Path planning method and device under cooperative work of multiple robots

Abstract

The invention provides a path planning method and device under multi-robot cooperative work, and relates to the technical field of path planning, and the method comprises the steps: constructing a two-dimensional layout map based on the structural layout feature information of a manufacturing workshop, and carrying out the rasterization processing through combining with the structural width of a predetermined robot, and obtaining a two-dimensional layout grid map; a task allocation evaluation mechanism is introduced to analyze the transportation task set to obtain a task list, a first task and a second task are extracted, the first task corresponds to the first robot, and the second task corresponds to the second robot; obtaining a first initial path and a second initial path; judging whether a first predetermined cooperative constraint is met or not, wherein the first predetermined cooperative constraint means that path intersection exists between the first initial path and the second initial path; and if yes, performing adjustment planning. According to the method, the technical problems of task delay and safety risk increase caused by the fact that multi-party path crossing and priority conflict cannot be processed when multi-robot cooperative work is coped with by traditional path planning are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a path planning method and device under the collaborative work of multiple robots. Background Art

[0002] In the fields of modern intelligent manufacturing, logistics and transportation, warehouse management, etc., there is an increasing demand for multiple robots to complete tasks collaboratively. Multi-robot systems can improve the automation of production lines and improve warehouse handling efficiency, while reducing labor costs and ensuring high precision and safety of operations. In a multi-robot collaborative working environment, path overlap and intersection are inevitable, which can easily lead to conflicts or collisions between multiple robots in the same area, especially in workshops and warehouses with limited space. This requires path planning algorithms to be able to detect path conflicts and effectively make dynamic adjustments when conflicts exist. However, traditional path planning cannot handle multi-party path intersections and priority conflicts when dealing with multi-robot collaborative work, resulting in task delays and increased safety risks. Summary of the invention

[0003] This application provides a path planning method and device for multi-robot collaborative work, aiming to solve the technical problem that traditional path planning cannot handle multi-party path intersections and priority conflicts when dealing with multi-robot collaborative work, resulting in task delays and increased safety risks.

[0004] The first aspect disclosed in the present application provides a path planning method for collaborative work of multiple robots, the method comprising: constructing a two-dimensional layout diagram based on structural layout feature information of a manufacturing production workshop, and rasterizing the two-dimensional layout diagram in combination with the structural width of a predetermined robot to obtain a two-dimensional layout grid diagram; introducing a task allocation evaluation mechanism to analyze a transport task set to obtain a task list, and extracting a first task and a second task from the task list, wherein the first task corresponds to a first robot and the second task corresponds to a second robot; analyzing the first task and the second task to obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout grid diagram, respectively; judging whether the first initial path corresponds to the second initial path Whether the path meets the first predetermined coordination constraint, wherein the first predetermined coordination constraint means that the first initial path and the second initial path have a path intersection; if so, calling the first coordination adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot; wherein, it also includes: obtaining the second predetermined coordination constraint, wherein the second predetermined coordination constraint means that the predicted blockage degree of the first initial path or the second initial path reaches a predetermined blocking threshold; judging whether the first initial path and the second initial path meet the second predetermined coordination constraint; if so, calling the second coordination adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

[0005] According to a second aspect disclosed in the present application, a path planning device for collaborative work of multiple robots is provided, and the device is used for the path planning method for collaborative work of multiple robots, and the device includes: a rasterization processing module, the rasterization processing module is used to construct a two-dimensional layout diagram based on the structural layout feature information of the manufacturing production workshop, and rasterize the two-dimensional layout diagram in combination with the structural width of a predetermined robot to obtain a two-dimensional layout raster diagram; a task extraction module, the task extraction module is used to introduce a task allocation evaluation mechanism to analyze the transport task set to obtain a task list, and extract the first task and the second task in the task list, wherein the first task corresponds to the first robot, and the second task corresponds to the second robot; an initial path acquisition module, the initial path acquisition module is used to analyze the first task and the second task, and obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout raster diagram respectively; an initial path acquisition module A path judgment module, the initial path judgment module is used to judge whether the first initial path and the second initial path meet the first predetermined coordination constraint, wherein the first predetermined coordination constraint refers to the existence of path intersection between the first initial path and the second initial path; an adjustment planning module, the adjustment planning module is used to, if it meets, call the first coordination adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot; wherein the adjustment planning module also includes the following operation steps: obtaining a second predetermined coordination constraint, wherein the second predetermined coordination constraint refers to the predicted blockage degree of the first initial path or the second initial path reaching a predetermined blocking threshold; judging whether the first initial path and the second initial path meet the second predetermined coordination constraint; if it meets, calling the second coordination adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

[0006] One or more technical solutions provided in this application have at least the following technical effects or advantages: By rasterizing the two-dimensional layout map and generating a two-dimensional layout grid map, the entire workshop environment can be discretized, and the path planning problem can be simplified to a discrete path search problem. This rasterization method makes it easier for robots to identify traffic areas and obstacle areas, which can significantly improve planning efficiency. A task allocation evaluation mechanism is introduced to reasonably allocate tasks to suitable robots based on task characteristics and robot status to ensure that tasks match robot capabilities, thereby reducing the possibility of conflicts at the task allocation level. The A* algorithm is used to calculate the shortest path from the starting point to the end point of the robot. This heuristic algorithm can not only quickly find the optimal path, but also adapt to complex raster environments and quickly generate initial paths for multiple robots, thereby improving the overall efficiency of path planning. By judging whether the path meets the collaborative constraints, a collaborative adjustment strategy is introduced when necessary to adjust the path in real time, thereby avoiding conflicts between multiple robots in the intersection area. By adjusting the path of the low-priority robot, it ensures that high-priority tasks can be completed smoothly, optimizes the orderliness and reliability of robot collaborative work, and thus improves the overall efficiency of task completion.

