Task planning method and device for robot cluster and storage medium

Abstract

The invention relates to the technical field of path planning, in particular to a task planning method and device for a robot cluster and a storage medium, and the method comprises the steps: obtaining the remaining time length and the task weight of each target, and determining a priority index according to the remaining time length and the task weight; obtaining a grid map, determining selectable paths from a starting grid of each robot to a target grid of each target in the robot cluster, and obtaining a selectable path set of each target; obtaining the distance and the congestion degree of the selectable path, determining the cost of the selectable path according to the distance and the congestion degree, and determining a preferred path and a robot corresponding to each target according to the cost of the selectable path; controlling the corresponding robot to run towards the target according to the optimal path, and adjusting the running path of the robot in real time according to the priority index of the target until the robot reaches the target grid of the corresponding target; according to the invention, cluster cooperation efficiency can be optimized, and task execution efficiency can be improved.

Description

Technical Field

[0001] The present invention relates to the technical field of path planning, and in particular to a task planning method, device and storage medium for a robot cluster. Background Art

[0002] Robotic task planning often relies on static presets and lacks dynamic adjustment capabilities. However, complex application scenarios require the use of robot swarms for multi-tasking. For example, in warehousing and logistics, robot swarms must simultaneously handle multiple tasks, such as picking, moving, and sorting. These tasks often have varying priorities and urgency, making it crucial to rationally arrange the robots' movement paths. This ensures efficient and orderly planning of each task.

[0003] In related technologies, when multi-tasking is performed in the form of robot clusters, it is difficult to adapt to complex and changing environmental requirements, resulting in low efficiency and waste of resources. Summary of the Invention

[0004] To solve the above problems, the present invention provides a task planning method, device and storage medium for a robot cluster, aiming to dynamically adjust the task allocation strategy through intelligent algorithms, optimize the cluster collaboration efficiency, ensure the efficient completion of task goals, and improve resource utilization.

[0005] In order to achieve the above object, the present invention provides the following technical solutions: In one aspect, an embodiment of the present invention provides a task planning method for a robot swarm, the method comprising the following steps: S100, obtaining the remaining time and task weight of each goal, and determining a priority index according to the remaining time and task weight; S200, obtaining a grid map, determining an optional path between a starting grid of each robot in the robot cluster and a target grid of each target, and obtaining a set of optional paths for each target; S300, obtaining the distance and congestion of an optional path, determining the cost of the optional path based on the distance and congestion, and determining the preferred path and robot corresponding to each target based on the cost of the optional path; S400: Control the corresponding robot to travel toward the target according to the preferred path, and adjust the robot's travel path in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target.

[0006] Optionally, in S100, determining the priority index according to the remaining time and the task weight includes: S110, obtaining the remaining time of the goal and the task weight, dividing the task weight by the sum of the remaining time of the goal and the preset reference time to obtain an initial index; S120, normalize the initial index to obtain the priority index.

[0007] Optionally, in S200, the step of obtaining the grid map, determining the optional paths between the starting grids of each robot in the robot cluster and the target grids of each target, and obtaining the set of optional paths for each target includes: S210, obtain the grid map, use the grids of each robot in the robot cluster in the grid map as the starting grids, and use the grids of each target in the grid map as the target grids; S220, for each target, traverse each robot, respectively determine the optional paths between the target and the traversed robot, obtain all the optional paths between the starting grid of the robot and the target grid of the target, and obtain a path group from the robot to the target; S230, use the path groups from each robot to the target as the set of optional paths for the target.

[0008] Optionally, in S300, the steps of obtaining the distance and congestion degree of the optional path, determining the cost of the optional path according to the distance and congestion degree, and determining the preferred paths and robots corresponding to each target according to the cost of the optional path include: S310, obtain the initialized pheromone matrix, determine the concentration information field of each grid on the optional path, and generate the pheromone content of the optional path based on the concentration information field of each grid; wherein, the pheromone matrix contains the concentration information fields of each grid, and the concentration information field represents the sum of the initial concentration value and the dynamic concentration value, and the dynamic concentration value is positively correlated with the number of times the grid is occupied; S320, multiply the ratio of the current number of robots on the optional path to the section capacity by the historical congestion coefficient and then normalize it to obtain the congestion degree of the optional path; S330, normalize the ratio of the distance of the optional path to the priority index to obtain the heuristic factor, and perform weighted normalization on the heuristic factor, congestion degree and pheromone content to obtain the cost of the optional path; S340, traverse each target in descending order of the priority index, select a path group and the corresponding robot from the set of optional paths of the target, and sequentially match to obtain the path groups and robots corresponding to each target, and use the cost of each optional path in the path group as the multiple costs corresponding to the target; S350, select one cost from the multiple costs corresponding to each target, combine them and then normalize them to obtain the total cost index of all targets; S360, select the combination with the smallest total cost index, determine the preferred paths corresponding to the costs of each target in the combination, and the robots corresponding to the preferred paths.

