Task scheduling method of robot and related equipment

Abstract

The embodiment of the invention provides a task scheduling method of a robot and related equipment, which are used for improving the task allocation and scheduling efficiency of a multi-robot system. The method comprises the following steps: acquiring current state information of each to-be-scheduled robot in a robot cluster; generating a target task allocation scheme corresponding to each to-be-scheduled robot according to the current state information; generating a current task allocation instruction of the target task allocation scheme, and allocating the target task allocation scheme to each corresponding to-be-scheduled robot according to the current task allocation instruction so as to control each to-be-scheduled robot to execute the target task allocation scheme; obtaining each to-be-scheduled robot, executing current task execution data in the target task allocation scheme, and adjusting the target task allocation scheme of each to-be-scheduled robot according to the current task execution data to obtain a target task scheduling strategy, and enabling each scheduling robot to execute the target task allocation scheme allocated by the target task scheduling strategy.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of intelligent device technology, and in particular, to a robot task scheduling method and related equipment. Background Art

[0002] In the field of modern industry and logistics, multi-robot systems have been widely used in task execution, warehouse management, automated production lines, etc. These robots usually interact with the control center through various communication protocols (such as Hypertext Transfer Protocol (HTTP), Transmission Control Protocol (TCP), User Datagram Protocol (User Datagram Protocol), etc.), report their own status information in real time, and receive task instructions. However, the existing technology has many problems in task allocation and scheduling optimization, especially when dealing with complex multi-robot task scheduling, it is easy to fall into the local optimal solution and it is difficult to achieve global optimal scheduling. Summary of the invention

[0003] The embodiments of the present application provide a robot task scheduling method and related equipment for improving the task allocation and scheduling efficiency of a multi-robot system.

[0004] A first aspect of an embodiment of the present application provides a task scheduling method for a robot, comprising:

[0005] Obtain the current status information of each robot to be scheduled in the robot cluster;

[0006] Generating a target task allocation scheme corresponding to each robot to be scheduled according to the current state information;

[0007] Generate a current task allocation instruction of the target task allocation scheme, and allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme;

[0008] Obtain the current task execution data of each robot to be scheduled in executing the target task allocation scheme, and adjust the target task allocation scheme of each robot to be scheduled according to the current task execution data to obtain the target task scheduling strategy, so that each scheduled robot executes the target task allocation scheme after being assigned by the target task scheduling strategy.

[0009] Optionally, generating a target task allocation scheme corresponding to each robot to be scheduled according to the current state information includes:

[0010] For each of the robots to be scheduled, generating an initial task allocation plan;

[0011] According to the current state information, an initial fitness value corresponding to the initial task allocation scheme is obtained; wherein the initial fitness value is used to characterize the current state information and the task execution data when each robot to be scheduled executes the initial task allocation scheme;

[0012] Control each of the robots to be scheduled to execute the initial task allocation plan to obtain a current fitness value;

[0013] Compare the current fitness value with the initial fitness value, and adjust the initial task allocation scheme of the current robot to be scheduled based on the particle swarm optimization PSO algorithm and the comparison result to obtain a target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme; wherein the current robot to be scheduled is any robot to be scheduled that executes the initial task allocation scheme;

[0014] If the target fitness value is the optimal fitness value, the adjusted initial task allocation scheme is determined as the target task allocation scheme.

[0015] Optionally, after obtaining the target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme, the method further includes:

[0016] Obtaining the target fitness value of all robots to be scheduled to execute the adjusted initial task allocation plan;

[0017] Compare all target fitness values, find the optimal fitness value among all target fitness values, and determine the task allocation scheme corresponding to the optimal fitness value; wherein the technical person to be scheduled corresponding to the optimal fitness value is the best scheduling robot;

[0018] According to the task allocation scheme corresponding to the optimal fitness value, the initial task allocation scheme of the valid robots to be scheduled is adjusted to obtain the target task allocation scheme corresponding to each robot to be scheduled; wherein the valid robots to be scheduled are the other robots to be scheduled among all the robots to be scheduled except the optimal scheduling robot.

[0019] Optionally, the obtaining of current task execution data of each robot to be scheduled in executing the target task allocation scheme includes:

[0020] During the execution of the target task allocation scheme, each robot to be scheduled is monitored based on the deep deterministic policy gradient DDPG algorithm to obtain real-time status information of all robots to be scheduled; wherein the real-time status information is used to characterize the real-time operating status of all robots to be scheduled;

[0021] Using the real-time status information as the state input of the DDPG algorithm to obtain an operation state vector; wherein the operation state vector includes the operation states of all the robots to be scheduled;

[0022] The current task execution data is obtained according to the running state vector.

[0023] Optionally, adjusting the target task allocation scheme of each to-be-scheduled robot according to the current task execution data to obtain a target task scheduling strategy includes:

[0024] The running state vector is evaluated to obtain the target state action expectation of all the robots to be scheduled to execute the target task allocation scheme in the real-time running state; wherein the target state action expectation is used to characterize the expectation of different actions taken by all the robots to be scheduled when executing the target task allocation scheme;

[0025] Obtaining historical state action expectations, and determining target task evaluation data according to the target state action expectations and the historical state action expectations; wherein the historical state actions are used to characterize the expectations of different actions taken by all robots to be scheduled when executing any task allocation scheme in history;

[0026] The target task scheduling strategy is generated according to the target task evaluation data.

[0027] Optionally, after obtaining the target task scheduling strategy, the method further includes:

[0028] Generate a real-time task allocation instruction according to the target task scheduling strategy; wherein the real-time task allocation instruction is used to instruct each robot to be scheduled to adjust one or more of the task priority, task planning path or task start and stop status of the target task allocation scheme;

[0029] According to the real-time task allocation instruction, one or more of the task priority, the task planning path or the task start / stop status in each robot to be scheduled corresponding to the target task allocation scheme is adjusted to obtain the allocated target task allocation scheme.

[0030] Optionally, after obtaining the target task scheduling strategy, the method further includes:

[0031] After all target task allocation plans are executed, target task execution data of all target task allocation plans are recorded;

[0032] Analyze the target task execution data to adjust the next task allocation plan according to the target task execution data to obtain the next task scheduling strategy; wherein the next task allocation plan is the next task allocation plan of each robot to be scheduled, and the next task scheduling strategy is the next task scheduling strategy of each robot to be scheduled after adjusting the next task allocation plan.

