Resource scheduling method and multi-robot system

Abstract

Embodiments of the present application provide a resource scheduling method and a multi-robot system. The resource scheduling method comprises: constructing a unified scheduling model based on the current running state of each robot node in the multi-robot system; performing unified scheduling on the resource scheduling behavior in the multi-robot system based on the unified scheduling model to obtain a global resource collaborative scheduling result; determining an initial resource scheduling strategy corresponding to each robot node based on the global resource collaborative scheduling result and the current running state of each robot node; dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future state prediction result of each robot node to obtain a target resource scheduling strategy corresponding to each robot node; and controlling the corresponding robot node to perform the resource scheduling behavior based on the target resource scheduling strategy, which can improve the resource collaborative capability, task execution continuity and overall running efficiency of the multi-robot system.

Description

Technical Field

[0001] This application relates to the field of multi-robot technology, and more particularly to a resource scheduling method and a multi-robot system. Background Technology

[0002] In recent years, with the development of robotics, edge intelligence, and agricultural automation, multi-robot systems have been widely used in smart agriculture, unmanned farms, and unmanned inspection scenarios because they can support collaborative operations, dynamic path planning, real-time environmental perception, and online intelligent decision-making.

[0003] However, taking multi-robot systems as an example, energy management, computing resource allocation, task scheduling, and path planning typically require separate modeling and control, with different modules making independent decisions based on remaining power, current computing load, task priority, or path cost. Due to the lack of a unified modeling and collaborative optimization mechanism for energy status, computing resources, task requirements, and path costs, effective linkage between energy replenishment, computing scheduling, task execution, and path planning is difficult to achieve, easily leading to low resource utilization and overall low operating efficiency. Summary of the Invention

[0004] The resource scheduling method and multi-robot system provided in this application are intended to address the shortcomings of existing resource scheduling methods, such as low resource utilization and low overall operating efficiency.

[0005] In a first aspect, embodiments of this application provide a resource scheduling method and a multi-robot system. The resource scheduling method is applied to a multi-robot system and includes: constructing a unified scheduling model based on the current operating state of each robot node in the multi-robot system; performing unified scheduling of resource scheduling behaviors in the multi-robot system based on the unified scheduling model to obtain a global resource collaborative scheduling result; determining an initial resource scheduling strategy corresponding to each robot node based on the global resource collaborative scheduling result and the current operating state of each robot node; dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future state prediction result of each robot node to obtain a target resource scheduling strategy corresponding to each robot node; and controlling the corresponding robot node to perform resource scheduling behaviors based on the target resource scheduling strategy.

[0006] In some embodiments, constructing a unified scheduling model based on the current operating state of each robot node in the multi-robot system includes: obtaining the energy state, computing power state, task state, and path state of each robot node; determining the resource constraints corresponding to each robot node and the collaborative scheduling relationships among all robot nodes based on the energy state, computing power state, task state, and path state; and constructing the unified scheduling model based on the resource constraints and the collaborative scheduling relationships.

[0007] In some embodiments, the step of uniformly scheduling resource scheduling behaviors in the multi-robot system based on the unified scheduling model to obtain a global resource collaborative scheduling result includes: performing associated scheduling on task execution behaviors, computing power allocation behaviors, energy replenishment behaviors, and path planning behaviors in the multi-robot system based on the unified scheduling model to obtain resource association scheduling results; determining resource occupancy information and execution constraint information corresponding to each resource scheduling behavior based on the resource occupancy information and the execution constraint information; determining the scheduling priority corresponding to each resource scheduling behavior based on the scheduling priority; and performing collaborative sorting and node allocation on each resource scheduling behavior based on the scheduling priority to obtain the global resource collaborative scheduling result.

[0008] In some embodiments, determining the initial resource scheduling strategy for each robot node based on the global resource collaborative scheduling result and the current operating state of each robot node includes: determining the node scheduling content for each robot node from the global resource collaborative scheduling result; determining whether the node scheduling content meets the execution conditions of the corresponding robot node based on the current operating state of each robot node; if the node scheduling content meets the execution conditions of the corresponding robot node, determining the initial resource scheduling strategy for each robot node based on the node scheduling content; if the node scheduling content does not meet the execution conditions of the corresponding robot node, correcting the node scheduling content, and determining the initial resource scheduling strategy for each robot node based on the corrected node scheduling content.

[0009] In some embodiments, the step of dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future state prediction result of each robot node to obtain the target resource scheduling strategy corresponding to each robot node includes: determining the future state prediction result of each robot node based on the historical and current operating states of each robot node; determining the future resource change state of each robot node when executing the initial resource scheduling strategy based on the future state prediction result; and dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future resource change state to obtain the target resource scheduling strategy corresponding to each robot node.

[0010] In some embodiments, the step of dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future resource change state to obtain the target resource scheduling strategy corresponding to each robot node includes: determining the energy change trend corresponding to each robot node based on the future resource change state; determining the computing power adjustment method corresponding to each robot node based on the energy change trend; and adjusting the computing power allocation content in the initial resource scheduling strategy based on the computing power adjustment method to obtain the target resource scheduling strategy.

[0011] In some embodiments, the step of dynamically adjusting the initial resource scheduling strategy corresponding to each robot node based on the future resource change state to obtain the target resource scheduling strategy corresponding to each robot node further includes: determining the task change trend and path change trend corresponding to each robot node based on the future resource change state; determining the task path collaborative adjustment method corresponding to each robot node based on the task change trend and the path change trend; and adjusting the task execution content and path execution content in the initial resource scheduling strategy based on the task path collaborative adjustment method to obtain the target resource scheduling strategy.