[0007] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0008] Figure 1 A schematic flow chart of a path planning method for multi-robot collaborative work provided in an embodiment of the present application.

[0009] Figure 2 A schematic diagram of the structure of a path planning device for multi-robot collaborative work provided in an embodiment of the present application.

[0010] Explanation of the reference numerals: rasterization processing module 10 , task extraction module 20 , initial path acquisition module 30 , initial path determination module 40 , adjustment planning module 50 . DETAILED DESCRIPTION

[0011] The embodiments of the present application provide a path planning method and device for multi-robot collaborative work, thereby solving the technical problem that traditional path planning cannot handle multi-party path intersections and priority conflicts when dealing with multi-robot interactions, resulting in task delays and increased safety risks.

[0012] After introducing the basic principles of the present application, various non-limiting implementation methods of the present application will be specifically introduced below in conjunction with the accompanying drawings of the specification. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0013] Embodiment 1, as Figure 1 As shown, the embodiment of the present application provides a path planning method for multi-robot collaborative work, the method comprising: A two-dimensional layout diagram is constructed based on the structural layout feature information of the manufacturing production workshop, and the two-dimensional layout diagram is rasterized in combination with the structural width of a predetermined robot to obtain a two-dimensional layout grid diagram.

[0014] Measure the physical environment of the manufacturing production workshop and collect structural layout feature information, including walls, equipment locations, passages, workstations, obstacles, and possible transportation routes. Based on the collected layout feature information, determine the boundaries of the entire layout diagram, delineate the range of robot activities, and the fixed obstacles that need to be avoided. Use CAD or other mapping software to digitize these structural layout feature information and generate a two-dimensional plane layout diagram with the physical boundaries of the workshop and obstacle locations.

[0015] According to the actual space size of the workshop and the structural width of the predetermined robot, the resolution of the grid, that is, the size of each grid unit, is determined. The selection of the grid size is set according to the width and turning radius of the robot to ensure that the robot can safely pass through each grid unit. The entire two-dimensional layout diagram is divided according to the set grid size, and the layout diagram is converted into a two-dimensional matrix containing multiple grid units. Each unit represents a specific spatial area. This rasterization process discretizes the entire workshop environment, allowing the path planning algorithm to be executed more efficiently. After the rasterization is completed, a two-dimensional layout grid diagram representing the workshop layout is obtained, in which each grid unit contains basic information about the environment. This grid diagram will become the input basis of the path planning algorithm for subsequent robot path calculation and optimization.

[0016] A task allocation evaluation mechanism is introduced to analyze the transport task set to obtain a task list, and a first task and a second task in the task list are extracted, wherein the first task corresponds to a first robot and the second task corresponds to a second robot.

[0017] The transport task set contains multiple tasks to be completed. The entire transport task set is analyzed to identify all transport tasks that need to be completed. Each task contains information such as the starting point location, target location, weight or volume of the material, urgency, etc. According to the specific requirements of each transport task and the current status of the robot, the suitability of each robot to perform the task is evaluated. The evaluation factors may include the distance between the robot's current geographical location and the starting point of the task, whether the robot's current load and power meet the task requirements, the urgency of the task, and whether the robot has the time and resources to complete the task.

[0018] Based on the evaluation, the most suitable robot is selected for each task. For example, if the first task in the task list is suitable for the first robot to complete, then the task will be assigned to the first robot, forming a "first task-first robot" correspondence relationship. The tasks in the task set are assigned in turn to ensure that all tasks are completed by a suitable robot. After the task assignment is completed, a task list is generated, which contains information about each task and its assigned robot.

[0019] The first task and the second task are extracted from the task list, where the first task and the second task can be any two tasks in the task list, wherein the first task corresponds to the first robot and the second task corresponds to the second robot, that is, the first task is completed by the first robot and the second task is completed by the second robot.

[0020] The first task and the second task are analyzed to obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout grid map, respectively.

[0021] According to the starting point and end point of each task in the task list, the starting point and the target position are marked on the two-dimensional layout grid map, and the A* algorithm is used to plan the path of the grid map. The A* algorithm is a commonly used heuristic path search algorithm that can find the shortest path from the starting point to the end point on the grid map. On the two-dimensional layout grid map, the shortest path from the current position of the first robot to the end point of the first task is calculated for the first robot to obtain the first initial path of the first robot. Similarly, the shortest path from the current position of the second robot to the end point of the second task is calculated for the second robot to obtain the second initial path of the second robot. These paths are used to guide the movement of the robots in the two-dimensional layout grid map and provide a basis for the next step of path adjustment.

