Multi-robot collaborative warehousing and carrying system

Abstract

The invention relates to the technical field of warehouse logistics, in particular to a multi-robot collaborative warehouse carrying system, which comprises a task state analysis module, a dynamic state mapping table generation module, a task management module and a task management module, wherein the task state analysis module is used for collecting operation data of each robot in real time and generating a dynamic state mapping table according to the data; the task distribution module is used for performing priority division on tasks based on information in the dynamic state mapping table and completing task matching in combination with the real-time state of the robot; the path planning module is used for acquiring a global map of a storage area through environment sensing equipment in a task execution process and planning an optimal advancing path for each robot; and the conflict detection and coordination module adjusts the path or task sequence according to a preset rule so as to avoid conflicts. Accurate task allocation is realized through a dynamic state mapping table and a comprehensive state scoring mechanism, a global map is constructed by utilizing a multi-sensor fusion technology, an optimal path is planned, and meanwhile, the problem of cross task conflicts is solved through virtual check points and conflict detection rules.

Description

TECHNICAL FIELD

[0001] The application belongs to the technical field of warehouse systems, and in particular to a multi-robot collaborative warehouse handling system. BACKGROUND

[0002] With the rapid development of e-commerce, intelligent manufacturing and other industries, the warehouse logistics industry has experienced explosive business growth, placing higher demands on the efficiency, precision and flexibility of cargo handling. Traditional single-robot warehouse handling mode has certain limitations in dealing with large-scale, multi-batch and high-dynamic warehouse operation requirements due to limited operation range and single task processing capability. Multi-robot collaborative warehouse handling systems have gradually become the mainstream development direction of the industry. In a multi-robot collaborative warehouse handling system, the manipulation control part, as the "brain" of the system, directly determines the operation efficiency, collaborative precision and running stability of the robot team, and its technical level is the core factor restricting the performance improvement of the entire system.

[0003] For example, the publication number CN118753699B provides an intelligent warehouse multi-layer robot system based on cloud edge collaboration, which focuses on using an adaptive Monte Carlo positioning method to plan robot paths and completing task allocation through a cloud edge warehouse management server.

[0004] For example, the publication number CN116605574B proposes a collaborative scheduling platform for large-scale robot picking systems, focusing on global path coordination optimization and collision avoidance.

[0005] In the manipulation control of multi-robot collaborative operation, task allocation and resource scheduling are key links, but the existing technology still has the phenomenon of uneven allocation and inefficient scheduling in actual application.

[0006] On the one hand, some systems use average allocation or fixed order allocation task allocation logic, which fails to fully consider the real-time state of each robot, such as remaining power, load capacity, current position and matching degree with task demand, resulting in some robots being in a high-load running state for a long time, while another part of the robots are idle, causing uneven resource utilization.

[0007] On the other hand, when multiple robots simultaneously execute cross tasks, such as converging in the same channel or jointly handling the same batch of goods, the control system lacks an efficient collaborative scheduling mechanism, which can easily cause robot congestion, task conflicts and other problems. These problems not only affect the overall operation efficiency, but also may cause potential safety risks such as equipment collision. Therefore, a multi-robot collaborative warehouse handling system is proposed to address the above problems. SUMMARY

[0008] The present application aims at the deficiencies of the prior art, and provides a multi-robot cooperative warehousing and carrying system to solve the technical problems in the background art.

[0009] To solve the above technical problems, the technical scheme is as follows: a multi-robot cooperative warehousing and carrying system, comprising the following modules:

[0010] A task state analysis module is configured to collect real-time operation data of each robot and generate a dynamic state mapping table according to the data;

[0011] A task allocation module is configured to prioritize tasks based on information in the dynamic state mapping table and complete task matching in combination with real-time states of the robots;

[0012] A path planning module is configured to obtain a global map of a warehousing area through an environment perception device during task execution and plan an optimal travel path for each robot;

[0013] A conflict detection and coordination module is configured to monitor motion states of multiple robots in the same task area in real time and adjust paths or task sequences of the robots through preset rules to avoid conflicts.