[0012] In some embodiments, controlling the corresponding robot node to perform resource scheduling behavior based on the target resource scheduling strategy includes: generating a scheduling control instruction for each robot node based on the target resource scheduling strategy; sending the scheduling control instruction to the corresponding robot node to control the corresponding robot node to perform resource scheduling behavior; obtaining execution feedback information after the corresponding robot node performs the resource scheduling behavior; and updating the parameters in the unified scheduling model based on the deviation information between the execution feedback information and the expected execution result corresponding to the target resource scheduling strategy.

[0013] In some embodiments, the method further includes: determining task coordination relationships and resource coordination relationships among multiple robot nodes based on the target resource scheduling strategy corresponding to each robot node; modifying the target resource scheduling strategy corresponding to at least one robot node based on the task coordination relationship and the resource coordination relationship to obtain a modified target resource scheduling strategy; and controlling the corresponding robot node to perform resource scheduling behavior based on the modified target resource scheduling strategy.

[0014] Secondly, embodiments of this application provide a multi-robot system, including: multiple robot nodes, a processor, and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are invoked by the processor, the processor executes the method described above.

[0015] The beneficial effects of the resource scheduling method provided in this application are as follows: On the one hand, by uniformly scheduling the resource scheduling behavior in a multi-robot system, a global resource collaborative scheduling result oriented towards the overall operating state can be obtained, reducing the impact of the fragmentation of different resource scheduling links; on the other hand, by combining the global resource collaborative scheduling result and the current operating state of all robot nodes to determine the initial resource scheduling strategy, and dynamically adjusting it according to the future state prediction result, the resource scheduling process can adapt to both the current operating situation and future state changes, thereby improving the resource collaboration capability, task execution continuity and overall operating efficiency of the multi-robot system. Attached Figure Description

[0016] One or more embodiments are illustrated by way of example with reference numerals in the accompanying drawings. These illustrations do not constitute a limitation on the embodiments. Elements with the same reference numerals in the drawings are denoted as similar elements. Unless otherwise stated, the figures in the drawings are not to be limited by scale.

[0017] Figure 1 A schematic diagram of the resource scheduling method provided in the embodiments of this application. Figure 1 ; Figure 2 A schematic diagram of a sub-process of step S110 of the resource scheduling method provided in the embodiments of this application; Figure 3 A schematic diagram of a sub-process of step S120 of the resource scheduling method provided in the embodiments of this application; Figure 4 A schematic diagram of a sub-process of step S130 of the resource scheduling method provided in the embodiments of this application; Figure 5 A schematic diagram of a sub-process of step S140 of the resource scheduling method provided in the embodiments of this application; Figure 6A schematic diagram of the sub-process of step S143 of the resource scheduling method provided in the embodiments of this application. Figure 1 ; Figure 7 A schematic diagram of the sub-process of step S143 of the resource scheduling method provided in the embodiments of this application. Figure 2 ; Figure 8 A schematic diagram of a sub-process of step S150 of the resource scheduling method provided in the embodiments of this application; Figure 9 A schematic diagram of the resource scheduling method provided in the embodiments of this application. Figure 2 ; Figure 10 This is a schematic diagram of the architecture of the resource scheduling device provided in the embodiments of this application; Figure 11 This is a schematic diagram of the structure of a multi-robot system provided in an embodiment of this application. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are for illustrative purposes only and are not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.

[0019] It should be noted that, unless there is a conflict, the various features in the embodiments of this application can be combined with each other, all of which are within the protection scope of this application. Furthermore, although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in a different order than the module division in the device or the order in the flowchart. Moreover, the terms "first," "second," and "third" used in this application do not limit the data or execution order, but only distinguish identical or similar items with essentially the same function and effect.

[0020] The technical solution of this application is described in detail below. The technical solution of this application can be applied to agricultural robots, unmanned farms or multi-device collaborative platforms to achieve functions such as autonomous operation, real-time perception, dynamic scheduling and online optimization.

[0021] Figure 1 This is a schematic diagram illustrating a resource scheduling method for a multi-robot system provided in an embodiment of this application. Figure 1 As shown, the resource scheduling method includes: S110. Based on the current operating state of each robot node in the multi-robot system, construct a unified scheduling model; Among them, "multi-robot system" refers to a system in which multiple robot nodes work together. This multi-robot system can be applied to agricultural operations, unmanned inspection, industrial handling, warehousing and logistics, or other scenarios that require multiple robots to perform tasks collaboratively.

[0022] A "robot node" refers to an individual robot in a multi-robot system that can independently perform tasks, move, compute, or schedule resources. Robot nodes can include agricultural robots, inspection robots, mobile operation robots, unmanned vehicles, or other mobile devices with autonomous operation capabilities.

[0023] "Current operating status" refers to information used to characterize the current operating status of a robot node. The current operating status can include the robot node's energy status, computing power status, task status, and path status. Specifically, energy status can include remaining battery power, charging requirements, and energy replenishment status; computing power status can include current computing load, available computing power, or computing task occupancy; task status can include pending tasks, task priorities, and task completion deadlines; and path status can include current location, target location, and feasible paths.

[0024] The "unified scheduling model" refers to a model used to uniformly describe and process the current operating status of multiple robot nodes.

[0025] In some embodiments, the multi-robot system acquires the current operating status of each robot node, such as the remaining power, current computing load, tasks to be executed, current position, and target position. Then, it performs a structured representation of different types of current operating status to form a unified scheduling model that can characterize the overall operating status of the multi-robot system. This avoids the low resource utilization and low overall operating efficiency caused by energy management, computing resource allocation, task scheduling, and path planning being handled independently by different modules.