[0033] A second aspect of an embodiment of the present application provides a robot task scheduling system, comprising:

[0034] An acquisition unit, used to acquire current status information of each robot to be scheduled in the robot cluster;

[0035] A generating unit, configured to generate a target task allocation scheme corresponding to each of the robots to be scheduled according to the current state information;

[0036] an allocation unit, configured to generate a current task allocation instruction of the target task allocation scheme, and allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme;

[0037] An adjustment unit is used to obtain the current task execution data of each robot to be scheduled in executing the target task allocation scheme, and adjust the target task allocation scheme of each robot to be scheduled according to the current task execution data to obtain a target task scheduling strategy so that each scheduled robot executes the target task allocation scheme after being assigned by the target task scheduling strategy.

[0038] The robot task scheduling system provided in the second aspect of the embodiment of the present application is used to execute the robot task scheduling method described in the first aspect.

[0039] A third aspect of an embodiment of the present application provides a task scheduling device for a robot, comprising:

[0040] CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply;

[0041] The memory is a transient storage memory or a persistent storage memory;

[0042] The central processing unit is configured to communicate with the memory and execute instruction operations in the memory to perform the robot task scheduling method described in the first aspect.

[0043] A fourth aspect of an embodiment of the present application provides a computer-readable storage medium, wherein the computer-readable storage medium includes instructions. When the instructions are executed on a computer, the computer executes the robot task scheduling method described in the first aspect.

[0044] A fifth aspect of an embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer, the computer executes the robot task scheduling method described in the first aspect.

[0045] It can be seen from the above technical scheme that the embodiment of the present application has the following advantages: through a robot task scheduling method disclosed in the embodiment of the present application, the current state information of each robot to be scheduled in the robot cluster is first obtained; then, according to the current state information, a target task allocation scheme corresponding to each robot to be scheduled is generated; then, the current task allocation instruction of the target task allocation scheme is generated, and the target task allocation scheme is allocated to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme; finally, the current task execution data of each robot to be scheduled in the target task allocation scheme is obtained, and the target task allocation scheme of each robot to be scheduled is adjusted according to the current task execution data, and the target task scheduling strategy is obtained, so that each scheduling robot executes the target task allocation scheme after the target task scheduling strategy is assigned. Thus, by adaptively adjusting Lili in a dynamic environment, the optimal task allocation scheme is found in the global scope, and the strategy is adjusted according to the real-time robot status information, so as to realize efficient scheduling of the multi-robot system, and has good adaptability. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the present application. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0047] Figure 1 A schematic diagram of a process flow of a robot task scheduling method disclosed in an embodiment of the present application;

[0048] Figure 2 A schematic flow chart of another robot task scheduling method disclosed in an embodiment of the present application;

[0049] Figure 3 A schematic flow chart of another robot task scheduling method disclosed in an embodiment of the present application;

[0050] Figure 4A schematic diagram of the structure of a robot task scheduling system disclosed in an embodiment of the present application;

[0051] Figure 5 A schematic diagram of the structure of a task scheduling device for a robot disclosed in an embodiment of the present application. DETAILED DESCRIPTION

[0052] In the field of modern industry and logistics, multi-robot systems have been widely used in task execution, warehouse management, automated production lines, etc. These robots usually interact with the control center through various communication protocols (such as HTTP, TCP, UDP, etc.), report their own status information in real time, and receive task instructions. However, the prior art has many problems in task allocation and scheduling optimization, especially when dealing with complex multi-robot task scheduling, it is easy to fall into the local optimal solution and it is difficult to achieve global optimal scheduling. With the development of intelligent algorithms, reinforcement learning and optimization algorithms are gradually applied to the scheduling optimization of multi-robot systems. However, it is often difficult to take into account both global search capabilities and dynamic environment adaptability when using reinforcement learning or traditional optimization algorithms alone. Specifically, the embodiment of the present application proposes a solution that combines deep deterministic policy gradient (DDPG, deep deterministic policy gradient) and reinforcement learning algorithm with particle swarm optimization (PSO, Particle Swarm Optimization) algorithm, aiming to improve the task allocation and scheduling efficiency of multi-robot systems.

[0053] The following will be combined with the drawings in the embodiments of the present application to clearly and completely describe the technical solutions in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0054] To solve the technical problems mentioned above, please refer to Figure 1 , Figure 1 This is a flowchart of a robot task scheduling method disclosed in an embodiment of the present application, including steps 101 to 104.

[0055] 101. Obtain the current status information of each robot to be scheduled in the robot cluster.

[0056] It should be noted that the present embodiment can be run in a multi-robot system. Among them, the multi-robot system includes multiple robots and a control device. The control device controls the movement or stop of the robot by sending instructions or other means, and the specifics are not limited here. Furthermore, the robot mentioned above is the robot to be scheduled described above. Among them, the robot to be scheduled can be an intelligent handling robot (AGV, automated guided vehicle), or other types of robots, which are not limited or repeated here.

[0057] In this embodiment, the multi-robot system needs to obtain the current state information of each robot to be scheduled in the robot cluster.

[0058] In one specific embodiment, when the multi-robot system is ready to start a task, task scheduling can be started to obtain the robot status. Specifically, the communication module in the multi-robot system will first collect the status information of each robot (i.e., the current status information mentioned above). This status information includes but is not limited to the current coordinate position of the robot, the battery power, the execution progress of the assigned task, or the current execution status. This status information provides key data support for task allocation and scheduling.

[0059] 102. Generate a target task allocation plan corresponding to each robot to be scheduled based on the current state information.

[0060] Based on step 101, a target task allocation plan corresponding to each robot to be scheduled can be generated according to the current state information.

[0061] In one specific embodiment, the multi-robot system can combine the current state information of each robot to be scheduled as a condition or initial value. The multi-robot system first uses the PSO algorithm to initialize the task allocation plan, and then simulates the current state information as a condition or initial value to obtain the target task allocation plan corresponding to each robot to be scheduled. Specifically, PSO determines a global optimal task allocation plan by simulating the process of a group of particles searching for the optimal solution in the search space.

[0062] In one specific embodiment, the multi-robot system regards each robot to be scheduled (AGV) as a "particle", and each "particle" represents a possible task allocation scheme. For example, task A can be completed by AGV1, task B can be completed by AGV2, and so on. Then, the multi-robot system combines the current state information of each robot to be scheduled, evaluates and simulates the current coordinate position, battery power, execution progress of the assigned tasks, or the current execution status, and then adjusts the task allocation scheme of each robot to be scheduled, so as to gradually find a task allocation scheme that will gradually approach the global optimal solution, thereby ensuring that the overall efficiency of task allocation is the highest and the energy consumption is the lowest. Among them, the task allocation scheme is the target task allocation scheme described above.