[0009] Optionally, in S310, the steps of obtaining the initialized pheromone matrix, determining the concentration information field of each grid on the optional path, and generating the pheromone content of the optional path based on the concentration information field of each grid include: S311, obtaining the number of times each grid is occupied at a set time interval to obtain the average number of times each grid is occupied; S312, determining the concentration change rate of the grid according to the ratio of the number of times occupied to the average number of times, and multiplying the initial concentration value of the grid by the concentration change rate to obtain the current concentration information field of the grid; S313, performing weighted normalization processing on the concentration information fields of each grid to obtain the pheromone content of the optional path; wherein, the weight of the grid decreases as the distance between the grid and the robot increases.

[0010] Optionally, in S400, the steps of controlling the corresponding robot to travel towards the target according to the preferred path and adjusting the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target include: S410, controlling the robot to travel along the preferred path and recording the grids passed by each robot; S420, for each grid, obtaining the robots passing through the grid at a certain time interval, and updating the concentration information field of the grid based on the priority index of the target corresponding to the robot, and the concentration information field is positively correlated with the priority index of the target corresponding to the robot; S430, obtaining each road section in the grid map, and determining the pheromone content of the road section based on the concentration information fields of the grids in the road section; S440, if it is determined that the pheromone content of the road section is lower than the preset content threshold, obtaining the passing grids of the road section, determining the passing grids corresponding to the robot based on the priority index of the target corresponding to the robot and the concentration information fields of each passing grid, and controlling each robot to pass through the road section according to the corresponding passing grids, and the passing grids are perpendicular to the traveling direction of the robot.

[0011] Optionally, in S440, the steps of determining the passing grid corresponding to the robot based on the priority index of the target corresponding to the robot and the concentration information fields of each passing grid include: S441, determining the passing index of each robot based on the priority index of the target corresponding to the robot and the grids passed by, and determining the passing order of each robot according to the passing index; S442, sorting the concentration information fields of each passing grid, and sequentially matching the passing grids corresponding to each robot according to the passing order.

[0012] On the other hand, an embodiment of the present invention provides a task planning system for a robot cluster, including: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0013] On the other hand, an embodiment of the present invention provides a computer-readable storage medium storing a program executable by a processor, and the program executable by the processor is used to execute the method described in any one of the above when executed by the processor.

[0014] The beneficial effects of the present invention are as follows: The present invention discloses a task planning method, device, and storage medium for a robot cluster. The present invention determines a priority index by obtaining the remaining duration and task weight of each target, determines an optional path between the target grid and each starting grid of each target, obtains the distance and congestion degree of the optional path, determines the cost of the optional path according to the distance and congestion degree, determines the preferred path and robot corresponding to each target according to the cost of the optional path; controls the corresponding robot to travel towards the target according to the preferred path, and adjusts the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid of the corresponding target, finally effectively planning the travel path of the robot and improving the task execution efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the embodiments. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings without creative efforts based on these drawings.

[0016] Figure 1 is a flowchart of a task planning method for a robot cluster according to an embodiment of the present invention; Figure 2 is a structural diagram of a task planning system for a robot cluster according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0017] The following will clearly and completely describe the concept, specific structure, and technical effects generated by the present invention disclosed in combination with the embodiments and the drawings, so as to fully understand the purpose, solution, and effects of the present invention disclosed. It should be noted that, without conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0018] In a dynamic environment, the challenges faced by a robot swarm are not limited to the judgment of task priorities, but also include the identification of path obstacles and the real-time optimization of path planning. The present invention provides a solution that can adapt to complex scenario changes and improve the cooperation efficiency of robots. By dynamically adjusting the driving order of robots, not only can the operation efficiency be improved, but also the timely completion of high-priority tasks can be ensured, thereby enhancing the flexibility and reliability of the overall operation process.