[0026] S120. Based on the unified scheduling model, the resource scheduling behavior of the multi-robot system is uniformly scheduled to obtain the global resource collaborative scheduling result. "Resource scheduling behavior" refers to the behaviors involving resource use, allocation, or adjustment during the operation of a multi-robot system. Resource scheduling behavior can include task execution behavior, computing power allocation behavior, energy replenishment behavior, and path planning behavior.

[0027] "Global resource collaborative scheduling result" refers to the resource scheduling result obtained based on the overall operating status of the multi-robot system. This global resource collaborative scheduling result can be used to characterize the resource allocation relationship, task collaboration relationship, energy supply relationship, or path execution relationship between multiple robot nodes.

[0028] In some embodiments, the multi-robot system, based on a unified scheduling model, performs a unified analysis of the energy status, computing power status, task status, and path status of each robot node. It also coordinates resource scheduling behaviors such as energy replenishment, computing resource allocation, task execution arrangement, and path selection to obtain a global resource collaborative scheduling result that improves the overall operational efficiency of the multi-robot system. For example, when the first robot node has low remaining power and the energy consumption of the task to be executed is high, while the second robot node has high remaining power and is closer to the target work area, the task execution node can be adjusted globally to reduce the risk of task interruption and path energy consumption. As another example, when the computing power load of a certain robot node is high, while other robot nodes have available computing power, computing resources can be coordinated among multiple robot nodes to improve the overall computing power utilization efficiency.

[0029] S130. Based on the global resource collaborative scheduling results and the current operating status of each robot node, determine the initial resource scheduling strategy corresponding to each robot node; The "initial resource scheduling strategy" refers to the preliminary scheduling arrangement determined based on the global resource collaborative scheduling results and the current operating status of the robot nodes, which guides the corresponding robot nodes to perform resource scheduling behavior.

[0030] In some embodiments, the multi-robot system determines the overall resource coordination relationship among all robot nodes in the system based on the global resource coordination scheduling results. Then, it generates corresponding initial resource scheduling strategies for different robot nodes by combining the current operating state of each robot node. For example, for robot nodes with high remaining power and sufficient available computing power, an initial resource scheduling strategy can be generated that prioritizes high-energy-consuming or high-computing-power tasks; for robot nodes with low remaining power or close to the charging location, an initial resource scheduling strategy can be generated that prioritizes energy replenishment or reduces the computing load.

[0031] S140. Based on the future state prediction results of each robot node, the initial resource scheduling strategy corresponding to each robot node is dynamically adjusted to obtain the target resource scheduling strategy corresponding to each robot node. Among them, "future state prediction results" refers to the results obtained by predicting the possible changes in the operating state of robot nodes in the future.

[0032] The "target resource scheduling strategy" refers to the scheduling strategy obtained by dynamically adjusting the initial resource scheduling strategy, which is used to actually control the robot nodes to perform resource scheduling behavior. Compared with the initial resource scheduling strategy, the target resource scheduling strategy further considers the impact of future state changes of the robot nodes on resource scheduling behavior.

[0033] In some embodiments, a multi-robot system can predict the energy status, computing power status, task status, and path status of robot nodes over a future period based on historical operating data, current operating status, and information on changes in the external environment. For example, if it is predicted that a robot node will experience insufficient energy supply or an increased task load in the future, unnecessary computing load on that robot node can be reduced in advance, or its energy replenishment sequence can be adjusted; if it is predicted that a robot node will have a long path to move in the future, some computing tasks can be rescheduled to be executed during the movement; if it is predicted that the risk of task completion for a robot node will increase, some tasks can be rescheduled to be executed on other robot nodes.

[0034] S150. Based on the target resource scheduling strategy, control the corresponding robot node to perform resource scheduling behavior.

[0035] In some embodiments, the multi-robot system, based on a target resource scheduling strategy, controls some robot nodes to prioritize high-priority tasks, moves some robot nodes to energy replenishment locations, reduces the computational load of some robot nodes, or performs computational tasks while moving. During execution, the multi-robot system can also continue to acquire the current operating status of all robot nodes and update the subsequent resource scheduling process based on changes in operating status.

[0036] In this embodiment, on the one hand, by uniformly scheduling the resource scheduling behavior in the multi-robot system, a global resource collaborative scheduling result oriented towards the overall operating state can be obtained, reducing the impact of the fragmentation of different resource scheduling links; on the other hand, by combining the global resource collaborative scheduling result and the current operating state of all robot nodes to determine the initial resource scheduling strategy, and making dynamic adjustments based on the future state prediction result, the resource scheduling process can adapt to both the current operating situation and future state changes, thereby improving the resource collaboration capability, task execution continuity and overall operating efficiency of the multi-robot system.

[0037] In some embodiments, such as Figure 2 As shown, S110 specifically includes: S111. Obtain the energy status, computing power status, task status, and path status of each robot node; Among them, "energy status" refers to information used to characterize the current energy supply and energy consumption of robot nodes.

[0038] "Computing power status" refers to information used to characterize the current computing resource usage of a robot node.

[0039] "Task status" refers to information used to characterize the current task execution status of a robot node.

[0040] "Path status" refers to information used to characterize the movement position of robot nodes and the execution status of the path.

[0041] In some embodiments, the multi-robot system can obtain the energy status, computing power status, task status, and path status of each robot node through the battery management unit, computing resource monitoring unit, task management unit, and positioning unit of each robot node, thereby forming the basic data for building a unified scheduling model.

[0042] S112. Based on the energy status, computing power status, task status, and path status, determine the resource constraints for each robot node and the collaborative scheduling relationships among all robot nodes. "Resource constraints" refer to the constraints formed by the robot node's own operating state, used to limit resource scheduling behavior. Resource constraints can include energy constraints, computing power constraints, task constraints, and path constraints.