[0063] 103. Generate a current task allocation instruction of the target task allocation plan, and allocate the target task allocation plan to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation plan.

[0064] Based on step 102, after obtaining the target task allocation scheme, the multi-robot system can generate the current task allocation quality of the target task allocation scheme, and then allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction. Thus, each robot to be scheduled can be controlled to execute the target task allocation scheme through the current task allocation instruction.

[0065] In one specific embodiment, after obtaining the task allocation scheme for each robot to be scheduled, it is necessary to perform task allocation, wherein it is necessary to send instructions for the task allocation. Specifically, according to the initial strategy provided by PSO, the multi-robot system allocates the target task scheme to each robot. Among them, specific instructions can be sent (i.e., the current task allocation quality described above). After receiving the instructions, the robot to be scheduled begins to execute the task as planned. This step marks the official start of task scheduling, and each robot to be scheduled is clear about its work goals.

[0066] 104. Obtain the current task execution data of each robot to be scheduled in the target task allocation plan, and adjust the target task allocation plan of each robot to be scheduled according to the current task execution data to obtain the target task scheduling strategy, so that each scheduled robot executes the target task allocation plan after being assigned by the target task scheduling strategy.

[0067] Based on step 103, the current task execution data of each robot to be scheduled in the process of executing the target task allocation scheme can be obtained. Then, the target task allocation scheme of each robot to be scheduled is adjusted according to the current task execution data to obtain the target task scheduling strategy. Thus, each scheduling robot executes the target task allocation scheme after being assigned by the target task scheduling strategy.

[0068] In one specific embodiment, after sending the current task allocation instruction, the multi-robot system will also monitor the status information of each robot to be scheduled in real time. In the process of task execution, the multi-robot system will use the DDPG algorithm for real-time monitoring and adjustment. In the DDPG algorithm, the multi-robot system will dynamically adjust the task allocation strategy according to the real-time data fed back during the robot's task execution. Specifically, the current task execution data of the robot to be scheduled in the process of executing the target task allocation scheme is first obtained, and then the current task execution data is used as real-time data and the state is input into the DDPG algorithm to evaluate the real-time data. Then, according to the subsequent evaluation results and combined with the current state of each robot to be scheduled, the strategy network of the multi-robot system will output or generate new action instructions, thereby obtaining the target task scheduling strategy. In this case, the target task scheduling strategy includes but is not limited to adjusting the order of task execution, changing the path, or transferring the task to other more suitable robots. Therefore, in combination with the above description, the multi-robot system can execute the new target task allocation scheme after redistribution through the target scheduling strategy.

[0069] According to a robot task scheduling method disclosed in this embodiment, the current state information of each robot to be scheduled in the robot cluster is first obtained; then, according to the current state information, a target task allocation scheme corresponding to each robot to be scheduled is generated; then, the current task allocation instruction of the target task allocation scheme is generated, and the target task allocation scheme is allocated to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme; finally, the current task execution data of each robot to be scheduled in the target task allocation scheme is obtained, and the target task allocation scheme of each robot to be scheduled is adjusted according to the current task execution data, and the target task scheduling strategy is obtained, so that each scheduling robot executes the target task allocation scheme after being assigned by the target task scheduling strategy. Therefore, by adaptively adjusting Lili in a dynamic environment, the optimal task allocation scheme is found in the global scope, and the strategy is adjusted according to the real-time robot status information, so as to realize efficient scheduling of the multi-robot system, and has good adaptability.

[0070] For the above Figure 1 For steps 101 to 103 shown in FIG. Figure 2 , Figure 2 This is a flow chart of another robot task scheduling method disclosed in an embodiment of the present application, including steps 201 to 207.

[0071] 201. Obtain current status information of each robot to be scheduled in the robot cluster.

[0072] In this embodiment, step 201 is the same as the aforementioned Figure 1 It is similar to step 101 in the above description and will not be described in detail here. However, it should be noted that in this embodiment, for the current status information, the urgency and complexity of the task, as well as the priority of the current robot position, can be used to quickly determine which robots are more suitable for performing specific tasks. For example, a robot that is closer to the target task and has sufficient power will be given priority. Based on the coordinate position of the robot currently to be scheduled, the multi-robot system can calculate the optimal path for the robot to reach the target task. For possible obstacles or path congestion, the multi-robot system will give priority to robots that can avoid these problems. If the robot is low on power, the multi-robot system will select other robots with more sufficient power to perform the task, or adjust the priority of the task of the current robot to be scheduled to avoid running out of power.

[0073] For ease of understanding, this will be described in detail later.

[0074] 202. For each robot to be scheduled, an initial task allocation plan is generated, and an initial fitness value corresponding to the initial task allocation plan is obtained according to the current state information.

[0075] Based on step 201, before obtaining the target task allocation plan, it is necessary to generate an initial task allocation plan for each robot to be scheduled, so as to obtain an initial fitness value corresponding to the initial task allocation plan according to the current state information. It should be noted that the initial fitness value is used to characterize the current state information and the task execution data when each robot to be scheduled executes the initial task allocation plan.

[0076] In one specific embodiment, the PSO algorithm may be used to initialize the task allocation scheme. To facilitate understanding of the PSO algorithm, a description is given below.

[0077] The PSO algorithm is an optimization algorithm that simulates the foraging behavior of bird flocks in nature. In this algorithm, "particles" are like individuals in a flock of birds. Each particle has its own position and speed. They will "fly" in the search space and find the optimal solution (just like a flock of birds looking for food). The global optimization of PSO means that by letting these particles communicate with each other in the entire search space, a globally optimal solution is finally found. In a multi-robot system, task allocation is a key issue. For example, there is a group of robots to complete different tasks, and we need to decide which robot performs which task. This process is not simple, because robots may have different capabilities and states, and we also need to consider the urgency of the task, the length of the path, etc. The role of PSO here is to help find the optimal task allocation method. Suppose there are three robots (A, B, C) and three tasks (T1, T2, T3). In the absence of PSO, it may be allocated through simple rules, such as allocating robots according to the closest task. However, this method is prone to fall into local optimality. For example, a robot may perform a certain type of task for a long time and ignore other more suitable tasks. After the introduction of PSO, each robot will run into different task allocation schemes (i.e. try different task combinations), and adjust and update the task allocation strategy based on feedback (such as task completion time, path length, etc.). After multiple iterations, the robot finally finds an overall better allocation method, such as minimizing energy consumption and task completion time at the same time.