[0022] It is determined whether the first initial path and the second initial path meet a first predetermined coordination constraint, wherein the first predetermined coordination constraint refers to a path intersection between the first initial path and the second initial path.

[0023] In a multi-robot collaboration scenario, the first predetermined collaborative constraint means that there is an intersection between the paths of the two robots. If there is an intersection between the paths of the two robots, a collision or blockage may occur. On the two-dimensional layout grid diagram, the first initial path and the second initial path are superimposed to check whether there are the same grid cells, that is, intersections, between the paths. If there are intersections, further analysis is performed to determine whether the two robots will reach the intersection at the same time. The time when the robots reach the intersection can be predicted by the number of path steps or the driving speed. If the paths do not intersect, it means that the two paths do not meet the first predetermined collaborative constraint, and no adjustment is required. The robot can travel along the initial path. If the paths intersect, the two paths meet the first predetermined collaborative constraint, and path adjustments are required to avoid conflicts between the robots at the intersection.

[0024] If it is in compliance, the first collaborative adjustment strategy is called to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

[0025] When the paths meet the collaborative constraints, it means that there is an intersection in the paths. Then, the first collaborative adjustment strategy is used to optimize the robot paths to avoid conflicts. Specifically, the tasks of the two robots are first prioritized to determine which task is more urgent. If the urgency of the first task is higher than that of the second task, the path of the second robot is adjusted first. Conversely, if the urgency of the second task is higher than that of the first task, the path of the first robot is adjusted first. According to the first collaborative adjustment strategy, the path of the low-priority robot is updated to form a new adjusted path. The new path will bypass intersecting roads or set waiting positions at necessary locations to ensure that the robots can pass without conflicts.

[0026] After the adjustment planning, a new path is obtained, that is, the adjusted path of the first robot or the adjusted path of the second robot, which will replace the initial path as the actual driving route of the robot. Under the guidance of the new path, the robot can continue to perform the task without conflict, ensuring the smooth progress of the collaborative workflow.

[0027] Furthermore, a task allocation evaluation mechanism is introduced to analyze the transportation task set to obtain a task list, including: Extract the first transport task from the transport task set; read the predetermined work index, and collect characteristics of the first transport task based on the predetermined work index to obtain first work characteristic information; obtain real-time robot status information, and optimize the first most suitable robot for the first transport task in combination with the first work characteristic information; generate the task list according to the first corresponding relationship between the first transport task and the first most suitable robot.

[0028] The transport task set contains multiple tasks to be completed, such as moving materials, delivering goods, etc. The basic information contained in each task includes the task starting point, target point, material type, weight, task priority or urgency, etc. A task is randomly extracted from the transport task set as the first transport task, and it is taken as the analysis object for further analysis.

[0029] The predetermined work indicators are the basic conditions required to complete the task, such as maximum load, minimum power requirement, time requirement for task completion, specific material handling capabilities, etc. Based on the predetermined work indicators, characteristic information of the first transportation task is extracted as the first work characteristic information to ensure that the robot can meet the basic requirements of the task.

[0030] The real-time status information of all current robots is obtained, including the position of each robot, the current load, the power level, etc. This information helps to determine whether each robot can perform the task. Based on the first work characteristic information, each robot is evaluated to determine whether it meets the task requirements. For example, check whether the current position of the robot is close to the starting point of the task, whether the current load of the robot can meet the weight of the task material, and whether the robot's power is sufficient to complete the entire task. Among the robots that meet the basic requirements of the task, the most suitable robot is selected through optimization calculation. The optimization goal can be the shortest distance, the best resources, etc. After analysis, the first most suitable robot for the first transportation task is obtained, that is, the selected robot best meets the requirements of the task.

[0031] The correspondence between the first transport task and the first most suitable robot is recorded, indicating that the first transport task is performed by the first most suitable robot. The above process is repeated to match a most suitable robot for each task. In this way, all tasks are matched with suitable robots and a task list is gradually generated. The task list is used in the path planning stage to guide the robots to perform their respective tasks.

[0032] Furthermore, obtaining real-time robot status information and optimizing the first most suitable robot for the first transport task in combination with the first work feature information includes: Extract any robot in the real-time robot status information, and the arbitrary robot corresponds to any status information, and the arbitrary status information includes any position, any load and any power; determine whether the arbitrary load and the arbitrary power satisfy the first working condition constraint in the first working feature information, wherein the first working condition constraint includes a first material weight constraint and a first material power constraint; issue an adaptation analysis instruction when and only when the arbitrary load satisfies the first material weight constraint and the arbitrary power satisfies the first material power constraint; analyze the arbitrary position and the first working position constraint in the first working feature information based on the adaptation analysis instruction to obtain any straight-line distance for the arbitrary robot to perform the first transportation task; take the shortest arbitrary straight-line distance as the optimization goal to obtain the first most suitable robot.

[0033] According to the real-time robot status information, any robot is randomly selected from all robots as the analysis object. Any robot corresponds to any status information, which includes any position, any load and any power. Among them, any position is the current coordinates or position information of the robot, which is used to judge its distance from the starting point of the task; any load is the weight currently carried by the robot, indicating its current task burden; any power is the power level of the robot, ensuring that it has enough power to complete the assigned task.