[0014] Preferably, the real-time collection of operation data of each robot and the generation of a dynamic state mapping table according to the data comprise the following specific method:

[0015] A sensor group installed on each robot is configured to obtain current position, remaining power, load weight and current task progress information of the robot;

[0016] The above information is sorted according to timestamps and stored in a distributed database to form a dynamic state mapping table;

[0017] For the dynamic state mapping table at any moment, a comprehensive state score of each robot is calculated, wherein the comprehensive state score is obtained by weighted summation of a remaining power proportion, a load capacity utilization rate and a distance to a next task point.

[0018] Preferably, the prioritization of tasks based on information in the dynamic state mapping table and the completion of task matching in combination with real-time states of the robots comprise the following specific method:

[0019] All to-be-allocated tasks are prioritized according to an emergency degree of the tasks, a cargo weight demand and a target position distance;

[0020] For any one to-be-allocated task, a robot set meeting a task demand is screened out from the dynamic state mapping table;

[0021] For the selected set of robots, calculate the time cost and energy cost for each robot to complete the task, and select the robot with the lowest total cost as the task performer.

[0022] Preferably, the global map of the warehouse area is obtained by the environmental perception device during task execution, and an optimal travel path is planned for each robot, including the specific method of:

[0023] A three-dimensional point cloud model of the warehouse area is constructed using a combination of laser radar and ultrasonic sensors, and noise points are removed by a noise reduction algorithm;

[0024] The point cloud model is converted into a two-dimensional grid map, where the state value of each grid cell is determined by the corresponding point cloud density;

[0025] For any robot, the A* algorithm is used to search for the shortest path from the current position to the target position on the two-dimensional grid map, and the path is decomposed into several straight line trajectories.

[0026] Preferably, the motion state of multiple robots in the same task area is monitored in real time, and the path or task order is adjusted to avoid conflicts through a pre-set rule, including the specific method of:

[0027] Multiple virtual checkpoints are set up in the warehouse area, each corresponding to a local monitoring area;

[0028] When two or more robots enter the same local monitoring area, calculate their relative speed and distance, and determine whether there is a potential conflict;

[0029] If there is a potential conflict, determine the adjustment strategy according to the priority of the robot's current task and the remaining path length, where the robot with high priority keeps the original path, and the robot with low priority replans the path or suspends the task.

[0030] Preferably, the specific method of generating the dynamic state mapping table is:

[0031] A data table is created in the distributed database, with each data table containing multiple fields recording the ID, current position coordinates, remaining battery percentage, load weight, current task ID and task progress of the robot;

[0032] For any robot, upload the data collected by its sensor group to the corresponding data table at fixed time intervals, and update the dynamic state mapping table;

[0033] Through the query interface of the distributed database, the latest state information of all robots is read in real time and the dynamic state mapping table is generated.

[0034] Preferably, the specific calculation method of the comprehensive state score is:

[0035] The remaining power ratio is defined as the ratio of the current power to the full power, the load capacity utilization rate is defined as the ratio of the current load weight to the maximum load weight, and the distance to the next task point is defined as the Euclidean distance between the current position of the robot and the next task point.

[0036] The three parameters are multiplied by the preset weight coefficients respectively and then added to obtain the comprehensive state score, wherein the weight coefficients are determined by experimental calibration.

[0037] Preferably, the specific division method of the task priority is:

[0038] The urgency of the task is defined as the difference between the task deadline and the current time, the cargo weight demand is defined as the weight of the cargo required to be carried by the task, and the target position distance is defined as the Manhattan distance between the task starting point and the target point.

[0039] The three parameters are normalized respectively and then multiplied by the preset weight coefficients to obtain the task priority score, wherein the weight coefficients are determined by actual operation demand.

[0040] Preferably, the specific calculation method of the time cost and energy consumption cost is:

[0041] The time cost is the path length from the current position of the robot to the task starting point divided by the average moving speed, plus the estimated time required for task execution.

[0042] The energy consumption cost is the power consumption required by the robot to complete the task, which is determined by the path length, the load weight and the robot energy consumption model.