[0078] In this embodiment, specifically, in the initialization phase, the system regards each robot to be scheduled, i.e., AGV, as a "particle", and each "particle" represents a possible task allocation scheme. For example, task A can be completed by AGV1, task B by AGV2, and so on. Initially, these "particles" will randomly generate different task allocation schemes, each of which has a corresponding "fitness value", which is determined by factors such as the completion time and energy consumption of the task. It should be noted that the fitness value is obtained based on the state information of the robot mentioned above. The fitness value is calculated based on the state information of the robot and the feedback of the actual execution of the task. For example: Task completion time: The faster the task is completed, the higher the fitness value. Energy consumption: The lower the energy consumption, the higher the fitness value. Path status: If the path passed by the robot is less blocked, the fitness value will be higher. Robot status: If a certain AGV is low in power or overloaded, its fitness value will be lower because it may not be able to complete the task efficiently. Each "particle" represents a possible task allocation scheme. For example, particle 1 may mean "task A is completed by AGV1, and task B is completed by AGV2"; particle 2 may mean "task A is completed by AGV2, and task B is completed by AGV1". These particles are initially generated randomly, that is, each robot is randomly assigned a task. Subsequently, the system will evaluate the scheme based on the fitness value of each particle (task allocation scheme). The adjustment and optimization of the task allocation scheme will be based on the execution feedback of each robot, and will be gradually adjusted in a better direction.

[0079] 203. Control each robot to be scheduled to execute the initial task allocation plan to obtain the current fitness value.

[0080] Based on step 202, after the initialization phase is completed, the task allocation scheme adjustment and evaluation phase is entered. The task allocation scheme of each AGV needs to be adjusted and evaluated. Specifically, each robot to be scheduled is controlled to execute the initial task allocation scheme to obtain the current fitness value.

[0081] In one specific embodiment, since each robot to be scheduled (AGV) will perform tasks according to the current task allocation plan and record the completion status of the task (such as task execution time, energy consumption, etc.), the multi-robot system can obtain the current fitness value of the AGV when executing the initial task allocation plan by controlling each robot to be scheduled to execute the initial task allocation plan.

[0082] 204. Compare the current fitness value with the initial fitness value, and adjust the initial task allocation plan of the current robot to be scheduled based on the particle swarm optimization PSO algorithm and the comparison result to obtain a target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation plan.

[0083] Based on step 203, by comparing the current fitness value with the initial fitness value, the initial task allocation scheme of the current robot to be scheduled can be adjusted based on the particle swarm optimization PSO algorithm and the comparison result, thereby obtaining the target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme. It should be noted that the current robot to be scheduled is a robot to be scheduled that executes the initial task allocation scheme, and can also be an AGV to be scheduled for other execution sequences arranged by the multi-robot system.

[0084] In one specific embodiment, the multi-robot system compares the current fitness value with the initial fitness value to determine whether the overall efficiency is improved when different robots to be scheduled execute a certain task allocation plan. The task allocation plan is constantly adjusted, and the result is compared with the previous plan to select a better plan as a reference. Furthermore, in each round of iteration, each AGV fine-tunes its task allocation to find a better solution under certain conditions, such as letting different robots handle different tasks.

[0085] 205. Obtain target fitness values ​​of all robots to be scheduled to execute the adjusted initial task allocation plan, compare the target fitness values, find the optimal fitness value among all target fitness values, and determine the task allocation plan corresponding to the optimal fitness value.

[0086] Then, the multi-robot system can obtain the target fitness values ​​of all robots to be scheduled to execute the adjusted initial task allocation plan, and then compare the target fitness values, find the optimal fitness value among all target fitness values, and determine the task allocation plan corresponding to the optimal fitness value. It should be noted that the technical person to be scheduled corresponding to the optimal fitness value is the best scheduling robot.

[0087] In one specific embodiment, in combination with step 204, the multi-robot system obtains the target fitness values ​​of all AGVs after executing the adjusted initial task allocation scheme, and then compares all the target fitness values, so as to screen out the fitness value that meets the overall optimum, that is, the optimal fitness value. The optimal fitness value is a task allocation scheme under a specific condition corresponding to all robots to be scheduled.

[0088] Furthermore, the multi-robot system finds the local optimal solution through group collaboration (sharing information between AGVs), and then gradually optimizes to the global optimal solution. This means that the task allocation solution finally found by the system can perform better under multiple indicators (such as time, energy consumption, path efficiency, etc.) at the same time. Therefore, after multiple rounds of iterations, the target fitness value of the initial task allocation solution after the current robot to be scheduled can be obtained.

[0089] It should be noted that step 204 to step 205 in this embodiment belong to the group collaboration process in the PSO algorithm, which will be described later.

[0090] 206. According to the task allocation scheme corresponding to the optimal fitness value, the initial task allocation scheme of the effective robots to be scheduled is adjusted to obtain a target task allocation scheme corresponding to each robot to be scheduled.

[0091] Then, the initial task allocation scheme of the effective robots to be scheduled can be adjusted according to the task allocation scheme corresponding to the optimal fitness value to obtain the target task allocation scheme corresponding to each robot to be scheduled. It should be noted that the effective robots to be scheduled are all robots to be scheduled except the best scheduled robot. For example, a multi-robot system includes AGV1, AGV2 and AGV3. Among them, the best robot to be scheduled is AGV2, then the effective robots to be scheduled are AGV1 and AGV3.

[0092] In one specific embodiment, after finding the task allocation scheme with the optimal fitness value, the initial task allocation of the valid robots to be scheduled can be adjusted, specifically, the task execution path or priority can be adjusted, which will not be described in detail here. Then, the target task allocation scheme corresponding to all robots to be scheduled can be determined.

[0093] Furthermore, all AGVs can share the optimal task allocation solutions they have found. If an AGV finds a better way to allocate tasks, other AGVs will refer to this optimal solution and adjust their task allocation accordingly. This step ensures that the global optimal task allocation solution can be quickly disseminated and applied.

[0094] 207. When the target fitness value is the optimal fitness value, the adjusted initial task allocation plan is determined as the target task allocation plan.