[0034] The first working characteristic information contains specific conditional constraints, that is, the basic conditions for executing the task, which are called the first working condition constraints. The first working condition constraints include the first material weight constraint and the first material power constraint. The first material weight constraint is the weight of the material required to be transported for the task. The robot load must be greater than this weight to ensure that the robot can stably complete the transport task; the first material power constraint is the minimum power required to execute the task. The robot power must be higher than this power to ensure that it will not shut down due to lack of power before the task is completed.

[0035] Check the load and power information of any robot and verify them one by one. If the current load of the robot can meet the first material weight constraint, the load check passes; if the current power of the robot meets the first material power constraint, the power check passes. Only when both the load and power meet the task requirements can the next step of adaptation analysis be continued.

[0036] If and only if any robot passes the dual constraint check of load and power at the same time, an adaptability analysis instruction is issued to instruct the robot to enter the adaptability analysis process. On the contrary, if any condition is not met, the robot is abandoned and the next robot is checked.

[0037] According to the adaptation analysis instruction, the straight-line distance between the current position of each qualified robot and the starting point of the task is analyzed. Specifically, for any qualified robot, its arbitrary position, that is, the current coordinates of the robot, is obtained, the starting position coordinates of the task are extracted from the first working feature information, and the straight-line distance from the current position coordinates of the robot to the starting position coordinates of the task is calculated according to the straight-line distance formula. The calculated straight-line distance is stored as the arbitrary straight-line distance of the arbitrary robot performing the first transport task for use in the optimization selection of subsequent steps.

[0038] Among all the robots that meet the conditions, a straight-line distance comparison is performed, and the robot with the shortest straight-line distance is selected as the first most suitable robot. This robot is assigned as the executor of the task, because a shorter distance means that the robot can reach the starting point of the task faster, reducing the task startup time and improving the task efficiency.

[0039] Furthermore, analyzing the arbitrary position and the first working position constraint in the first working feature information based on the adaptation analysis instruction to obtain an arbitrary straight-line distance for the arbitrary robot to perform the first transport task includes: Obtain a first straight-line distance between the arbitrary position and the first material position in the first work position constraint; obtain a second straight-line distance between the first material position and the first destination position in the first work position constraint; and add the first straight-line distance and the second straight-line distance to obtain the arbitrary straight-line distance.

[0040] Get the coordinates of any position of any current robot, recorded as (x, y), extract the first material position of the task from the task feature information, that is, the location of the material, as the starting position, recorded as (x1, y1), and use the Euclidean distance formula to calculate the first straight-line distance from any position to the first material position. The first straight-line distance represents the straight-line distance from the current position of the robot to the material position, which is used to further calculate the total straight-line distance.

[0041] The first destination position is the end position of the task, denoted as (x2, y2). The second straight-line distance from the first material position to the first destination position is calculated using the Euclidean distance formula. The second straight-line distance represents the shortest path distance from the material position to the end point of the task after the robot transports the material.

[0042] The first straight-line distance is added to the second straight-line distance to obtain the total straight-line distance that the robot completes the task, which represents the overall path length for the robot to start from the current position, reach the material position, and complete the transportation to the destination position.

[0043] Furthermore, the first initial path and the second initial path refer to the shortest driving paths of the first task and the second task respectively obtained by analyzing the two-dimensional layout grid graph based on the A* algorithm principle.

[0044] The A* (A-star) algorithm is a heuristic search algorithm that is often used to find the shortest path from the start point to the end point in a grid or graph. It can effectively calculate the optimal path on a two-dimensional grid map. Mark the start point and end point of the first task and the start point and end point of the second task on the two-dimensional layout grid map, and start searching on the grid map from the start point. During the search process, the A* algorithm calculates each feasible node and skips all inaccessible grid cells when expanding the node, thereby automatically avoiding obstacle areas. When the search reaches the end point of the task, the algorithm stops and generates the shortest path from the start point to the end point. Finally, the shortest path from the start point to the end point of the first task is obtained according to the A* algorithm, which is called the first initial path; the shortest path from the start point to the end point of the second task is obtained in the same way, which is called the second initial path.

[0045] Further, if the conditions are met, calling a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot includes: Step A: Obtain the first urgency of the first task and the second urgency of the second task in sequence; Step B1: According to the first collaborative adjustment strategy, when the first urgency is greater than the second urgency, adjust and plan the second initial path of the second robot; Step B2: According to the first collaborative adjustment strategy, when the first urgency is less than the second urgency, adjust and plan the first initial path of the first robot.

[0046] The urgency of the first task is obtained from the task feature information, which is called the first urgency. Similarly, the urgency of the second task is obtained from the task feature information, which is called the second urgency. The urgency can be quantified based on factors such as the task completion time requirement and the task priority setting. For example, the urgency can be a numerical value, and the larger the numerical value, the higher the priority of the task.