[0043] Preferably, the specific generation method of the two-dimensional grid map is:

[0044] The three-dimensional point cloud model is projected onto the horizontal plane and divided into a plurality of square grids of the same size.

[0045] For any grid, the number of point clouds contained in the grid is counted, and the state value of the grid is set according to the number of point clouds, wherein the higher the state value, the more likely the area is an obstacle.

[0046] Preferably, the specific execution method of the adjustment strategy is:

[0047] For the robot that needs to re-plan the path, the path planning module is called to re-calculate the shortest path from the current position to the target position of the robot.

[0048] For the robot that needs to suspend the task, the current task of the robot is marked as suspended state, and the task progress information of the robot is updated in the dynamic state mapping table.

[0049] Advantages of the present application:

[0050] By introducing a dynamic state mapping table and a comprehensive state scoring mechanism, the accuracy of task allocation and resource utilization are significantly improved, avoiding the waste of resources caused by traditional average allocation or fixed order allocation. The system uses multi-sensor fusion technology to construct a global map of the warehouse area, and combines A* algorithm to realize efficient path planning, so that the robot can quickly find the optimal travel path in a complex environment. At the same time, by setting virtual checkpoints and conflict detection rules, the congestion and collision problems in multi-robot cross-task are effectively solved, further improving the running stability and safety of the system. BRIEF DESCRIPTION OF DRAWINGS

[0051] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application, and other drawings can be obtained by those skilled in the art without creative labor.

[0052] In the drawings:

[0053] Figure 1 The module structure diagram of the multi-robot cooperative warehouse handling system in the embodiment of the present application.

[0054] Figure 2 The two-dimensional grid map of the warehouse area in the embodiment of the present application.

[0055] The reference signs are as follows:

[0056] 1, task state analysis module; 2, task allocation module; 3, path planning module; 4, conflict detection and coordination module; 5, two-dimensional grid map; 6, robot travel path; 7, obstacle distribution area. DETAILED DESCRIPTION

[0057] The exemplary embodiments of the present disclosure are described below in conjunction with the accompanying drawings, which include various details of the embodiments of the present disclosure to help understanding, and should be considered only as exemplary. Therefore, those skilled in the art should recognize that various changes and modifications can be made to the embodiments described herein without departing from the scope and spirit of the present disclosure. Also, for the sake of clarity and conciseness, the description below omits the description of well-known functions and structures.

[0058] The specific embodiments are given below.

[0059] Please refer to Figures 1-2 The present application provides a multi-robot cooperative warehouse handling system, combined with the accompanying Figure 1 and the accompanyingFigure 2 The composition, connection relationships, and operation process of each module in the system are described in detail. For example... Figure 1 As shown, the system includes a task status analysis module 1, a task allocation module 2, a path planning module 3, and a conflict detection and coordination module 4. These modules are interconnected through a data transmission link to form a complete closed-loop control system, used to achieve efficient collaborative handling of multiple robots in a warehouse environment.

[0060] Based on this, a comprehensive status score is calculated for each robot using the following formula: Comprehensive Status Score = Remaining Battery Percentage × Weighting Coefficient 1 + Load Capacity Utilization Rate × Weighting Coefficient 2 + Distance to Next Task Point × Weighting Coefficient 3. The weighting coefficients are determined through experimental calibration; for example, weighting coefficient 1 is set to 0.4, weighting coefficient 2 to 0.3, and weighting coefficient 3 to 0.3. Using this method, the task status analysis module 1 can accurately reflect the real-time status of each robot, providing a basis for subsequent task allocation.

[0061] Task allocation module 2 completes task priority division and task matching based on the dynamic status mapping table generated by task status analysis module 1.

[0062] First, prioritize all tasks to be assigned based on their urgency, cargo weight requirements, and distance to the target location.

[0063] For example, if task T002 has a deadline of 1 hour, a cargo weight requirement of 50kg, and a target location distance of 10 meters, its priority score is: Priority Score = Normalized Urgency Level × Weight Coefficient 1 + Normalized Cargo Weight Requirement × Weight Coefficient 2 + Normalized Target Location Distance × Weight Coefficient 3. The weight coefficients are determined by the actual operational requirements; for example, weight coefficient 1 can be set to 0.5, weight coefficient 2 to 0.3, and weight coefficient 3 to 0.2. For any task to be assigned, a set of robots that meet the task requirements is selected from the dynamic state mapping table.