[0043] "Cooperative scheduling relationship" refers to the cooperative relationship formed between multiple robot nodes during resource scheduling. Cooperative scheduling relationships can include task allocation relationships, computing resource coordination relationships, energy supply coordination relationships, and path coordination relationships.

[0044] In some embodiments, the multi-robot system determines the available capabilities and limitations of each robot node in terms of energy, computing power, task, and path based on its energy status, computing power status, task status, and path status, and establishes corresponding resource constraints. Simultaneously, the multi-robot system further compares the remaining power, available computing power, task load, and location differences among different robot nodes to determine relationships enabling task collaboration, computing power collaboration, energy collaboration, or path collaboration among multiple robot nodes, thereby obtaining the collaborative scheduling relationships among all robot nodes.

[0045] S113. Based on the resource constraint relationship and the cooperative scheduling relationship, construct the unified scheduling model.

[0046] In some embodiments, the multi-robot system uses the resource constraints corresponding to all robot nodes as model constraints and the collaborative scheduling relationships between all robot nodes as node associations in the model to construct a unified scheduling model, enabling subsequent resource scheduling behaviors to be uniformly scheduled globally.

[0047] In this embodiment, by constructing a unified scheduling model based on resource constraints and collaborative scheduling relationships, a unified model foundation can be provided for resource scheduling behavior in multi-robot systems, thereby helping to improve the coordination of resource scheduling and overall operational efficiency.

[0048] In some embodiments, such as Figure 3 As shown, S120 specifically includes: S121. Based on the unified scheduling model, the task execution behavior, computing power allocation behavior, energy supply behavior and path planning behavior of the multi-robot system are correlated and scheduled to obtain resource correlation scheduling results. "Task execution behavior" refers to the behavior of robot nodes in performing corresponding operations or processing tasks according to task requirements. Task execution behavior can include agricultural operation task execution, inspection task execution, identification task execution, material handling task execution, or other task execution processes completed by robot nodes.

[0049] "Computing power allocation behavior" refers to the behavior of allocating, adjusting, or utilizing computing resources in robot nodes or multi-robot systems. Computing power allocation behavior can include assigning computing tasks to robot nodes with available computing power, adjusting the computing load of robot nodes, or coordinating the use of computing resources among multiple robot nodes.

[0050] "Energy replenishment behavior" refers to the energy replenishment behavior performed by robot nodes to maintain continuous operation. Energy replenishment behavior can include moving to a charging location, waiting to charge, performing charging, or adjusting the charging sequence according to the energy status.

[0051] "Path planning behavior" refers to the behavior of determining a movement path based on the current position of a robot node, the target position, the path cost, and the task requirements. Path planning behavior can include determining the path to the work area, the path to the energy replenishment location, or coordinating movement routes among multiple robot nodes.

[0052] "Resource association scheduling result" refers to the scheduling result obtained after associating and scheduling task execution behavior, computing power allocation behavior, energy supply behavior, and path planning behavior. This resource association scheduling result can characterize the matching relationship between different resource scheduling behaviors, such as the matching relationship between tasks and robot nodes, the matching relationship between computing tasks and available computing power, the matching relationship between energy supply behavior and task execution timing, and the matching relationship between path planning behavior and task execution location.

[0053] In some embodiments, the multi-robot system, based on a unified scheduling model, simultaneously analyzes the relationships between task execution behavior, computing power allocation behavior, energy replenishment behavior, and path planning behavior to obtain resource association scheduling results that reflect the collaborative relationships between multiple types of resource scheduling behaviors. For example, when a robot node needs to execute a high-priority task with high computational requirements, its available computing power and remaining power can be combined to determine whether the robot node should execute the task or whether to allocate some of the computational tasks to other robot nodes. As another example, when a robot node has low remaining power and the target work area is far away, energy replenishment behavior and path planning behavior can be combined to determine whether the robot node should replenish its energy before executing the task.

[0054] S122. Based on the resource association scheduling results, determine the resource occupancy information and execution constraint information corresponding to each resource scheduling behavior; "Resource occupancy information" refers to the resource information required or consumed during the execution of resource scheduling actions. This information may include power consumption, computing power usage, execution time, path distance, path cost, and robot node occupancy status.

[0055] "Execution constraint information" refers to the restrictions that resource scheduling behavior must meet during execution. Execution constraint information may include task completion time limits, energy minimums, computing power capacity, and node location limitations.

[0056] In some embodiments, the multi-robot system analyzes task execution behavior, computing power allocation behavior, energy replenishment behavior, and path planning behavior based on resource association scheduling results, determines the resource occupancy information corresponding to each resource scheduling behavior in terms of power, computing power, time, path, and node occupancy, and further determines the execution constraint information corresponding to each resource scheduling behavior in terms of task time limit, energy lower limit, computing power capacity, and path reachability.

[0057] S123. Based on the resource occupancy information and the execution constraint information, determine the scheduling priority corresponding to each resource scheduling behavior; "Scheduling priority" refers to information used to characterize the degree to which resource scheduling behavior is prioritized or resources are allocated in a multi-robot system.

[0058] In some embodiments, a multi-robot system can determine the degree of power, computing power, time, or path resource consumption of resource scheduling behavior based on resource occupancy information, and determine whether the resource scheduling behavior has strong time constraints, energy constraints, computing power constraints, or path constraints based on execution constraint information. For resource scheduling behaviors with short task completion deadlines, high energy risks, or significant impacts on the overall system operation, higher scheduling priorities can be assigned; for resource scheduling behaviors with low resource consumption, ample execution time, or those that can be postponed, lower scheduling priorities can be assigned.

[0059] S124. Based on the scheduling priority, each resource scheduling action is collaboratively sorted and node is allocated to obtain the global resource collaborative scheduling result.