[0095] Based on step 204, if the target fitness value at this time is the optimal fitness value, the adjusted initial task allocation scheme can be directly determined as the target task allocation scheme.

[0096] In one specific embodiment, when the target fitness value is determined to be the optimal fitness value, it can be determined that the adjusted initial task allocation scheme is the global optimal task allocation scheme, and thus the global optimal task allocation scheme is determined as the target allocation scheme.

[0097] In other feasible technical solutions, global updates and optimizations can also be performed. Specifically, the multi-robot will record the currently found global optimal task allocation solution (among all possible task allocation solutions, find one that performs best under specific conditions, such as minimizing task completion time, energy consumption, path optimization, etc.), and in the next round of allocation, all AGVs will move closer to this optimal solution. Through continuous iterative optimization, the task allocation solution will gradually approach the global optimal solution, ensuring that the overall efficiency of task allocation is the highest and the energy consumption is the lowest. After multiple iterations, the AGV will find a globally optimal task allocation solution. Under this solution, each AGV is assigned the most suitable task, so that the task can be completed in the shortest time and energy savings can be maximized.

[0098] It should be noted that, in this embodiment, steps 204 to 207 are the group collaboration stage in the PSO algorithm. In the group collaboration stage, all AGVs share the optimal task allocation plan and try to adjust it. Then the multi-robot system performs global updates and optimizations to determine whether the best task allocation plan is achieved under the characteristic conditions. The best performing task allocation plan is the target task allocation plan described above.

[0099] Through the task scheduling method of a robot disclosed in this embodiment, through the PSO algorithm, the AGV system can effectively find and optimize the task allocation scheme to ensure the efficiency and intelligence of the entire task scheduling. This method combines the advantages of global search and local adjustment, and ultimately enables the AGV system to quickly find the best task allocation strategy in a complex task environment. This embodiment assigns each task to the most suitable robot and inputs it into the DDPG algorithm as an initial strategy. This step ensures that the basic scheme for task allocation has been optimized and fully considered. Furthermore, after combining the PSO algorithm, the multi-robot system can comprehensively consider the collaborative relationship between multiple robots and find a better task allocation scheme on a global scale. For example, the system may find that robot B is allowed to perform task T1 instead of A, even though A is closer to T1, but B has a higher execution efficiency. This global optimization can significantly improve the overall efficiency of the system and reduce the time and energy consumption for task completion.

[0100] For the above Figure 1 Step 104 shown in FIG. Figure 3 , Figure 3 This is a flow chart of another robot task scheduling method disclosed in an embodiment of the present application, including steps 301 to 307.

[0101] 301. During the execution of the target task allocation plan, each robot to be scheduled is monitored based on the deep deterministic policy gradient DDPG algorithm to obtain the real-time status information of all robots to be scheduled.

[0102] It should be noted that in this embodiment, steps 301 to 307 are mainly implemented by the DDPG algorithm. Specifically, during the execution of the target task allocation scheme, each robot to be scheduled can be monitored based on the deep deterministic policy gradient DDPG algorithm to obtain the real-time status information of all robots to be scheduled. It should be noted that the real-time status information is used to characterize the real-time operating status of all robots to be scheduled.

[0103] In one specific embodiment, during the execution of the target task allocation scheme, corresponding to Figure 1 In step 103, according to the initial strategy provided by PSO, the system assigns tasks to each robot and sends specific instructions. After receiving the instructions, the robot starts to execute the task as planned. This step marks the official start of task scheduling, and each robot is clear about its work goals. Then, the multi-robot system needs to adjust the task allocation strategy in real time. Specifically, during the task execution process, the system will use the DDPG algorithm for real-time monitoring and adjustment. In the DDPG algorithm, the system will dynamically adjust the task allocation strategy based on the real-time data fed back during the robot's task execution. To understand the DDPG algorithm, the following is a detailed description.

[0104] The DDPG algorithm is a policy gradient-based reinforcement learning method suitable for problems in continuous action spaces. Compared with traditional Q learning, DDPG can handle complex high-dimensional states and action spaces, and has a higher convergence speed and stability. Specific method description: 1. Policy network and Q network: DDPG uses a policy network (Actor) to generate continuous actions, and a Q network (Critic) to evaluate the quality of these actions. The policy network is responsible for deciding what action should be taken at each step, while the Q network guides the adjustment of the policy network by evaluating the value of each action. 2. Target network: In order to stabilize the training process, DDPG introduces a target policy network and a target Q network. The parameters of these networks approach the parameters of the main network with a smaller step size each time they are updated, reducing the instability of training. 3. Experience replay: DDPG also uses an experience replay mechanism to randomly extract small batches of data from past experience to update the network. This reduces the correlation between training data and improves the learning effect of the algorithm. Using DDPG technology: It has the characteristics of adaptive task allocation and path optimization. Specifically, after using DDPG technology, the jacking AGV system can dynamically adjust task allocation and path planning through reinforcement learning. When performing tasks, AGV will not only consider the current cargo location and target location, but also make adaptive adjustments based on the real-time warehouse environment (such as the status of other AGVs, current traffic conditions, the presence of obstacles, etc.). For example, suppose an AGV is going to the storage area to pick up goods, and finds that another AGV is occupying its planned path on the way. At this time, the DDPG algorithm will allow the AGV to quickly evaluate multiple feasible paths and select the best alternative path. At the same time, the DDPG algorithm can also reallocate tasks according to the real-time system status, such as assigning another urgent task to an AGV that is closer. Therefore, through DDPG, the AGV system can continuously optimize its own behavior during task execution, making task allocation more flexible and efficient, and reducing delays caused by path conflicts or traffic congestion. In the end, the overall task completion efficiency of the system has been significantly improved, and resource utilization is also higher.

[0105] 302. Use the real-time status information as the status input of the DDPG algorithm to obtain a running state vector, and obtain the current task execution data according to the running state vector.

[0106] Based on step 301, the real-time data input link of the DDPG algorithm is entered. Specifically, the real-time state information is used as the state input of the DDPG algorithm to obtain the running state vector, and the current task execution data is obtained according to the running state vector.