[0064] For task T002, robots B and C were selected, both meeting the cargo weight requirements and having sufficient power. Next, the time and energy costs for each robot to complete the task were calculated. The time cost is the path length from the robot's current position to the task's starting point divided by its average speed, plus the estimated time required to complete the task. The energy cost is the electricity consumed by the robot to complete the task, determined by the path length, load weight, and the robot's energy consumption model. Finally, the robot with the lowest total cost was selected as the task executor. For example, robot B has a time cost of 15 minutes and an energy cost of 10%, while robot C has a time cost of 20 minutes and an energy cost of 12%. Therefore, task T002 is assigned to robot B.

[0065] The path planning module 3 is responsible for obtaining the global map of the warehouse area through the environment perception device during task execution, and planning the optimal travel path for each robot.

[0066] As shown in Figure 2 The two-dimensional grid map 5 of the warehouse area is constructed by combining laser radar and ultrasonic sensors, and its generation process is as follows:

[0067] First, the warehouse area is scanned by laser radar to obtain a three-dimensional point cloud model, and noise points are removed by a noise reduction algorithm. Then the point cloud model is projected onto the horizontal plane and divided into a number of square grids of the same size. For example, each grid has a side length of 0.5 meters, and the entire warehouse area is divided into a 100x100 grid matrix.

[0068] For any grid, the number of point clouds contained in it is counted, and the state value of the grid is set according to the number of point clouds, where a higher state value indicates that the area is more likely to be an obstacle distribution area 7. For example, if the number of point clouds contained in a certain grid exceeds the threshold, its state value is set to 1, indicating that the area is an obstacle; otherwise, it is set to 0, indicating that the area is passable.

[0069] After the construction of the two-dimensional grid map is completed, the A* algorithm is used to search for the shortest path from the current position to the target position on the map. For example, robot B starts from the current position (10, 5) and the target position is (30, 20), and the shortest path 6 is calculated by the A* algorithm. The path is decomposed into several straight line trajectories for easy execution by the robot.

[0070] The conflict detection and coordination module 4 adjusts the robot path or task sequence to avoid conflicts by setting virtual checkpoints and preset rules. Multiple virtual checkpoints are set in the warehouse area, each corresponding to a local monitoring area.

[0071] For example, a virtual checkpoint P1 is set at the center of the warehouse area, with a monitoring range of a circular area with a radius of 5 meters. When two or more robots enter the same local monitoring area, their relative speed and distance are calculated, and it is determined whether there is a potential conflict.

[0072] For example, if robots B and D simultaneously enter the P1 monitoring area, their relative speed is calculated to be 1 m / s and their distance is 3 meters, indicating a potential conflict. If a potential conflict exists, an adjustment strategy is determined based on the priority of the robot's current task and the remaining path length. For instance, if robot B's task priority is higher than robot D's and its remaining path length is shorter, robot B maintains its original path, while robot D replans its path or suspends its task. For robots that need to replan their paths, the path planning module 3 is invoked to recalculate the shortest path from their current position to the target position; for robots that need to suspend their tasks, their current task is marked as suspended, and their task progress information is updated in the dynamic state mapping table.

[0073] Through the collaboration of the above modules, this system achieves efficient collaborative handling by multiple robots in complex warehousing environments. For example, in a practical application scenario, five robots in a warehousing area perform different handling tasks. Task status analysis module 1 collects the operating data of each robot in real time and generates a dynamic status mapping table. Task allocation module 2 completes task priority division and task matching based on the dynamic status mapping table. Path planning module 3 plans the optimal travel path for each robot. Conflict detection and coordination module 4 monitors the robot's motion status in real time and adjusts the path or task order to avoid conflicts. Ultimately, all robots successfully complete their tasks, avoiding problems such as uneven task allocation, inefficient scheduling, and conflicts between overlapping tasks.