[0019] Reference Figure 1 , such as Figure 1 shown is a task planning method for a robot swarm provided by an embodiment of the present invention. The method includes the following steps: S100. Obtain the remaining duration and task weight of each target, and determine the priority index according to the remaining duration and task weight; It should be noted that taking the warehousing and logistics scenario as an example, the target is goods, and tasks such as picking up and delivering goods need to be performed on the goods. The remaining duration reflects the urgency of the target's task, and the task weight represents the importance of the task; the priority index is set according to the urgency and importance of the goods.

[0020] S200. Obtain a grid map, determine the optional paths between the starting grids of each robot in the robot swarm and the target grids of each target, and obtain the optional path set of each target; It should be noted that the robot swarm includes multiple robots. After obtaining the environmental map where the robot swarm is located, convert the environmental map into a grid map; the grids in the grid map are square, and the size of the grid matches the floor area of the robot; the driving speeds of each robot during driving are equal; use the grids of each robot in the grid map as the starting grids, and use the grids of each target in the grid map as the target grids; then, determine the optional paths between the target grids of each target and the starting grids of each robot in the robot swarm, and obtain the optional path set of each target.

[0021] S300. Obtain the distance and congestion degree of the optional path, determine the cost of the optional path according to the distance and congestion degree, and determine the optimal path and robot corresponding to each target according to the cost of the optional path; Specifically, sort all targets from high to low according to the priority index to obtain a sorted list; select targets from the sorted list in order, calculate the optional paths between the target grid and the starting grid, and ensure that the higher the priority index of the target, the earlier the optional path is determined. Then, from an overall perspective, optimize the optional paths of each target with the minimum overall cost, determine the optimal path of each target, and assign the robot corresponding to the optimal path to the target; after determining the optimal path and robot corresponding to each target, control the robot to execute the task corresponding to the target.

[0022] The S400 controls the corresponding robot to travel towards the target according to the preferred path, and adjusts the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target.

[0023] It should be noted that by obtaining the remaining duration and task weight of the target, the priority index is adjusted in real time to ensure that the task planning of the robot cluster always adapts to the current working environment, ensure that the robot cluster dynamically adjusts the path, and improve the task execution efficiency. The task planning based on priority and cost analysis ensures that the robot cluster can dynamically adjust the action strategy when executing multiple tasks to cope with the complex and changeable environment, thereby improving the overall task execution efficiency and reducing resource waste.

[0024] The present invention is applicable to complex and changeable environments, such as traffic jams or path obstacles. The robot cluster can quickly recalculate the path and adjust the route plan to maintain the continuity and efficiency of the task. Through this dynamic response mechanism, even in a dynamically changing environment, the robot can work efficiently to ensure the timely and safe delivery of goods. Effectively improve the collaborative operation ability of the robot cluster and ensure the efficient completion of tasks. The task planning of the present invention based on priority and cost analysis ensures that the robot cluster can dynamically adjust the action strategy when executing multiple tasks to cope with the complex and changeable environment, thereby improving the overall task execution efficiency and reducing resource waste.

[0025] In some embodiments, in S100, the determining the priority index according to the remaining duration and task weight includes: S110, obtaining the remaining duration and task weight of the target, dividing the task weight by the sum of the remaining duration of the target and a preset reference duration to obtain an initial index; S120, performing a normalization process on the initial index to obtain a priority index.

[0026] Specifically, the calculation formula of the priority index is: pj = norm(wj / (tj + e)); where norm() represents the normalization process, tj is the remaining duration of target j, e is the preset reference duration, e is a very small positive number to prevent the denominator from being zero; wj is the task weight; the shorter the remaining time and the greater the task weight (such as task importance, resource consumption), the larger the pj value and the higher the priority. By balancing the time urgency and task weight through the priority index. After determining the priority index pj, all targets are sorted from high to low according to the priority index pj to form a priority queue; queue = {target 1, target 2,..., target N}, p1 ≥ p2 ≥... ≥ pN.

[0027] In some embodiments, in S200, obtaining a grid map, determining optional paths between the starting grids of each robot in the robot cluster and the target grids of each target, and obtaining an optional path set for each target includes: S210, obtaining a grid map, using the grids of each robot in the robot cluster in the grid map as starting grids, and using the grids of each target in the grid map as target grids; S220, for each target, traversing each robot, respectively determining the optional paths between the target and the traversed robot, obtaining all optional paths between the starting grid of the robot and the target grid of the target, and obtaining a path group from the robot to the target; S230, using the path groups from each robot to the target as the optional path set for the target.