[0107] In one of the specific embodiments, feedback data such as path congestion and insufficient power will be used as state input. Specifically, in the DDPG algorithm, the system will continuously obtain real-time status data from the feedback of the robot performing tasks. These data include but are not limited to: power level (detecting whether the robot's battery power is sufficient to complete the task), task progress (tracking the real-time status of task completion and whether it is completed within the scheduled time) and position coordinates (obtaining the precise position of the robot in real time to ensure that it has not deviated from the path). These state data will be used as state inputs of the DDPG algorithm. Then, DDPG mathematically models the current state of all AGVs to form a multi-dimensional state vector, namely the motion state vector. This motion state vector contains real-time information of all AGVs, such as current tasks, positions, power, etc.

[0108] 303. Evaluate the running state vector to obtain the target state action expectation of all robots to be scheduled to execute the target task allocation plan in the real-time running state.

[0109] After obtaining the motion state vector, it is necessary to evaluate the motion state vector to obtain the target state action expectation of all robots to be scheduled when executing the target task allocation scheme in real-time operation. It should be noted that the target state action expectation is used to characterize the expectation of different actions taken by all robots to be scheduled when executing the target task allocation scheme.

[0110] In one specific embodiment, the DDPG algorithm can use the Q function in value assessment. Specifically, DDPG uses a Q network to evaluate the value of each state and action combination. By evaluating the reward that the robot can get after taking a certain action in the current state, the target state action expectation is obtained. Among them, the Q network can help the system select high-value actions. High rewards usually mean shorter task completion time, less energy consumption, or a more efficient path.

[0111] 304. Obtain historical state action expectations, and determine target task evaluation data according to the target state action expectations and the historical state action expectations.

[0112] Therefore, in combination with step 303, it is also necessary to combine the historical state action expectations and determine the target task evaluation data based on the target state action expectations and the historical state action expectations. It should be noted that the historical state action is used to characterize the expectations of different actions taken by all robots to be scheduled when executing any task allocation scheme in history.

[0113] In one specific embodiment, the value network in DDPG predicts different actions that may be taken in the current state and estimates the cumulative rewards that can be obtained in the future under these actions. This evaluation is based on the current feedback data of the robot and combined with historical experience, that is, historical state action expectations, to make the evaluation more accurate and obtain the target task evaluation data.

[0114] 305. Generate a target task scheduling strategy based on the target task evaluation data.

[0115] Then, the target task scheduling strategy can be generated based on the target task evaluation data.

[0116] In one specific embodiment, the multi-robot system generates a new task scheduling strategy, namely, a target task scheduling strategy, based on the target task evaluation data by using the policy network in the DDPG algorithm.

[0117] 306. Generate a real-time task allocation instruction according to the target task scheduling strategy, and adjust one or more of the task priority, task planning path or task start and stop status in each robot to be scheduled corresponding to the target task allocation plan according to the real-time task allocation instruction to obtain the allocated target task allocation plan.

[0118] Based on step 305, the multi-robot system generates a real-time task allocation instruction according to the target task scheduling strategy by combining the DDPG algorithm. At the same time, according to the real-time task allocation instruction, one or more of the task priority, task planning path or task start-stop state in each robot to be scheduled corresponding to the target task allocation scheme is adjusted to obtain the assigned target task allocation scheme. It should be noted that the real-time task allocation instruction is used to instruct each robot to be scheduled to adjust one or more of the task priority, task planning path or task start-stop state in the target task allocation scheme.

[0119] In one of the specific embodiments, DDPG will first update the policy network and then control the robot to perform the action. Specifically, the policy network in DDPG generates real-time task allocation instructions, that is, new action instructions, based on the current state, such as adjusting the order of task execution, changing the path, or transferring the task to other more suitable robots to be scheduled AGV. Then, the multi-robot system will issue new instructions (real-time task allocation instructions) to the robot in real time based on the output of the policy network. This may include replanning the path, modifying task priorities, and even emergency stopping of a robot's task to avoid failures. These supplements can make the entire process clearer and more detailed, helping to better understand how to use this data, how the algorithm works, and how the task allocation strategy is dynamically adjusted. Ensure that the system is highly adaptable to environmental changes and ensure that the task can be completed smoothly.

[0120] 307. After all target task allocation plans are executed, the target task execution data of all target task allocation plans are recorded, and the target task execution data are analyzed to adjust the next task allocation plan according to the target task execution data to obtain the next task scheduling strategy.

[0121] Based on step 305, it should be added that after all target task allocation plans are executed, the target task execution data of all target task allocation plans can also be recorded, and the target task execution data can be analyzed to adjust the next task allocation plan according to the target task execution data to obtain the next task scheduling strategy. It should be noted that the next task allocation plan is the next task allocation plan for each robot to be scheduled, and the next task scheduling strategy is the next task scheduling strategy for each robot to be scheduled after adjusting the next task allocation plan.

[0122] In one of the specific embodiments, the DDPG algorithm can also perform continuous learning and optimization. At the same time, the execution data of the task allocation scheme during execution is recorded and analyzed. Specifically, DDPG not only adjusts tasks in real time, but also continuously learns and optimizes strategies during task execution. Through repeated learning and adjustment, DDPG can enable the system to gradually optimize the overall efficiency of task scheduling. For example, a multi-robot system can identify which paths are more prone to congestion and which tasks are more suitable for which type of robot, thereby avoiding inefficient decisions in future task scheduling. This process makes the system increasingly intelligent and efficient in dealing with complex tasks.

[0123] Furthermore, when all tasks are successfully completed, the multi-robot system will record detailed data on the entire task execution. This data includes the task completion time, the resources consumed by the robots, unexpected situations that occurred, and the adjustments made by the system during the process. By analyzing this data, the system can find optimization space and improve the strategy in the next task scheduling. In the long run, these records and analyses will significantly improve the overall performance of the system.

[0124] Through the task scheduling method of a robot disclosed in this embodiment, the whole process can be made clearer and more detailed, helping to better understand how to use these data, how the algorithm works, and how the task allocation strategy is dynamically adjusted. Ensure that the system is highly adaptable to environmental changes and ensure that the task can be completed smoothly. At the same time, DDPG not only adjusts tasks in real time, but also continuously learns and optimizes strategies during task execution. Through repeated learning and adjustment, DDPG can enable the system to gradually optimize the overall efficiency of task scheduling. By analyzing these data, the system can find optimization space and improve the strategy in the next task scheduling. In the long run, these records and analyses will significantly improve the overall performance of the system. Furthermore, this embodiment realizes dynamic task allocation and optimization in a multi-robot system by combining the DDPG reinforcement learning algorithm with the particle swarm optimization (PSO) algorithm. PSO is responsible for initially finding the global optimal task allocation solution, while reinforcement learning (such as the DDPG algorithm) is fine-tuned according to real-time data during task execution to ensure that the system can adapt to dynamically changing environments. This combination not only improves the efficiency of global search, but also enhances the adaptability and real-time optimization capabilities of the system.