[0074] To enable those skilled in the art to fully understand and implement this invention, the specific implementation principle of this invention will be further explained below in conjunction with a specific application scenario.

[0075] In a real-world warehousing environment, suppose a large e-commerce logistics center is equipped with a multi-robot collaborative warehousing and handling system. This system includes a task status analysis module 1, a task allocation module 2, a path planning module 3, and a conflict detection and coordination module 4. These modules are interconnected through data transmission links to form a closed-loop control system, used to achieve efficient collaborative handling by multiple robots.

[0076] First, in Task Status Analysis Module 1, each robot is equipped with a sensor array, including a position sensor, a power monitoring device, a load sensor, and a task progress recorder. For example, when robot A is performing task T001 at its current position (10,5), its sensor array collects data showing that the remaining power is 80%, the load weight is 30kg, and the task progress is 50%. This data is uploaded to a distributed database, generating a dynamic status mapping table. Based on this, a comprehensive status score is calculated using the formula: Comprehensive Status Score = Remaining Power Percentage × Weighting Coefficient 1 + Load Capacity Utilization Rate × Weighting Coefficient 2 + Distance to Next Task Point × Weighting Coefficient 3.

[0077] If weight coefficient 1 is 0.4, weight coefficient 2 is 0.3, and weight coefficient 3 is 0.3, then the comprehensive status score of robot A is determined to be a specific value based on the above parameters. Using this method, task status analysis module 1 can reflect the real-time operating status of each robot, providing a basis for subsequent task allocation.

[0078] Next, in task allocation module 2, the urgency level of task T002 is 1 hour, the cargo weight requirement is 50kg, and the target location distance is 10 meters. According to the task priority scoring formula: Priority Score = Normalized Urgency Level × Weight Coefficient 1 + Normalized Cargo Weight Requirement × Weight Coefficient 2 + Normalized Target Location Distance × Weight Coefficient 3, if we set weight coefficient 1 to 0.5, weight coefficient 2 to 0.3, and weight coefficient 3 to 0.2, then the priority score for task T002 can be obtained.

[0079] Subsequently, a set of robots that meet the task requirements, such as Robot B and Robot C, are selected from the dynamic state mapping table. For these two robots, the time cost and energy cost required to complete the task are calculated respectively. The time cost is the path length from the robot's current position to the task's starting point divided by its average moving speed, plus the estimated time required to execute the task; the energy cost is determined by the path length, the load weight, and the robot's energy consumption model. If Robot B's time cost is 15 minutes and its energy cost is 10%, while Robot C's time cost is 20 minutes and its energy cost is 12%, then Robot B is ultimately selected as the task executor.

[0080] In the path planning module 3, the two-dimensional grid map 5 of the warehouse area is constructed using a combination of LiDAR and ultrasonic sensors. For example, LiDAR is used to scan the warehouse area to obtain a three-dimensional point cloud model, and noise points are removed using a noise reduction algorithm. Then, the point cloud model is projected onto a horizontal plane and divided into several square grids with a side length of 0.5 meters. The entire warehouse area is divided into a 100×100 grid matrix. For any given grid, the number of point clouds contained within it is counted, and a state value is set. If the number of point clouds contained in a grid exceeds a threshold, its state value is set to 1, indicating that the area is an obstacle distribution area 7; otherwise, it is set to 0, indicating that the area is passable.

[0081] After constructing the 2D grid map, the A* algorithm is used to search for the shortest path from the current position to the target position on the map. For example, robot B starts from the current position (10,5) and the target position is (30,20). The shortest path 6 is calculated by the A* algorithm. The path is decomposed into several straight line trajectories, which facilitates the robot's execution.

[0082] The A* (A-Star) algorithm is one of the most classic and efficient heuristic search algorithms in the fields of artificial intelligence and path planning. It combines the "global optimality" of Dijkstra's algorithm with the "local heuristic efficiency" of Greedy Best-First Search, and can quickly find the optimal path (such as the shortest distance or minimum cost) from the starting point to the destination in complex graphs or grids.

[0083] Finally, in the conflict detection and coordination module 4, multiple virtual checkpoints are set up within the storage area.