[0028] Specifically, after selecting a target, taking the target as an object, searching for all passable paths between the robot and the target to obtain a path group for the target; the path groups from the target to each robot form a database according to the corresponding robots, and the optional path set for the target is obtained.

[0029] In some embodiments, in S300, obtaining the distance and congestion degree of the optional path, determining the cost of the optional path according to the distance and congestion degree, and determining the preferred path and robot corresponding to each target according to the cost of the optional path includes: S310, obtaining an initialized pheromone matrix, determining the concentration information field of each grid on the optional path, and generating the pheromone content of the optional path based on the concentration information field of each grid; wherein, the pheromone matrix includes the concentration information field of each grid, the concentration information field represents the sum of the initial concentration value and the dynamic concentration value, and the dynamic concentration value is positively correlated with the number of times the grid is occupied; S320, multiplying the ratio of the current number of robots on the optional path to the section capacity by the historical congestion coefficient and then performing normalization processing to obtain the congestion degree of the optional path; S330, normalizing the ratio of the distance of the optional path to the priority index to obtain a heuristic factor, and performing weighted normalization processing on the heuristic factor, congestion degree, and pheromone content to obtain the cost of the optional path; Specifically, the cost of the alternative path is: Costijk = norm(λ1 * ηijk + λ2 * Conijk + λ3 * τijk); where Costijk is the cost of the k-th alternative path from robot i to target j, ηijk is the heuristic factor, Conijk is the congestion degree, and τijk is the pheromone content. The congestion degree Conijk = norm(h * Nijk / N), where h is the historical congestion coefficient, Nijk is the number of robots on the k-th alternative path from robot i to target j, N is the road section capacity (the maximum number of robots that can be carried), and the historical congestion coefficient is obtained by averaging the ratio of the average number of robots on the road section and the road section capacity over multiple past time periods; the heuristic factor ηijk = norm(dijk / pj), where dijk is the distance of the k-th alternative path from robot i to target j. τijk represents the pheromone content of the k-th alternative path from robot i to target j. The cost of the path includes the heuristic factor, the congestion degree, and the pheromone content, and guides the robot to select alternative paths through the heuristic factor, the congestion degree, and the pheromone content, avoiding relying on a single factor only.

[0030] S340. Traverse each target in descending order of the priority index, select a path group and the corresponding robot from the set of alternative paths of this target, and sequentially match to obtain the path group and robot corresponding to each target, and use the cost of each alternative path in the path group as the multiple costs corresponding to this target; Specifically, first select the target with the highest priority index, select a robot from the set of alternative paths of this target, obtain the path group from this robot to this target, determine the robot corresponding to this path group as the robot corresponding to this target, and mark this robot as occupied; then, sequentially select the next target, select a path group from the set of alternative paths of this target, determine the robot corresponding to this path group, exclude the occupied robot and the corresponding path group, and select a path group and the robot corresponding to this path group from the remaining path groups, and so on until all targets have selected the corresponding path group and robot; then, use the cost of each alternative path in the path group as the multiple costs corresponding to this target.

[0031] S350. Combine and normalize by selecting one cost from each of the multiple costs corresponding to each target to obtain the total cost index of all targets; S360. Select the combination with the smallest total cost index, determine the preferred paths corresponding to the costs of each target in this combination, and the robots corresponding to these preferred paths.

[0032] Specifically, regarding the robot and target allocation as a multi-objective allocation problem, the goal is to minimize the total cost index. Through this process, it is ensured that the robot can navigate efficiently in a complex environment, reducing the randomness and repetition of path selection. Finally, when all targets are effectively allocated to the corresponding robots, the preferred paths and robots corresponding to each target are output.

[0033] In this embodiment, predictions are made based on the road section capacity and expected traffic flow to generate dynamic pheromones, ensuring that the pheromone content can accurately reflect the actual traffic capacity of the path. In this way, the robot can effectively avoid high-congestion areas and select the optimal path, thereby improving the operation efficiency of the entire cluster and its ability to cope with complex environments. Finally, through continuous iterative optimization, stable and reliable navigation and path planning can be provided in a dynamically changing environment.