[0060] "Collaborative sorting" refers to the process of determining the execution order of multiple resource scheduling actions based on their scheduling priorities, resource occupancy relationships, and execution constraints. Collaborative sorting is not limited to a single robot node or a single resource type; rather, it involves the overall sorting of tasks, computing power allocation, energy replenishment, and path planning across a multi-robot system.

[0061] "Node allocation" refers to the process of assigning resource scheduling actions to corresponding robot nodes for execution.

[0062] In some embodiments, the multi-robot system coordinates the prioritization of task execution, computing power allocation, energy replenishment, and path planning behaviors based on their respective scheduling priorities. It then allocates different resource scheduling behaviors to corresponding robot nodes based on the operating status of each robot node, thereby obtaining a global resource coordination scheduling result that simultaneously reflects the execution order and node allocation relationships. For example, high-priority task execution behaviors with high energy consumption can be preferentially allocated to robot nodes with higher remaining power and lower path costs; computing tasks with high computing power consumption can be preferentially allocated to robot nodes with higher available computing power; and energy replenishment behaviors can be prioritized for robot nodes with low remaining power.

[0063] In this embodiment, on the one hand, by coordinating the scheduling of task execution, computing power allocation, energy replenishment, and path planning behaviors based on a unified scheduling model, the originally independent resource scheduling behaviors in a multi-robot system can be incorporated into the same scheduling process, enabling a collaborative relationship between different resource scheduling behaviors. On the other hand, by determining the scheduling priority based on resource occupancy information and execution constraint information, the execution order of resource scheduling behaviors can be made more in line with the overall system operation requirements. Based on this, collaborative sorting and node allocation based on scheduling priority can coordinate the execution order and execution nodes of different resource scheduling behaviors within the multi-robot system, thereby improving the overall system operating efficiency.

[0064] In some embodiments, such as Figure 4 As shown, S130 specifically includes: S131. Determine the node scheduling content corresponding to each robot node from the global resource collaborative scheduling results; Among them, "node scheduling content" refers to the scheduling content corresponding to a single robot node, which is decomposed from the global resource collaborative scheduling results.

[0065] In some embodiments, after obtaining the global resource collaborative scheduling result, the multi-robot system can decompose the global resource collaborative scheduling result into multiple node scheduling contents according to the task allocation relationship, resource occupation relationship, or execution order of the robot nodes. For example, the global resource collaborative scheduling result may include the first robot node performing the first task, the second robot node performing the second task, and the third robot node performing energy replenishment, etc.

[0066] S132. Based on the current running state of each robot node, determine whether the node scheduling content meets the execution conditions of the corresponding robot node; "Execution conditions" refer to the conditions that a robot node must meet when executing the scheduled content of its corresponding node. Execution conditions can include energy conditions, computing power conditions, task conditions, and path conditions.

[0067] In some embodiments, the multi-robot system matches the node scheduling content of each robot node with the current operating status of that robot node. For example, when the node scheduling content includes high-energy-consuming tasks, the system can determine whether the remaining power of the robot node meets the task execution requirements based on its energy status; when the node scheduling content includes high-computing-power tasks, the system can determine whether the available computing power of the robot node meets the computing requirements based on its computing power status.

[0068] S133. If the node scheduling content satisfies the execution conditions of the corresponding robot node, determine the initial resource scheduling strategy for each robot node based on the node scheduling content. In some embodiments, when the multi-robot system determines that the remaining power, available computing power, and path reachability of the first robot node all meet the execution requirements of its node scheduling content, the node scheduling content can be transformed into the initial resource scheduling strategy of the first robot node to guide the robot node to subsequently execute tasks, use computing resources, replenish energy, or move along the execution path.

[0069] S134. If the node scheduling content does not meet the execution conditions of the corresponding robot node, the node scheduling content is modified, and the initial resource scheduling strategy corresponding to each robot node is determined based on the modified node scheduling content.

[0070] In some embodiments, when a multi-robot system determines that the node scheduling content of a certain robot node does not meet the corresponding execution conditions, the node scheduling content can be modified according to the reason for not meeting the execution conditions. For example, when a robot node has insufficient remaining power to execute the corresponding task, the node scheduling content can be modified to replenish energy first and then execute the task, or some tasks can be transferred to other robot nodes; when a robot node has insufficient available computing power to complete the corresponding computational task, the computational load of the robot node can be reduced, or some computational tasks can be transferred to other robot nodes; when the path state of a robot node does not meet the execution requirements, the path planning behavior can be adjusted or the execution node can be replaced.

[0071] After the correction is completed, the multi-robot system can determine the initial resource scheduling strategy for the corresponding robot node based on the corrected node scheduling content, so that the initial resource scheduling strategy matches the actual operating state of the robot node.

[0072] In this embodiment, by determining whether the node scheduling content meets the corresponding execution conditions based on the current running state of each robot node, it is possible to avoid directly issuing scheduling content that is not suitable for the current node state for execution. When the node scheduling content meets the execution conditions, the initial resource scheduling strategy is determined. When the node scheduling content does not meet the execution conditions, it is corrected before determining the initial resource scheduling strategy. This makes the initial resource scheduling strategy corresponding to each robot node more in line with the actual running state, thereby improving the executability of the resource scheduling process.

[0073] In some embodiments, such as Figure 5 As shown, S140 specifically includes: S141. Based on the historical and current operating states of each robot node, determine the future state prediction result for each robot node; Among them, "historical operating status" refers to the state information generated by the robot node during its operation before the current moment.

[0074] "Current running status" refers to the running status of a robot node at the current moment.