[0084] For example, a virtual checkpoint P1 can be set up at the center of the warehouse area, with a monitoring range of a circular area with a radius of 5 meters. When two or more robots enter the same local monitoring area, their relative speed and distance are calculated, and it is determined whether there is a potential conflict.

[0085] For example, if robots B and D simultaneously enter the P1 monitoring area, their relative speed is calculated to be 1 m / s and their distance is 3 meters, indicating a potential conflict. If a potential conflict exists, an adjustment strategy is determined based on the priority of the robot's current task and the remaining path length. For instance, if robot B's task priority is higher than robot D's and its remaining path length is shorter, robot B maintains its original path, while robot D replans its path or suspends its task. For robots that need to replan their paths, the path planning module 3 is invoked to recalculate the shortest path from their current position to the target position; for robots that need to suspend their tasks, their current task is marked as suspended, and their task progress information is updated in the dynamic state mapping table.

[0086] Through the collaboration of the above modules, this system enables efficient collaborative handling of multiple robots in complex warehousing environments.

[0087] For example, in a real-world application scenario, five robots in a warehouse area perform different handling tasks. Task status analysis module 1 collects real-time operational data for each robot and generates a dynamic status mapping table. Task allocation module 2 prioritizes and matches tasks based on this dynamic status mapping table. Path planning module 3 plans the optimal path for each robot. Conflict detection and coordination module 4 monitors the robot's movement status in real-time and adjusts paths or task sequences to avoid conflicts. Ultimately, all robots successfully complete their tasks, avoiding problems such as uneven task allocation, inefficient scheduling, and conflicts between overlapping tasks.

[0088] The contents not described in detail in this specification are existing technologies known to those skilled in the art, and the model parameters of each electrical appliance are not specifically limited; conventional equipment can be used. Electrical control components not mentioned in this technical solution are existing technologies and are therefore not shown in the figures, nor will they be described further here.

[0089] In the description of this invention, it should be understood that the various forms of processes shown above can be used, with steps reordered, added, or deleted. For example, the steps described in this disclosure can be executed in parallel, sequentially, or in different orders, as long as the desired results of the technical solution provided in this disclosure can be achieved, and no limitation is imposed herein.

[0090] The above description is merely a preferred embodiment of the present invention and does not constitute a limitation on the scope of protection of this disclosure. Those skilled in the art should understand that various modifications, combinations, sub-combinations, and substitutions can be made according to design requirements and other factors. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of this disclosure should be included within the scope of protection of this disclosure.

Claims

1. A multi-robot collaborative warehouse handling system, comprising the following modules: a task state analysis module for collecting real-time operation data of each robot and generating a dynamic state mapping table according to the data; a task allocation module for prioritizing tasks based on information in the dynamic state mapping table and completing task matching in combination with real-time states of the robots; a path planning module for obtaining a global map of the warehouse area through an environmental perception device during task execution and planning an optimal travel path for each robot; a conflict detection and coordination module for monitoring the motion state of multiple robots in the same task area in real time and adjusting their paths or task sequences to avoid conflicts through preset rules; the specific method of collecting real-time operation data of each robot and generating a dynamic state mapping table according to the data comprises: obtaining the current position, remaining power, load weight and current task progress information of each robot through a sensor group installed on each robot; sorting the above information according to the time stamp and storing it in a distributed database to form a dynamic state mapping table; calculating the comprehensive state score of each robot for any moment of the dynamic state mapping table, wherein the comprehensive state score is obtained by weighted summation of the remaining power ratio, the load capacity utilization rate and the distance to the next task point; the specific calculation method of the comprehensive state score is: defining the remaining power ratio as the ratio of the current power to the full power, the load capacity utilization rate as the ratio of the current load weight to the maximum load weight, and the distance to the next task point as the Euclidean distance between the current position of the robot and the next task point; adding the above three parameters multiplied by the preset weight coefficients to obtain the comprehensive state score, wherein the weight coefficients are determined by experiment calibration.