[0034] In some embodiments, in S310, the obtaining of the initialized pheromone matrix, determining the concentration information field of each grid on the optional path, and generating the pheromone content of the optional path based on the concentration information field of each grid includes: S311, obtaining the number of times each grid is occupied at a set time interval to obtain the average number of times each grid is occupied; Specifically, within the most recent past time period, obtain the number of times the grid is occupied at a set time interval to obtain the average number of times the grid is occupied; by statistically counting the occupancy frequency of each grid within a specific time, accurately reflect the recent usage status of the grid as the initialization information.

[0035] S312, determining the concentration change rate of the grid according to the ratio of the number of times occupied to the average number of times, and multiplying the initial concentration value of the grid by the concentration change rate to obtain the current concentration information field of the grid; Specifically, by calculating the ratio of the number of occupancy times to the average number of times, determine the current concentration change rate, ensure that the concentration information field is updated in real time, accurately reflect the actual occupancy situation of the grid, and provide a reliable basis for path planning.

[0036] S313, performing weighted normalization processing on the concentration information fields of each grid to obtain the pheromone content of the optional path; wherein, the weight of the grid decreases as the distance between the grid and the robot increases.

[0037] In the calculation of the pheromone concentration change rate, consider the number of times and time when the robot passes through the grid to simulate the volatilization and accumulation process of the pheromone. By calculating the concentration change rate of each grid and updating the pheromone matrix according to the set time interval. It can simulate the real-time state change of the path and ensure that the pheromone content can dynamically respond to environmental changes.

[0038] In some embodiments, in S400, controlling the corresponding robot to travel towards the target according to the preferred path and adjusting the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target includes: S410, controlling the robot to travel along the preferred path and recording the grids passed by each robot; S420, for each grid, obtaining the robots passing through the grid at each time interval, and updating the concentration information field of the grid based on the priority index of the target corresponding to the robot, where the concentration information field is positively correlated with the priority index of the target corresponding to the robot; Specifically, after accumulating the priority indexes of the targets corresponding to each robot passing through the grid, dividing by the maximum value among the priority indexes of each target to obtain a concentration index, and multiplying the concentration information field of the grid by the concentration index to obtain the updated concentration information field of the grid.

[0039] To avoid premature convergence of pheromones leading to local optimal solutions, a pheromone dynamic adjustment mechanism is introduced. In each update, the pheromone content of each section is dynamically adjusted to ensure that the algorithm has the ability to continuously explore. Specifically, the pheromone content released by the robot corresponding to the target with a high priority index is relatively high, indicating that the possibility of this grid being occupied by the robot corresponding to the target with a higher priority index in the future is lower, and other robots can give priority to passing through this grid, thereby increasing the guidance of this grid to other robots; conversely, the pheromone released by the robot with a low priority index decreases, thus driving the possibility of the ant colony algorithm to explore other grids. Through pheromone update, the optimization of the task assignment scheme is ensured.

[0040] S430, obtaining each section in the grid map and determining the pheromone content of the section based on the concentration information fields of the grids in the section; Specifically, the paths with intersections in the grid map are regarded as one section. In this way, the preferred path can be considered as multiple continuous sections. By obtaining the concentration information fields of the grids in the section and performing fusion processing on the concentration information fields of each grid, the pheromone content of the section is obtained, and the pheromone content of each section is updated in real time.

[0041] S440, if it is determined that the pheromone content of the section is lower than the preset content threshold, obtaining the passing grids of the section, determining the passing grids corresponding to the robot based on the priority index of the target corresponding to the robot and the concentration information fields of each passing grid, and controlling each robot to pass through the section according to the corresponding passing grids, where the passing grids are perpendicular to the traveling direction of the robot.

[0042] Specifically, the content threshold is pre-set according to the actual situation, and a reasonable threshold is obtained through experiments based on whether the road section can ensure the smooth passage of the robot. If the pheromone content of the road section is greater than or equal to the content threshold, it means that the robot can pass through the road section smoothly, and the robot can continue to drive. If the pheromone content of the road section is lower than the content threshold, the robot's driving path needs to be further optimized. Before the robot is about to enter the next road section according to the preferred path, the traffic grid of the road section, that is, multiple side-by-side grids perpendicular to the robot's driving direction, determines the traffic index of the traffic grid, and allocates the grid that the robot is about to enter. For each robot to enter the road section, it passes through the road section according to the corresponding traffic grid.