[0075] In some embodiments, a multi-robot system can predict the energy, computing power, task, and path changes of a robot node in the future based on its historical power consumption, historical computing load changes, historical task completion, and historical path movement, combined with its current remaining power, current computing load, current pending tasks, and current location, thereby obtaining the corresponding future state prediction results.

[0076] S142. Based on the future state prediction results, determine the future resource change state of each robot node when executing the initial resource scheduling strategy; Among them, "future resource change status" refers to the resource changes that may occur in the future when the robot node executes resource scheduling behavior according to the initial resource scheduling strategy.

[0077] In some embodiments, a multi-robot system can match and analyze the initial resource scheduling strategy of each robot node with the corresponding future state prediction results to determine the possible resource changes that may occur after the robot node executes the initial resource scheduling strategy. For example, when the initial resource scheduling strategy arranges for a robot node to continuously execute multiple high-energy-consuming tasks, the future state prediction results can be used to determine whether the robot node will run out of power; when the initial resource scheduling strategy arranges for a robot node to execute a high-computing-power task, the future state prediction results can be used to determine whether the robot node will experience excessive computing power load in the future; when the initial resource scheduling strategy arranges for a robot node to go to a target area to perform a task, the future state prediction results can be used to determine whether the future path cost of the robot node will increase.

[0078] S143. Based on the future resource change status, dynamically adjust the initial resource scheduling strategy corresponding to each robot node to obtain the target resource scheduling strategy corresponding to each robot node.

[0079] In some embodiments, when future resource changes indicate that a robot node may run out of power after executing the initial resource scheduling strategy, the multi-robot system can advance the energy replenishment of that robot node or reduce the high-energy-consuming tasks undertaken by that robot node; when future resource changes indicate that a robot node will have a high future computing load, the system can reduce the number of computing tasks for that robot node or transfer some computing tasks to other robot nodes; when future resource changes indicate that the future path cost of a certain path will increase, the system can adjust the path execution arrangement or assign the relevant tasks to robot nodes that are more suitable for execution.

[0080] In this embodiment, by predicting the future state of each robot node, the changes in energy, computing power, tasks, and paths that may occur during the subsequent execution of the initial resource scheduling strategy can be determined in advance. Then, the initial resource scheduling strategy can be dynamically adjusted based on the future resource changes, making the target resource scheduling strategy more adaptable to future state changes, thereby improving the stability of resource scheduling and the overall operating efficiency of the multi-robot system.

[0081] In some embodiments, such as Figure 6 As shown, S143 specifically includes: S211. Based on the future resource change status, determine the energy change trend corresponding to each robot node; Among them, "energy change trend" refers to the direction and magnitude of changes in the energy status of robot nodes in the future.

[0082] In some embodiments, a multi-robot system can determine the energy change trend for each robot node based on future power changes, future task execution status, and future path costs included in future resource change states. For example, when a robot node needs to continuously perform high-energy-consuming tasks or move long distances in the future, a continuous downward trend in energy consumption for that robot node can be determined; when a robot node undergoes an energy replenishment process in the future, an upward trend in energy consumption for that robot node can be determined.

[0083] S212. Based on the energy change trend, determine the computing power adjustment method for each robot node; Among them, "computing power adjustment method" refers to the method of adjusting the computing power usage of robot nodes according to the energy change trend of robot nodes.

[0084] In some embodiments, the multi-robot system determines a computing power adjustment method that matches the energy change trend corresponding to each robot node. For example, when the energy change trend of a robot node indicates that its future energy state is sufficient or rising, it can be determined to increase the computing power usage level or execute more computing tasks; when the energy change trend of a robot node indicates that its future energy state is continuously declining, it can be determined to reduce the computing power usage level, suspend unnecessary computing tasks, or postpone some computing tasks; when the energy change trend of a robot node indicates that its future energy state is basically stable, it can be determined to maintain the current computing power usage level.

[0085] S213. Based on the computing power adjustment method, adjust the computing power allocation content in the initial resource scheduling strategy to obtain the target resource scheduling strategy.

[0086] Among them, "computing power allocation content" refers to the content related to the use of computing resources, execution of computing tasks, or allocation of computing load in the initial resource scheduling strategy.

[0087] In some embodiments, the multi-robot system adjusts the computing power allocation in the initial resource scheduling strategy according to the computing power adjustment method. For example, when the computing power adjustment method is to increase the computing power utilization level, the number of computing tasks executed by the corresponding robot node can be increased, or the intensity of its computing task execution can be increased; when the computing power adjustment method is to decrease the computing power utilization level, the number of computing tasks undertaken by the corresponding robot node can be reduced, its computing load can be reduced, or some computing tasks can be transferred to other robot nodes; when the computing power adjustment method is to postpone some computing tasks, the execution time of the computing tasks can be adjusted so that the computing tasks are executed when the energy state of the robot node is more suitable.

[0088] In this embodiment, by determining the computing power adjustment method based on the future energy change trend of robot nodes, the computing power usage can be dynamically constrained by changes in energy status. Adjusting the computing power allocation in the initial resource scheduling strategy based on the computing power adjustment method can make the computing power allocation in the target resource scheduling strategy match the future energy change trend, thereby reducing computing interruptions caused by insufficient energy or resource waste caused by insufficient computing power utilization despite sufficient energy.

[0089] In some embodiments, such as Figure 7 As shown, S143 also includes: S221. Based on the future resource change status, determine the task change trend and path change trend corresponding to each robot node; Among them, "task change trend" refers to the direction and extent to which the task status of robot nodes may change in the future.

[0090] "Path change trend" refers to the direction and extent to which the path state of a robot node may change over a period of time.