2. The multi-robot collaborative warehousing handling system according to claim 1, characterized in that: the specific method of prioritizing tasks based on information in the dynamic state mapping table and completing task matching in combination with real-time states of the robots comprises: prioritizing all tasks to be allocated according to the urgency of the task, the weight demand of the goods and the distance to the target position; for any one task to be allocated, selecting a set of robots that meet the task demand from the dynamic state mapping table; calculating the time cost and energy cost required for each robot to complete the task for the selected robot set and selecting the robot with the lowest total cost as the task performer.

3. The multi-robot collaborative warehousing handling system of claim 2, wherein: the specific method of prioritizing tasks is: defining the urgency of the task as the difference between the task deadline and the current time, the weight demand of the goods as the weight of the goods to be handled by the task, and the distance to the target position as the Manhattan distance between the task starting point and the target point; normalizing the above three parameters and multiplying them by the preset weight coefficients to obtain the task priority score, wherein the weight coefficients are determined by actual job requirements.

4. The multi-robot collaborative warehousing handling system of claim 2, wherein: the specific calculation method of the time cost and energy cost is: the time cost is the path length from the current position of the robot to the task starting point divided by the average moving speed of the robot plus the estimated time required for task execution; the energy cost is the power consumption required for the robot to complete the task, which is determined by the path length, the load weight and the energy consumption model of the robot.

5. The multi-robot collaborative warehousing handling system of claim 1, wherein: The specific method for obtaining a global map of the storage area by an environmental perception device during task execution and planning an optimal travel path for each robot comprises: A three-dimensional point cloud model of the storage area is constructed using a combination of laser radar and ultrasonic sensors, and noise points are removed by a noise reduction algorithm; The point cloud model is converted into a two-dimensional grid map, wherein the state value of each grid cell is determined by the corresponding point cloud density; For any robot, the A* algorithm is used to search for the shortest path from the current position to the target position on the two-dimensional grid map, and the path is decomposed into several straight-line trajectories.

6. A multi-robot cooperative warehousing handling system according to claim 5, wherein: The specific method for generating the two-dimensional grid map comprises: The three-dimensional point cloud model is projected onto the horizontal plane and divided into several square grid cells of the same size; For any grid, the number of point clouds contained in the grid is counted, and the state value of the grid is set according to the number of point clouds, wherein a higher state value indicates that the area is more likely to be an obstacle distribution area.

7. The multi-robot collaborative warehousing handling system of claim 1, wherein: The specific method for real-time monitoring of the motion state of multiple robots in the same task area and adjusting their paths or task sequences to avoid conflicts comprises: Multiple virtual checkpoints are set in the storage area, each corresponding to a local monitoring area; When two or more robots enter the same local monitoring area, the relative speed and distance between them are calculated to determine whether there is a potential conflict; If there is a potential conflict, the adjustment strategy is determined according to the priority of the robot's current task and the remaining path length, wherein the robot with high priority maintains the original path, and the robot with low priority re-plans the path or suspends the task.

8. A multi-robot collaborative warehousing handling system according to claim 7, wherein: The specific execution method of the adjustment strategy comprises: For robots that need to re-plan the path, the path planning module is called to recalculate the shortest path from the current position to the target position; For robots that need to suspend the task, the current task is marked as suspended, and the task progress information is updated in the dynamic state mapping table.

9. The multi-robot cooperative warehousing handling system of claim 1, wherein: The specific generation method of the dynamic state mapping table comprises: A data table is created in the distributed database, with each data table containing multiple fields recording the ID, current position coordinates, remaining battery percentage, load weight, current task ID, and task progress of the robot; For any robot, the data collected by the sensor group is uploaded to the corresponding data table at fixed time intervals, and the dynamic state mapping table is updated; The latest state information of all robots is read in real time through the query interface of the distributed database, and the dynamic state mapping table is generated.

10. The multi-robot collaborative warehousing handling system of claim 1, wherein: The conflict detection and coordination module is implemented through virtual checkpoints and conflict detection rules, wherein the number and location of virtual checkpoints are pre-set according to the actual layout of the storage area, and the conflict detection rules are based on the comparison results of robot priority and path length.

Images