[0043] In some embodiments, in S440, determining the passage grid corresponding to the robot based on the priority index of the robot's corresponding target and the concentration information field of each passage grid includes: S441, determining a passage index of each robot based on the priority index of the robot's corresponding target and the grids it has traveled through, and determining a passage order of each robot according to the passage index; Specifically, if multiple robots choose the same road section at the same time, the shorter the grid they travel through and the higher the priority index, the higher the pass index. In this way, when a customer's urgent order triggers a high priority index, the pass index of each robot is determined based on the task completion status and robot performance to ensure that orders with a high priority index are processed first.

[0044] S442, sorting the concentration information fields of each pass grid, and matching the pass grids corresponding to each robot in order of passage.

[0045] Specifically, robots with higher pass orders preferentially match pass grids with higher concentration information fields, thereby improving the communication efficiency of the robots.

[0046] refer to Figure 2 , an embodiment of the present invention further provides a task planning system for a robot cluster, comprising: at least one processor; at least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the above method.

[0047] The contents of the above method embodiments are all applicable to this embodiment. The functions specifically implemented by this embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those achieved by the above method embodiments, which will not be repeated here.

[0048] In addition, an embodiment of the present invention further provides a computer-readable storage medium storing computer-executable instructions for executing the above-described air conditioner control method. Exemplarily, it executes the method steps in Figure 1 the above.

[0049] Since the computer-readable storage medium of the embodiment of the present invention can execute the air conditioner control method of any of the above embodiments, the specific implementation manners and technical effects of the computer-readable storage medium of the embodiment of the present invention can refer to the specific implementation manners and technical effects of the air conditioner control method of any of the above embodiments.

[0050] In addition, an embodiment of the present application also discloses a computer program product or a computer program stored in a computer-readable storage medium. The processor of the computer device can read the computer program from the computer-readable storage medium, and the processor executes the computer program to enable the computer device to execute the above method. Similarly, the content in the above method embodiments is applicable to this storage medium embodiment. The functions specifically implemented by this storage medium embodiment are the same as those of the above method embodiments, and the beneficial effects achieved are also the same as those of the above method embodiments.

[0051] Those of ordinary skill in the art can understand that all or some of the methods and systems disclosed above can be implemented as software, firmware, hardware, and appropriate combinations thereof. Some physical components or all physical components can be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or can be implemented as hardware, or can be implemented as an integrated circuit, such as an application-specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or a non-transitory medium) and a communication medium (or a transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include, but are not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disc (DVD) or other optical disc storage, magnetic cassette, tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those of ordinary skill in the art, a communication medium typically includes computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

[0052] The above is a specific description of the preferred embodiments of the present disclosure. However, the present disclosure is not limited to the above embodiments. Those skilled in the art can make various equivalent deformations or substitutions without departing from the spirit of the present disclosure, and these equivalent deformations or substitutions are all included within the scope defined by the claims of the present disclosure.

Claims

1. A task planning method for a robot cluster, characterized in that, The method includes the following steps: S100. Obtain the remaining duration and task weight of each target, and determine a priority index according to the remaining duration and task weight; S200. Obtain a grid map, determine optional paths between the starting grids of the robots in the robot cluster and the target grids of each target, and obtain a set of optional paths for each target; S300. Obtain the distance and congestion degree of the optional paths, determine the cost of the optional paths according to the distance and congestion degree, and determine the optimal paths and robots corresponding to each target according to the cost of the optional paths; S400. Control the corresponding robot to travel towards the target according to the optimal path, and adjust the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target.

2. The task planning method for a robot cluster according to claim 1, characterized in that, In S100, the determining the priority index according to the remaining duration and task weight includes: S110. Obtain the remaining duration and task weight of the target, divide the task weight by the sum of the remaining duration of the target and a preset reference duration, and obtain an initial index; S120. Perform a normalization process on the initial index to obtain a priority index.

3. The task planning method for a robot cluster according to claim 1, characterized in that, In S200, the obtaining the grid map, determining the optional paths between the starting grids of the robots in the robot cluster and the target grids of each target, and obtaining a set of optional paths for each target includes: S210. Obtain the grid map, use the grids of the robots in the robot cluster in the grid map as starting grids, and use the grids of each target in the grid map as target grids; S220. For each target, traverse each robot, respectively determine the optional paths between the target and the traversed robot, obtain all the optional paths between the starting grid of the robot and the target grid of the target, and obtain a path group from the robot to the target; S230. Use the path groups from each robot to the target as the set of optional paths for the target.