[0125] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0126] See also Figure 4 , Figure 4 A schematic diagram of the structure of a robot task scheduling system disclosed in an embodiment of the present application.

[0127] An acquisition unit 401 is used to acquire current status information of each robot to be scheduled in the robot cluster;

[0128] A generating unit 402, configured to generate a target task allocation scheme corresponding to each robot to be scheduled according to the current state information;

[0129] The allocation unit 403 is used to generate a current task allocation instruction of the target task allocation scheme, and allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme;

[0130] The adjustment unit 404 is used to obtain the current task execution data of each robot to be scheduled in the target task allocation plan, and adjust the target task allocation plan of each robot to be scheduled according to the current task execution data to obtain the target task scheduling strategy, so that each scheduled robot executes the target task allocation plan after being assigned by the target task scheduling strategy.

[0131] Exemplarily, the system further includes: a control unit 405 and a determination unit 406;

[0132] A generating unit 402 is specifically used to generate an initial task allocation plan for each robot to be scheduled;

[0133] The acquisition unit 401 is specifically used to obtain an initial fitness value corresponding to the initial task allocation scheme according to the current state information; wherein the initial fitness value is used to represent the current state information and the task execution data when each robot to be scheduled executes the initial task allocation scheme;

[0134] A control unit 405 is used to control each robot to be scheduled to execute the initial task allocation plan to obtain a current fitness value;

[0135] The adjustment unit 404 is specifically used to compare the current fitness value with the initial fitness value, and adjust the initial task allocation scheme of the current robot to be scheduled based on the particle swarm optimization PSO algorithm and the comparison result to obtain the target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme; wherein the current robot to be scheduled is any robot to be scheduled that executes the initial task allocation scheme;

[0136] The determination unit 406 is configured to determine the adjusted initial task allocation scheme as the target task allocation scheme if the target fitness value is the optimal fitness value.

[0137] Exemplarily, the system further includes:

[0138] The acquisition unit 401 is further used to obtain the target fitness values ​​of all robots to be scheduled to execute the adjusted initial task allocation scheme;

[0139] The determination unit 406 is further used to compare all target fitness values, find the optimal fitness value among all target fitness values, and determine the task allocation scheme corresponding to the optimal fitness value; wherein the technical person to be scheduled corresponding to the optimal fitness value is the best scheduling robot;

[0140] The adjustment unit 404 is also used to adjust the initial task allocation plan of the effective robots to be scheduled according to the task allocation plan corresponding to the optimal fitness value, so as to obtain the target task allocation plan corresponding to each robot to be scheduled; wherein the effective robots to be scheduled are the other robots to be scheduled among all the robots to be scheduled except the best scheduling robot.

[0141] Exemplarily, the system includes:

[0142] The acquisition unit 401 is specifically used to monitor each robot to be scheduled based on the deep deterministic policy gradient DDPG algorithm during the execution of the target task allocation plan to obtain the real-time status information of all robots to be scheduled; wherein the real-time status information is used to characterize the real-time operating status of all robots to be scheduled;

[0143] The acquisition unit 401 is further used to use the real-time state information as the state input of the DDPG algorithm to obtain an operation state vector; wherein the operation state vector includes the operation states of all robots to be scheduled;

[0144] The acquisition unit 401 is further used to obtain current task execution data according to the running state vector.

[0145] Exemplarily, the system includes:

[0146] The acquisition unit 401 is specifically used to evaluate the running state vector to obtain the target state action expectation of all robots to be scheduled to execute the target task allocation scheme in the real-time running state; wherein the target state action expectation is used to characterize the expectation of different actions taken by all robots to be scheduled when executing the target task allocation scheme;

[0147] The acquisition unit 401 is further used to acquire the historical state action expectation, and determine the target task evaluation data according to the target state action expectation and the historical state action expectation; wherein the historical state action is used to characterize the expectation of different actions taken by all robots to be scheduled when executing any task allocation scheme in history;

[0148] The generating unit 402 is specifically configured to generate a target task scheduling strategy according to the target task evaluation data.

[0149] Exemplarily, the system further includes:

[0150] The generating unit 402 is further used to generate a real-time task allocation instruction according to the target task scheduling strategy; wherein the real-time task allocation instruction is used to instruct each robot to be scheduled to adjust one or more of the task priority, task planning path or task start and stop status of the target task allocation scheme;

[0151] The adjustment unit 404 is also used to adjust one or more of the task priority, task planning path or task start and stop status in each robot to be scheduled corresponding to the target task allocation plan according to the real-time task allocation instruction, so as to obtain the allocated target task allocation plan.

[0152] Exemplarily, the system further includes: a recording unit 407;

[0153] The recording unit 407 is used to record the target task execution data of all target task allocation schemes after all target task allocation schemes are executed;

[0154] The acquisition unit 401 is also used to analyze the target task execution data so as to adjust the next task allocation plan according to the target task execution data to obtain the next task scheduling strategy; wherein the next task allocation plan is the next task allocation plan for each robot to be scheduled, and the next task scheduling strategy is the next task scheduling strategy for each robot to be scheduled after adjusting the next task allocation plan.

[0155] See below Figure 5 , a schematic diagram of a robot task scheduling device disclosed in an embodiment of the present application includes:

[0156] CPU 501, memory 505, input / output interface 504, wired or wireless network interface 503 and power supply 502;

[0157] The memory 505 is a temporary storage memory or a permanent storage memory;

[0158] The CPU 501 is configured to communicate with the memory 505 and execute the instructions in the memory 505 to perform the aforementioned Figures 1 to 3 A method for scheduling tasks for a robot in any of the illustrated embodiments.

[0159] The embodiment of the present application also provides a chip system, the chip system includes at least one processor and a communication interface, the communication interface and the at least one processor are interconnected through a line, and the at least one processor is used to run a computer program or instruction to execute the aforementioned Figures 1 to 3 A method for scheduling tasks for a robot in any of the illustrated embodiments.

[0160] The embodiment of the present application also provides a computer-readable storage medium, the computer-readable storage medium includes instructions, when the instructions are executed on a computer, the computer executes the aforementioned Figures 1 to 3 A method for scheduling tasks for a robot in any of the illustrated embodiments.