[0047] When the first urgency of the first task is greater than the second urgency of the second task, it means that the first task is more urgent than the second task. In this case, it is necessary to prioritize the smooth completion of the first task. According to the first collaborative adjustment strategy, the second initial path of the second robot is modified to avoid conflict with the path of the first robot.

[0048] The second robot can be made to avoid the path of the first robot by detouring, or the second robot can be made to wait at the intersection and continue driving after the first robot passes. If the detour or waiting strategy is not enough to solve the problem, the path of the second robot can be replanned to find an alternative route that does not affect the first task. After the path adjustment is completed, a new path for the second robot is generated to ensure that the first task is completed first.

[0049] On the contrary, when the first urgency of the first task is less than the second urgency of the second task, it means that the second task is more urgent than the first task. In this case, it is necessary to prioritize the smooth completion of the second task. According to the first collaborative adjustment strategy, the first initial path of the first robot is modified to avoid conflict with the path of the second robot. The specific path adjustment method is the same as the aforementioned second initial path. For the sake of brevity of the specification, it will not be repeated here. After completing the path adjustment, a new path of the first robot is generated to ensure that the second task is completed first.

[0050] Further, if the conditions are met, calling a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot includes: Obtain a target intersection road corresponding to a target intersection point of the first initial path and the second initial path, wherein the target intersection road includes a first intersection road of the first robot and a second intersection road of the second robot; perform adjustment planning of the first initial path by adjusting the first intersection road; and perform adjustment planning of the second initial path by adjusting the second intersection road.

[0051] Find the target intersection point in the overlapping area of ​​the first initial path and the second initial path, that is, the intersection position of the two paths. The target intersection point is a grid unit shared by the two paths, and the robots may collide here. Obtain the target intersection road corresponding to the target intersection point, including the first intersection road and the second intersection road, wherein the path portion of the first robot where the target intersection point is located is called the first intersection road; the path portion of the second robot where the target intersection point is located is called the second intersection road.

[0052] If the urgency of the first task is low, then in the path of the first robot, based on the A* algorithm, the first intersection road is adjusted, such as bypassing the grid of the target intersection, to generate a new conflict-free path so that it does not overlap with the second intersection road. The adjusted first initial path is completed to ensure that the first robot avoids conflicts through the new path.

[0053] On the contrary, if the second task has a lower urgency, the second robot's path is adjusted based on the A* algorithm for the second intersection, such as bypassing the grid of the target intersection, to generate a new conflict-free path so that it does not overlap with the first intersection. The adjusted second initial path is completed to ensure that the second robot avoids conflicts through the new path.

[0054] Furthermore, adjusting the first intersection road to adjust the first initial path includes: Acquire a first road starting point and a first road end point of the first intersection road; perform path analysis of the first road starting point and the first road end point in the two-dimensional layout grid map based on the A* algorithm principle to obtain multiple candidate routes; arrange the multiple candidate routes in ascending order with route distance as a constraint to obtain an ascending order of candidate routes; extract the second candidate route in the ascending order of candidate routes as a first adjustment road for the first intersection road; replace the first intersection road with the first adjustment road to implement adjustment planning of the first initial path.

[0055] The first intersection road is the part of the first robot's path that overlaps with the path of the second robot. The first road starting point and the first road end point of the first intersection road are obtained, wherein the first road starting point is the starting position of the first intersection road, located in the first grid unit where the path of the first robot approaches the intersection, and the first road end point is the end position of the first intersection road, located in the last grid unit of the intersection area, that is, the position where the first robot leaves the intersection road.

[0056] Based on the A* algorithm, multiple candidate routes from the first road starting point to the first road end point are generated in the two-dimensional layout grid map to provide multiple options for path optimization. Specifically, starting from the first road starting point and searching toward the first road end point, the A* algorithm usually searches for an optimal path, but the search parameters or heuristic functions can be adjusted to generate multiple slightly different paths. During the search, an offset option is added to make some paths deviate from the center of the intersecting roads, thereby generating multiple different candidate routes, each of which represents a different path plan from the first road starting point to the end point.

[0057] The total path distance from the starting point to the end point of the road is calculated for each of the candidate routes obtained, and all candidate routes are sorted from short to long according to the total path distance to generate a candidate route list sorted in ascending order of distance. The first route in the list is the shortest path, and the distances of subsequent routes increase in sequence.

[0058] The first shortest path in the list is the first intersection road mentioned above, which intersects with the second robot. Therefore, the second-ranked path is selected as an alternative, which can ensure that the path is shorter and has a higher probability of bypassing the intersection and avoiding conflicts. The second-ranked candidate route in the ascending order of the candidate routes is extracted as the first adjustment route of the first intersection road.

[0059] In the first initial path, the original first intersection road portion is replaced with the first adjustment road, so that the first robot travels on the adjusted path, thereby avoiding intersection with the path of the second robot, realizing the path adjustment planning of the first robot, and ensuring smooth and conflict-free collaborative work between robots.