[0091] In some embodiments, the multi-robot system can determine the task change trend and path change trend for each robot node based on future task execution status, future task requirements, future target location changes, future path distances, and future path costs contained in the future resource change status. For example, when the future resource change status indicates that the number of subsequent tasks for a robot node increases and the task completion time is shortened, it can be determined that the robot node has an increasing task pressure trend; when the future resource change status indicates that the subsequent path distance for a robot node increases or the target location changes, it can be determined that the robot node has an increasing path execution burden trend.

[0092] S222. Based on the task change trend and the path change trend, determine the task path collaborative adjustment method for each robot node; Among them, "task path collaborative adjustment method" refers to the method of collaboratively adjusting the task execution content and path execution content according to the task change trend and path change trend of robot nodes.

[0093] In some embodiments, the multi-robot system determines a collaborative adjustment method for task paths that matches the future task and path changes of each robot node, based on the task change trend and path change trend corresponding to that robot node. For example, when the task change trend of a robot node indicates that its subsequent tasks are highly urgent, and the path change trend indicates that the path cost to the target location is low, the execution order of the relevant tasks of that robot node can be increased; when the task change trend of a robot node indicates that the number of its subsequent tasks is increasing, and the path change trend indicates that its path cost is significantly increasing, some tasks can be adjusted to other robot nodes with lower path costs; when the path change trend of a robot node indicates that it needs to pass through multiple task areas during movement, the task execution order can be adjusted to match the path movement process with the task execution process.

[0094] S223. Based on the task path collaborative adjustment method, adjust the task execution content and path execution content in the initial resource scheduling strategy to obtain the target resource scheduling strategy.

[0095] "Task execution content" refers to the content related to task execution in the initial resource scheduling strategy.

[0096] "Path execution content" refers to the content related to path execution in the initial resource scheduling strategy.

[0097] In some embodiments, when the task path coordination adjustment method is to adjust the task execution order, tasks with lower path costs and shorter task completion times can be executed earlier; when the task path coordination adjustment method is to adjust the task execution nodes, a task can be moved from a robot node with a higher path cost to a robot node with a lower path cost; when the task path coordination adjustment method is to adjust the path execution order, the order in which robot nodes arrive at different target locations can be rearranged so that the path execution content matches the task execution content.

[0098] In this embodiment, by determining the task path coordination adjustment method based on the task change trend and path change trend corresponding to each robot node, a dynamic match can be formed between the task execution content and the path execution content. Adjusting the task execution content and path execution content in the initial resource scheduling strategy based on the task path coordination adjustment method can make the target resource scheduling strategy more adaptable to future task and path changes, thereby improving the continuity of task execution and the efficiency of path execution.

[0099] In some embodiments, such as Figure 8 As shown, S150 specifically includes: S151. Based on the target resource scheduling strategy, generate scheduling control instructions for each robot node; Among them, "scheduling control instructions" refer to instruction information generated according to the target resource scheduling strategy and used to control robot nodes to perform resource scheduling behavior.

[0100] In some embodiments, when a target resource scheduling policy instructs a robot node to perform a target task, a task execution instruction can be generated; when a target resource scheduling policy instructs a robot node to reduce its computing load or perform a specified computing task, a computing power allocation instruction can be generated; when a target resource scheduling policy instructs a robot node to go to an energy replenishment location, an energy replenishment instruction can be generated; and when a target resource scheduling policy instructs a robot node to move along a specified path, a path execution instruction can be generated.

[0101] S152. Send the scheduling control command to the corresponding robot node to control the corresponding robot node to perform resource scheduling behavior; In some embodiments, the multi-robot system can send different scheduling control instructions to the corresponding robot nodes based on the robot node identifier, so that each robot node can perform corresponding resource scheduling behavior according to the received scheduling control instructions. For example, the first robot node executes the corresponding task after receiving the task execution instruction, the second robot node goes to the energy replenishment location after receiving the energy replenishment instruction, and the third robot node moves according to the specified path after receiving the path execution instruction.

[0102] S153. Obtain the execution feedback information after the corresponding robot node performs the resource scheduling behavior; Among them, "execution feedback information" refers to the actual execution result information fed back by the robot node after executing resource scheduling behavior.

[0103] In some embodiments, after performing resource scheduling, the robot node can report back to the multi-robot system whether the task is completed, the task completion time, and the resource consumption during task execution; after completing computing power allocation, it can report the actual computing load, the completion status of the computing task, and changes in computing power usage; after completing energy replenishment, it can report the charging result, replenishment time, and the energy status after replenishment; and after completing path execution, it can report the actual movement path, the actual arrival time, and any abnormal situations during path execution.

[0104] S154. Based on the deviation information between the execution feedback information and the expected execution result corresponding to the target resource scheduling strategy, update the parameters in the unified scheduling model.

[0105] "Expected execution result" refers to the execution result predetermined according to the target resource scheduling strategy.

[0106] "Deviation information" refers to the difference between the execution feedback information and the expected execution result.

[0107] "Parameters" refer to the parameters used in the unified scheduling model to describe resource consumption, execution constraints, scheduling priorities, future state predictions, or node collaboration relationships.

[0108] In some embodiments, the multi-robot system compares the execution feedback information from robot nodes with the expected execution results corresponding to the target resource scheduling strategy to obtain deviation information and update the parameters in the unified scheduling model. This allows subsequent resource scheduling processes to better align with the actual operation of the multi-robot system. For example, when actual energy consumption exceeds expected energy consumption, the energy consumption parameters in the unified scheduling model can be updated; when the actual task completion time is longer than the expected task completion time, the task execution time parameters can be updated; when the actual path cost is higher than the expected path cost, the path cost parameters can be updated; and when there is a difference between actual and pre-calculated computing power usage, the computing power usage parameters can be updated.