[0161] The present application also provides a computer program product including instructions, which, when executed on a computer, enables the computer to execute the aforementioned Figures 1 to 3 A method for scheduling tasks for a robot in any of the illustrated embodiments.

[0162] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0163] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be an indirect coupling or communication connection through some interfaces, devices or units, which can be electrical, mechanical or other forms.

[0164] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0165] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0166] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application is essentially or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, read-only memory), random access memory (RAM, random access memory), disk or optical disk and other media that can store program code.

Claims

1. A robot task scheduling method, characterized in that: The method comprises: Obtain the current status information of each robot to be scheduled in the robot cluster; Generating a target task allocation scheme corresponding to each robot to be scheduled according to the current state information; Generate a current task allocation instruction of the target task allocation scheme, and allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme; Obtain the current task execution data of each robot to be scheduled in executing the target task allocation scheme, and adjust the target task allocation scheme of each robot to be scheduled according to the current task execution data to obtain the target task scheduling strategy, so that each scheduled robot executes the target task allocation scheme after being assigned by the target task scheduling strategy.

2. The robot task scheduling method according to claim 1, characterized in that: The step of generating a target task allocation scheme corresponding to each robot to be scheduled according to the current state information includes: For each of the robots to be scheduled, generating an initial task allocation plan; According to the current state information, an initial fitness value corresponding to the initial task allocation scheme is obtained; wherein the initial fitness value is used to characterize the current state information and the task execution data when each robot to be scheduled executes the initial task allocation scheme; Control each of the robots to be scheduled to execute the initial task allocation plan to obtain a current fitness value; Compare the current fitness value with the initial fitness value, and adjust the initial task allocation scheme of the current robot to be scheduled based on the particle swarm optimization PSO algorithm and the comparison result to obtain a target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme; wherein the current robot to be scheduled is any robot to be scheduled that executes the initial task allocation scheme; If the target fitness value is the optimal fitness value, the adjusted initial task allocation scheme is determined as the target task allocation scheme.

3. The robot task scheduling method according to claim 2, characterized in that: After obtaining the target fitness value of the current robot to be scheduled to execute the adjusted initial task allocation scheme, the method further includes: Obtaining the target fitness value of all robots to be scheduled to execute the adjusted initial task allocation plan; Compare all target fitness values, find the optimal fitness value among all target fitness values, and determine the task allocation scheme corresponding to the optimal fitness value; wherein the technical person to be scheduled corresponding to the optimal fitness value is the best scheduling robot; According to the task allocation scheme corresponding to the optimal fitness value, the initial task allocation scheme of the valid robots to be scheduled is adjusted to obtain the target task allocation scheme corresponding to each robot to be scheduled; wherein the valid robots to be scheduled are the other robots to be scheduled among all the robots to be scheduled except the optimal scheduling robot.

4. The robot task scheduling method according to claim 1, characterized in that: The obtaining of the current task execution data of each robot to be scheduled in executing the target task allocation scheme includes: During the execution of the target task allocation scheme, each robot to be scheduled is monitored based on the deep deterministic policy gradient DDPG algorithm to obtain real-time status information of all robots to be scheduled; wherein the real-time status information is used to characterize the real-time operating status of all robots to be scheduled; Using the real-time status information as the state input of the DDPG algorithm to obtain an operation state vector; wherein the operation state vector includes the operation states of all the robots to be scheduled; The current task execution data is obtained according to the running state vector.

5. The robot task scheduling method according to claim 4, characterized in that: The step of adjusting the target task allocation scheme of each robot to be scheduled according to the current task execution data to obtain a target task scheduling strategy includes: The running state vector is evaluated to obtain the target state action expectation of all the robots to be scheduled to execute the target task allocation scheme in the real-time running state; wherein the target state action expectation is used to characterize the expectation of different actions taken by all the robots to be scheduled when executing the target task allocation scheme; Obtaining historical state action expectations, and determining target task evaluation data according to the target state action expectations and the historical state action expectations; wherein the historical state actions are used to characterize the expectations of different actions taken by all robots to be scheduled when executing any task allocation scheme in history; The target task scheduling strategy is generated according to the target task evaluation data.

6. The robot task scheduling method according to claim 1, characterized in that: After obtaining the target task scheduling strategy, the method further includes: Generate a real-time task allocation instruction according to the target task scheduling strategy; wherein the real-time task allocation instruction is used to instruct each robot to be scheduled to adjust one or more of the task priority, task planning path or task start and stop status of the target task allocation scheme; According to the real-time task allocation instruction, one or more of the task priority, the task planning path or the task start / stop state in each robot to be scheduled corresponding to the target task allocation scheme is adjusted to obtain the allocated target task allocation scheme.

7. The robot task scheduling method according to claim 1, characterized in that: After obtaining the target task scheduling strategy, the method further includes: When all target task allocation plans are executed, target task execution data of all target task allocation plans are recorded; Analyze the target task execution data to adjust the next task allocation plan according to the target task execution data to obtain the next task scheduling strategy; wherein the next task allocation plan is the next task allocation plan of each robot to be scheduled, and the next task scheduling strategy is the next task scheduling strategy of each robot to be scheduled after adjusting the next task allocation plan.

8. A robot task scheduling system, characterized in that: The system comprises: An acquisition unit, used to acquire current status information of each robot to be scheduled in the robot cluster; A generating unit, configured to generate a target task allocation scheme corresponding to each of the robots to be scheduled according to the current state information; an allocation unit, configured to generate a current task allocation instruction of the target task allocation scheme, and allocate the target task allocation scheme to each corresponding robot to be scheduled according to the current task allocation instruction, so as to control each robot to be scheduled to execute the target task allocation scheme; An adjustment unit is used to obtain the current task execution data of each robot to be scheduled in executing the target task allocation plan, and adjust the target task allocation plan of each robot to be scheduled according to the current task execution data to obtain a target task scheduling strategy so that each scheduled robot executes the target task allocation plan after being assigned by the target task scheduling strategy.

9. A robot task scheduling device, characterized in that: The device comprises: CPU, memory, input and output interfaces, wired or wireless network interfaces, and power supply; The memory is a short-term storage memory or a persistent storage memory; The central processing unit is configured to communicate with the memory and execute instruction operations in the memory to perform the robot task scheduling method described in any one of claims 1 to 7.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes instructions, and when the instructions are executed on a computer, the computer executes the robot task scheduling method according to any one of claims 1 to 7.

Images