[0060] Furthermore, it also includes: Obtain a second predetermined collaborative constraint; determine whether the first initial path and the second initial path comply with the second predetermined collaborative constraint; if so, call a second collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

[0061] Predict the congestion of the road at each stage and obtain the second predetermined collaborative constraint. The congestion degree is a quantification of the degree of traffic congestion on a certain road section, which is usually used to represent the travel speed or load condition of the road. A high congestion degree indicates that more robots may pass through the road, resulting in a slower travel speed. For example, based on the aforementioned two-dimensional layout grid map, a robot driving road map can be generated, wherein the driving road map includes multiple road sections. The distribution of robots in each road section can be monitored in real time to directly calculate the expected congestion degree of each road in a certain time period in the future, or the number of robots and the frequency of path use can be simulated and predicted in the path planning system to obtain the congestion degree information of each road section, or the principle of graph neural network can be used to perform intelligent fusion prediction of the congestion degree of the corresponding road section of the robot's driving channel in the driving road map. For example, an arbitrary road section in the first initial path is obtained, and a first-level associated road section that intersects or connects with the arbitrary road section is obtained in combination with the driving road map, and the real-time number of robots in each associated road section in the first-level associated road section is obtained, and the number of robots is used to represent the congestion degree of the corresponding associated road section. Furthermore, the congestion degree data of each associated road section are fused, and the fused result is used as the predicted congestion degree of the arbitrary road section.

[0062] Check the blockage degree of each road on the first initial path and the second initial path to determine whether the path will be affected by the blockage. A blockage threshold can be set. When the blockage degree on the path is higher than the threshold, it means that the path may cause task delays or blockages. In this case, the path is judged to meet the coordination constraints and needs to be adjusted.

[0063] When the initial path meets the second predetermined collaborative constraint, the path is replanned through the second collaborative adjustment strategy to optimize the traffic efficiency. Specifically, if the blockage degree of some sections is high, the relevant paths can be diverted and adjusted to select alternative roads to avoid congestion; the path is adjusted to allow the robot to bypass the sections with high blockage degree. For example, alternative routes with low blockage degree can be given priority, and the traffic efficiency can be improved even if the path is slightly longer; in the case of emergency tasks, the emergency task robot can be given priority to pass through the blocked section, while other robots avoid or delay passing. According to the adjustment strategy, the first initial path or the second initial path with a lower priority is replanned, and a new path is generated to replace the initial path, thereby ensuring that the robot can complete the task efficiently and without obstruction. This process makes multi-robot collaboration more adaptable and robust, and can optimize traffic efficiency even in a high-load environment.

[0064] In summary, the path planning method for multi-robot collaborative work provided by the embodiment of the present application has the following technical effects: By rasterizing the two-dimensional layout map and generating a two-dimensional layout grid map, the entire workshop environment can be discretized, and the path planning problem can be simplified to a discrete path search problem. This rasterization method makes it easier for robots to identify traffic areas and obstacle areas, which can significantly improve planning efficiency. A task allocation evaluation mechanism is introduced to reasonably allocate tasks to suitable robots based on task characteristics and robot status to ensure that tasks match robot capabilities, thereby reducing the possibility of conflicts at the task allocation level. The A* algorithm is used to calculate the shortest path from the starting point to the end point of the robot. This heuristic algorithm can not only quickly find the optimal path, but also adapt to complex raster environments and quickly generate initial paths for multiple robots, thereby improving the overall efficiency of path planning. By judging whether the path meets the collaborative constraints, a collaborative adjustment strategy is introduced when necessary to adjust the path in real time, thereby avoiding conflicts between multiple robots in the intersection area. By adjusting the path of the low-priority robot, it ensures that high-priority tasks can be completed smoothly, optimizes the orderliness and reliability of robot collaborative work, and thus improves the overall efficiency of task completion.

[0065] Embodiment 2 is based on the same inventive concept as the path planning method for multi-robot collaborative work in the above embodiment. Figure 2 As shown, an embodiment of the present application provides a path planning device for multi-robot collaborative work, the device comprising: A rasterization processing module 10, the rasterization processing module 10 is used to construct a two-dimensional layout diagram based on the structural layout feature information of the manufacturing production workshop, and rasterize the two-dimensional layout diagram in combination with the structural width of the predetermined robot to obtain a two-dimensional layout grid diagram; a task extraction module 20, the task extraction module 20 is used to introduce a task allocation evaluation mechanism to analyze the transportation task set to obtain a task list, and extract the first task and the second task in the task list, wherein the first task corresponds to the first robot, and the second task corresponds to the second robot; an initial path acquisition module 30, the initial path acquisition module 30 is used to analyze the first task and the second task, and obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout grid diagram respectively; an initial path judgment module 40, the initial path judgment module 40 is used to judge the first initial path. Whether the path and the second initial path meet the first predetermined collaborative constraint, wherein the first predetermined collaborative constraint means that there is a path intersection between the first initial path and the second initial path; an adjustment planning module 50, wherein the adjustment planning module 50 is used to, if it meets, call the first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot; wherein the adjustment planning module 50 also includes the following operation steps: obtaining the second predetermined collaborative constraint, wherein the second predetermined collaborative constraint means that the predicted blockage degree of the first initial path or the second initial path reaches a predetermined blocking threshold; judging whether the first initial path and the second initial path meet the second predetermined collaborative constraint; if it meets, calling the second collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

[0066] Furthermore, the device further includes a task list generating module to perform the following operation steps: Extract the first transport task from the transport task set; read the predetermined work index, and collect characteristics of the first transport task based on the predetermined work index to obtain first work characteristic information; obtain real-time robot status information, and optimize the first most suitable robot for the first transport task in combination with the first work characteristic information; generate the task list according to the first corresponding relationship between the first transport task and the first most suitable robot.