4. A task planning method for a robot cluster according to claim 1, characterized in that In S300, the obtaining the distance and congestion degree of the optional paths, determining the cost of the optional paths according to the distance and congestion degree, and determining the optimal paths and robots corresponding to each target according to the cost of the optional paths includes: S310. Obtain an initialized pheromone matrix, determine the concentration information field of each grid on the optional path, and generate the pheromone content of the optional path based on the concentration information field of each grid; wherein, the pheromone matrix contains the concentration information fields of each grid, the concentration information field represents the sum of the initial concentration value and the dynamic concentration value, and the dynamic concentration value is positively correlated with the number of times the grid is occupied; S320. Multiply the ratio of the current number of robots on the optional path to the section capacity by the historical congestion coefficient and then perform a normalization process to obtain the congestion degree of the optional path; S330. Perform a normalization process on the ratio of the distance of the optional path to the priority index to obtain a heuristic factor, and perform a weighted normalization process on the heuristic factor, congestion degree, and pheromone content to obtain the cost of the optional path; S340. Traverse each target in order from the highest to the lowest priority index, select a path group and the corresponding robot from the set of optional paths of the target, and sequentially match to obtain the path group and robot corresponding to each target. Use the cost of each optional path in the path group as the multiple costs corresponding to the target. S350. Select one cost from the multiple costs corresponding to each target for combination and then perform normalization processing to obtain the total cost index of all targets. S360. Select the combination with the smallest total cost index, determine the preferred path corresponding to the cost of each target in the combination, and the robot corresponding to the preferred path.

5. The task planning method for a robot cluster according to claim 4, characterized in that, In S310, the step of obtaining the initialized pheromone matrix, determining the concentration information field of each grid on the optional path, and generating the pheromone content of the optional path based on the concentration information field of each grid includes: S311. Obtain the number of times each grid is occupied at a set time interval to obtain the average number of times each grid is occupied. S312. Determine the concentration change rate of the grid according to the ratio of the number of times occupied to the average number of times, and multiply the initial concentration value of the grid by the concentration change rate to obtain the current concentration information field of the grid. S313. Perform weighted normalization processing on the concentration information fields of each grid to obtain the pheromone content of the optional path; wherein, the weight of the grid decreases as the distance between the grid and the robot increases.

6. The task planning method for a robot cluster according to claim 1, wherein In S400, the step of controlling the corresponding robot to travel towards the target according to the preferred path and adjusting the travel path of the robot in real time according to the priority index of the target until the robot reaches the target grid corresponding to the target includes: S410. Control the robot to travel according to the preferred path and record the grids passed by each robot. S420. For each grid, obtain the robots passing through the grid at each time interval, and update the concentration information field of the grid based on the priority index of the target corresponding to the robot. The concentration information field is positively correlated with the priority index of the target corresponding to the robot. S430. Obtain each road section in the grid map, and determine the pheromone content of the road section based on the concentration information fields of the grids in the road section. S440. If it is determined that the pheromone content of the road section is lower than the preset content threshold, obtain the passing grids of the road section, determine the passing grids corresponding to the robot based on the priority index of the target corresponding to the robot and the concentration information fields of each passing grid, and control each robot to pass through the road section according to the corresponding passing grids. The passing grids are perpendicular to the traveling direction of the robot.

7. The task planning method for a robot cluster according to claim 6, wherein, In S440, the step of determining the passing grids corresponding to the robot based on the priority index of the target corresponding to the robot and the concentration information fields of each passing grid includes: S441. Determine the passing index of each robot based on the priority index of the target corresponding to the robot and the grids passed through, and determine the passing order of each robot according to the passing index. S442. Sort the concentration information fields of each passing grid, and sequentially match the passing grids corresponding to each robot according to the passing order.

8. A task planning system for a robot cluster, characterized in that, Includes: At least one processor; At least one memory for storing at least one program; When the at least one program is executed by the at least one processor, the at least one processor implements the method according to any one of claims 1 to 7.

9. A computer-readable storage medium storing a program executable by a processor, characterized in that, The program executable by the processor is used to execute the method according to any one of claims 1 to 7 when executed by the processor.

Images