memory and processor can be coupled via a bus.

[0151] Those skilled in the art will appreciate that the embodiments of the present application can be provided as methods, systems, or computer program products. Therefore, the present application can adopt the form of a complete hardware embodiment, a complete software embodiment, or an embodiment in combination with software and hardware. Moreover, the present application can adopt the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) that contain computer-usable program code.

[0152] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or box in the flowchart and / or block diagram, as well as the combination of the processes and / or boxes in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the steps in the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0153] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0154] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0155] Optionally, a non-transitory computer-readable storage medium having a computer program stored thereon.

[0156] The above describes some specific embodiments of the present application. It should be understood that the present application is not limited to the specific embodiments described above, and those skilled in the art may make various variations or modifications within the scope of the claims, which do not affect the substantive content of the present application. The above preferred features may be used in any combination as long as they do not conflict with each other.

Claims

1. A method for joint planning and scheduling of task paths for multiple trailer robots, characterized in that: include: According to the preset task execution constraints, the initial tasks are assigned to multiple towed robots to determine the initial task assignment plan; Determine a current task allocation plan based on the initial task allocation plan; Performing path backtracking on each of the towed robots according to the current task allocation plan to obtain posture information of each of the towed robots; performing collision detection on the plurality of tractor-mounted robots according to the posture information of each tractor-mounted robot, and determining anti-collision information of each tractor-mounted robot; An optimal task allocation scheme and an optimal task execution path for the plurality of tractor robots are determined according to the current task allocation scheme and the anti-collision information of each tractor robot.

2. The method according to claim 1, characterized in that The method of allocating the initial task to the plurality of towed robots according to the preset task execution constraints and determining the initial task allocation scheme includes: Constructing a task graph for the plurality of towed robots according to the preset task execution constraints; According to the task graphs of the plurality of towed robots, a preset algorithm is used to allocate and sort the initial tasks to determine the initial task allocation scheme, which includes task allocation and task sorting.

3. The method according to claim 1, characterized in that Determining a current task allocation plan based on the initial task allocation plan includes: If there are more than a preset number of path conflicts between the tractor robots in the initial task allocation plan and the total path completion time of the tractor robots after adding the anti-collision information is greater than a preset time threshold, a new task allocation plan is determined and the new task allocation plan is used as the current task allocation plan; If there are no more than a preset number of path conflicts between the towed robots in the initial task allocation plan and the total time for completing the path of the towed robots after adding the anti-collision information is no more than a preset time threshold, the initial task allocation plan will be used as the current task allocation plan.

4. The method according to claim 1, wherein The performing collision detection on the plurality of tractor-mounted robots according to the posture information of each tractor-mounted robot and determining the anti-collision information of each tractor-mounted robot includes: determining, based on the posture information of each of the tractor-mounted robots, collision information of the plurality of tractor-mounted robots; According to the collision information of the plurality of tractor-type robots, anti-collision information of the tractor-type robots associated with the collision information is determined, wherein the anti-collision information includes anti-collision information during a non-task execution period and anti-collision information during a task execution period.

5. The method according to claim 4, characterized in that Determining the optimal task allocation scheme and the optimal task execution path for the plurality of tractor robots based on the current task allocation scheme and the anti-collision information of each tractor robot includes: In the current task allocation plan, if there is no path conflict between the multiple towed robots, the current task allocation plan is used as the optimal task allocation plan, and the optimal task allocation plan includes optimal task allocation and optimal task sorting; According to the optimal task allocation plan, the multiple towed robots are controlled to perform all tasks, the total path completion time of the multiple towed robots is minimized, and the path corresponding to the minimum total path completion time of the multiple towed robots is used as the optimal task execution path of the multiple towed robots.

6. The method according to claim 4, characterized in that The determining of the optimal task allocation scheme and the optimal task execution path for the plurality of tractor-mounted robots based on the current task allocation scheme and the anti-collision information of each tractor-mounted robot further includes: If there is a path conflict between the multiple tractor robots in the current task allocation plan, establishing a path search space for the multiple tractor robots according to the current task allocation plan and the anti-collision information of each of the tractor robots; In the path search space of the multiple towed robots, the multiple towed robots are controlled to perform all tasks, the total time for completing the collision-free paths of the multiple towed robots is minimized, the path corresponding to the minimum total time for completing the collision-free paths of the multiple towed robots is used as the optimal task execution path of the multiple towed robots, and the current task allocation plan is used as the optimal task allocation plan.

7. The method according to claim 6, characterized in that The determining of the optimal task allocation scheme and the optimal task execution path for the plurality of tractor-mounted robots based on the current task allocation scheme and the anti-collision information of each tractor-mounted robot further includes: If there is a path conflict between the multiple towed robots in the current task allocation plan, and there is no collision-free path for the multiple towed robots to perform all tasks, then return to the step of determining the current task allocation plan based on the initial task allocation plan.

8. A multi-trailer robot task path joint planning and scheduling system, characterized in that: include: An initial task allocation module is used to allocate initial tasks to multiple towed robots according to preset task execution constraints and determine an initial task allocation plan; A task allocation updating module is used to determine a current task allocation plan based on the initial task allocation plan; a posture information acquisition module, configured to perform path retracement on each of the towed robots according to the current task allocation plan, and acquire posture information of each of the towed robots; an anti-collision information determination module, configured to perform collision detection on the plurality of tractor-mounted robots based on the posture information of each tractor-mounted robot, and determine anti-collision information of each tractor-mounted robot; The task path joint planning module is used to determine the optimal task allocation plan and the optimal task execution path of the multiple tractor robots based on the current task allocation plan and the anti-collision information of each tractor robot.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. An electronic device, characterized in that: include: a memory having a computer program stored thereon; A processor, configured to execute the computer program in the memory to implement the steps of the method according to any one of claims 1 to 7.

Images