[0067] Furthermore, the device further comprises a first most suitable robot acquisition module to perform the following operation steps: Extract any robot in the real-time robot status information, and the arbitrary robot corresponds to any status information, and the arbitrary status information includes any position, any load and any power; determine whether the arbitrary load and the arbitrary power satisfy the first working condition constraint in the first working feature information, wherein the first working condition constraint includes a first material weight constraint and a first material power constraint; issue an adaptation analysis instruction when and only when the arbitrary load satisfies the first material weight constraint and the arbitrary power satisfies the first material power constraint; analyze the arbitrary position and the first working position constraint in the first working feature information based on the adaptation analysis instruction to obtain any straight-line distance for the arbitrary robot to perform the first transportation task; take the shortest arbitrary straight-line distance as the optimization goal to obtain the first most suitable robot.

[0068] Furthermore, the device further includes an arbitrary straight-line distance acquisition module to perform the following operation steps: Obtain a first straight-line distance between the arbitrary position and the first material position in the first work position constraint; obtain a second straight-line distance between the first material position and the first destination position in the first work position constraint; and add the first straight-line distance and the second straight-line distance to obtain the arbitrary straight-line distance.

[0069] Furthermore, the first initial path and the second initial path refer to the shortest driving paths of the first task and the second task respectively obtained by analyzing the two-dimensional layout grid graph based on the A* algorithm principle.

[0070] Furthermore, the device further includes a first adjustment planning module to perform the following operation steps: Step A: Obtain the first urgency of the first task and the second urgency of the second task in sequence; Step B1: According to the first collaborative adjustment strategy, when the first urgency is greater than the second urgency, adjust and plan the second initial path of the second robot; Step B2: According to the first collaborative adjustment strategy, when the first urgency is less than the second urgency, adjust and plan the first initial path of the first robot.

[0071] Furthermore, the device further includes a second adjustment planning module to perform the following operation steps: Obtain a target intersection road corresponding to a target intersection point of the first initial path and the second initial path, wherein the target intersection road includes a first intersection road of the first robot and a second intersection road of the second robot; perform adjustment planning of the first initial path by adjusting the first intersection road; and perform adjustment planning of the second initial path by adjusting the second intersection road.

[0072] Furthermore, the device further includes a third adjustment planning module to perform the following operation steps: Acquire a first road starting point and a first road end point of the first intersection road; perform path analysis of the first road starting point and the first road end point in the two-dimensional layout grid map based on the A* algorithm principle to obtain multiple candidate routes; arrange the multiple candidate routes in ascending order with route distance as a constraint to obtain an ascending order of candidate routes; extract the second candidate route in the ascending order of candidate routes as a first adjustment road for the first intersection road; replace the first intersection road with the first adjustment road to implement adjustment planning of the first initial path.

[0073] Through the above-mentioned detailed description of the path planning method under the collaborative work of multiple robots in this specification, those skilled in the art can clearly understand the path planning device under the collaborative work of multiple robots in this embodiment. Since it corresponds to the method disclosed in the embodiment, the description is relatively simple, and the relevant parts can be referred to the description of the method part.

[0074] The above description of the disclosed embodiments enables those skilled in the art to implement or use the present application. Various modifications to these embodiments will be apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present application. Therefore, the present application will not be limited to the embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A path planning method for multi-robot collaborative work, characterized in that: include: Constructing a two-dimensional layout diagram based on the structural layout feature information of the manufacturing production workshop, and performing rasterization processing on the two-dimensional layout diagram in combination with the structural width of the predetermined robot to obtain a two-dimensional layout grid diagram; A task allocation evaluation mechanism is introduced to analyze the transport task set to obtain a task list, and a first task and a second task in the task list are extracted, wherein the first task corresponds to the first robot and the second task corresponds to the second robot; Analyze the first task and the second task, and obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout grid map respectively; Determining whether the first initial path and the second initial path meet a first predetermined coordination constraint, wherein the first predetermined coordination constraint refers to a path intersection between the first initial path and the second initial path; If it is in compliance, calling a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot; Among them, it also includes: Acquire a second predetermined coordination constraint, wherein the second predetermined coordination constraint refers to that a predicted blocking degree of the first initial path or the second initial path reaches a predetermined blocking threshold; determining whether the first initial path and the second initial path meet the second predetermined coordination constraint; If it is in compliance, the second collaborative adjustment strategy is called to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

2. The path planning method for multi-robot collaborative work according to claim 1, characterized in that: The task allocation evaluation mechanism is introduced to analyze the transportation task set to obtain a task list, including: Extracting the first transport task from the transport task set; Reading a predetermined work indicator, and collecting characteristics of the first transportation task based on the predetermined work indicator to obtain first work characteristic information; Acquire real-time robot status information, and optimize and obtain a first most suitable robot for the first transport task in combination with the first work characteristic information; The task list is generated according to a first corresponding relationship between the first transport task and the first most suitable robot.

3. The path planning method for multi-robot collaborative work according to claim 2, characterized in that: Acquiring real-time robot status information, and optimizing the first most suitable robot for the first transport task in combination with the first work feature information, including: Extracting any robot from the real-time robot status information, wherein the any robot corresponds to any status information, and the any status information includes any position, any load, and any power; Determine whether the arbitrary load and the arbitrary power satisfy the first working condition constraint in the first working characteristic information, wherein the first working condition constraint includes a first material weight constraint and a first material power constraint; If and only if the arbitrary load satisfies the first material weight constraint and the arbitrary power quantity satisfies the first material power quantity constraint, issuing an adaptation analysis instruction; Analyze the arbitrary position and the first working position constraint in the first working feature information based on the adaptation analysis instruction to obtain an arbitrary straight-line distance for the arbitrary robot to perform the first transport task; The first most suitable robot is obtained by taking the shortest arbitrary straight line distance as the optimization goal.

4. The path planning method for multi-robot collaborative work according to claim 3, characterized in that: Analyzing the arbitrary position and the first working position constraint in the first working feature information based on the adaptation analysis instruction to obtain an arbitrary straight-line distance for the arbitrary robot to perform the first transport task includes: Obtaining a first straight-line distance between the arbitrary position and the first material position in the first working position constraint; Obtaining a second straight-line distance between the first material position and the first destination position in the first working position constraint; The arbitrary straight-line distance is obtained by adding the first straight-line distance and the second straight-line distance.

5. The path planning method for multi-robot collaborative work according to claim 1, characterized in that: The first initial path and the second initial path refer to the shortest driving paths of the first task and the second task respectively obtained by analyzing the two-dimensional layout grid diagram based on the A* algorithm principle.

6. The path planning method for multi-robot collaborative work according to claim 5, characterized in that: If it is in compliance, calling a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot, including: Step A: sequentially obtaining a first urgency of the first task and a second urgency of the second task; Step B1: According to the first collaborative adjustment strategy, when the first urgency is greater than the second urgency, adjusting and planning the second initial path of the second robot; Step B2: According to the first collaborative adjustment strategy, when the first urgency is less than the second urgency, adjusting and planning the first initial path of the first robot.

7. The path planning method for multi-robot collaborative work according to claim 6, characterized in that: If it is in compliance, calling a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot, including: Acquire a target intersection road corresponding to a target intersection point of the first initial path and the second initial path, wherein the target intersection road includes a first intersection road of the first robot and a second intersection road of the second robot; Perform adjustment planning of the first initial path by adjusting the first intersecting road; Adjustment planning of the second initial path is performed by adjusting the second intersecting road.

8. The path planning method for multi-robot collaborative work according to claim 7, characterized in that: The adjusting plan of the first initial path is performed by adjusting the first intersection road, including: Acquire a first road starting point and a first road end point of the first intersection road; Based on the principle of A* algorithm, a path analysis is performed on the starting point of the first road and the end point of the first road in the two-dimensional layout grid map to obtain a plurality of candidate routes; Arranging the plurality of candidate routes in ascending order based on route distance as a constraint to obtain an ascending order of the candidate routes; extracting the second candidate route ranked in ascending order of the candidate routes as a first adjustment route for the first intersection road; The first intersecting road is replaced by the first adjusted road to implement adjustment planning of the first initial path.

9. A path planning device for multi-robot collaborative work, characterized in that: The device is used to implement the path planning method for multi-robot collaborative work according to any one of claims 1 to 8, comprising: A rasterization processing module, the rasterization processing module is used to construct a two-dimensional layout diagram based on the structural layout feature information of the manufacturing production workshop, and perform rasterization processing on the two-dimensional layout diagram in combination with the structural width of a predetermined robot to obtain a two-dimensional layout grid diagram; A task extraction module, the task extraction module is used to introduce a task allocation evaluation mechanism to analyze the transportation task set to obtain a task list, and extract a first task and a second task from the task list, wherein the first task corresponds to the first robot, and the second task corresponds to the second robot; an initial path acquisition module, the initial path acquisition module being used to analyze the first task and the second task, and respectively obtain a first initial path of the first robot and a second initial path of the second robot in the two-dimensional layout grid map; an initial path judgment module, the initial path judgment module being used to judge whether the first initial path and the second initial path meet a first predetermined coordination constraint, wherein the first predetermined coordination constraint refers to a path intersection between the first initial path and the second initial path; an adjustment planning module, wherein if the conditions are met, the adjustment planning module is used to call a first collaborative adjustment strategy to adjust and plan the first initial path of the first robot or the second initial path of the second robot; The adjustment planning module further includes the following operation steps: Acquire a second predetermined coordination constraint, wherein the second predetermined coordination constraint refers to that a predicted blocking degree of the first initial path or the second initial path reaches a predetermined blocking threshold; determining whether the first initial path and the second initial path meet the second predetermined coordination constraint; If it is in compliance, the second collaborative adjustment strategy is called to adjust and plan the first initial path of the first robot or the second initial path of the second robot.

Images