[0109] In this embodiment, by obtaining execution feedback information after the robot node performs resource scheduling behavior, and updating the parameters in the unified scheduling model based on the deviation information between the execution feedback information and the expected execution result, the unified scheduling model can be continuously updated according to the actual execution situation, thereby improving the accuracy of subsequent resource scheduling.

[0110] In some embodiments, such as Figure 9 As shown, the resource scheduling method also includes: S311. Based on the target resource scheduling strategy corresponding to each robot node, determine the task coordination relationship and resource coordination relationship between multiple robot nodes; Among them, "task collaboration relationship" refers to the collaboration relationship formed by multiple robot nodes during task execution.

[0111] "Resource collaboration relationship" refers to the collaborative relationship formed by multiple robot nodes during the process of resource use or resource allocation.

[0112] In some embodiments, the multi-robot system can summarize and analyze the target resource scheduling strategy corresponding to each robot node to identify the collaborative needs and resource occupancy relationships between different robot nodes. For example, when multiple robot nodes are scheduled to perform tasks in the same or adjacent areas within a similar time period, a task collaboration relationship can be determined between the multiple robot nodes; when multiple robot nodes need to occupy the same energy supply resource or pass through the same path area within a similar time period, a resource collaboration relationship can be determined between the multiple robot nodes.

[0113] S312. Based on the task coordination relationship and the resource coordination relationship, the target resource scheduling strategy corresponding to at least one robot node is modified to obtain the modified target resource scheduling strategy. Among them, the "revised resource scheduling strategy" refers to the scheduling strategy obtained by readjusting the target resource scheduling strategy based on task coordination relationships and resource coordination relationships.

[0114] In some embodiments, when the tasks performed by two robot nodes are duplicated or conflict, the duplicate task of one robot node can be canceled, or one of the tasks can be rescheduled to another time for execution; when a robot node is unable to complete a task on time while other robot nodes have the conditions to execute it, the corresponding task can be rescheduled to other robot nodes for execution.

[0115] In other embodiments, when multiple robot nodes need to be recharged at the same time but energy recharge resources are limited, the energy recharge order can be adjusted according to the remaining power of each robot node and the urgency of the task; when the computing power load of a certain robot node is high and other robot nodes have available computing power, some computing tasks can be transferred to other robot nodes; when the path execution content of multiple robot nodes conflicts, the path execution content or execution time of at least one robot node can be adjusted.

[0116] S313. Based on the modified target resource scheduling strategy, control the corresponding robot node to perform resource scheduling behavior.

[0117] In this embodiment, by modifying the target resource scheduling strategy corresponding to at least one robot node based on the task coordination and resource coordination relationships among multiple robot nodes, the resource scheduling strategy of each robot node can not only adapt to its own future state changes, but also coordinate with the task execution and resource usage of other robot nodes. Controlling the corresponding robot node to perform resource scheduling behavior based on the modified target resource scheduling strategy can improve the overall consistency of task execution, resource usage and node coordination in the multi-robot system.

[0118] Figure 10 This is a schematic diagram of the architecture of the resource scheduling device provided in an embodiment of this application. Figure 10 As shown, the resource scheduling device 200 includes: a construction module 210, a scheduling module 220, a first determination module 230, an adjustment module 240, and a first control module 250, wherein: The construction module 210 is used to construct a unified scheduling model based on the current running state of each robot node in the multi-robot system; The scheduling module 220 is used to perform unified scheduling of resource scheduling behavior in the multi-robot system based on the unified scheduling model, so as to obtain a global resource collaborative scheduling result. The first determining module 230 is used to determine the initial resource scheduling strategy corresponding to each robot node based on the global resource collaborative scheduling result and the current running status of each robot node. The adjustment module 240 is used to dynamically adjust the initial resource scheduling strategy corresponding to each robot node based on the future state prediction results of each robot node, so as to obtain the target resource scheduling strategy corresponding to each robot node. The first control module 250 is used to control the corresponding robot node to perform resource scheduling behavior based on the target resource scheduling strategy.

[0119] In some embodiments, the construction module 210 is specifically used to: obtain the energy status, computing power status, task status, and path status of each robot node; determine the resource constraint relationship corresponding to each robot node and the collaborative scheduling relationship among all robot nodes based on the energy status, computing power status, task status, and path status; and construct the unified scheduling model based on the resource constraint relationship and the collaborative scheduling relationship.

[0120] In some embodiments, the scheduling module 220 is specifically used to: perform associated scheduling on the task execution behavior, computing power allocation behavior, energy supply behavior, and path planning behavior in the multi-robot system based on the unified scheduling model, to obtain resource association scheduling results; determine the resource occupancy information and execution constraint information corresponding to each resource scheduling behavior based on the resource association scheduling results; determine the scheduling priority corresponding to each resource scheduling behavior based on the resource occupancy information and the execution constraint information; and perform collaborative sorting and node allocation on each resource scheduling behavior based on the scheduling priority, to obtain the global resource collaborative scheduling result.

[0121] In some embodiments, the first determining module 230 is specifically configured to: determine the node scheduling content corresponding to each robot node from the global resource collaborative scheduling result; determine whether the node scheduling content meets the execution conditions of the corresponding robot node based on the current running state of each robot node; if the node scheduling content meets the execution conditions of the corresponding robot node, determine the initial resource scheduling strategy corresponding to each robot node based on the node scheduling content; if the node scheduling content does not meet the execution conditions of the corresponding robot node, correct the node scheduling content, and determine the initial resource scheduling strategy corresponding to each robot node based on the corrected node scheduling content.

[0122] In some embodiments, the adjustment module 240 includes: a first determining unit, a second determining unit, and an adjustment unit, wherein: the first determining unit is configured to determine the future state prediction result of each robot node based on the historical operating state and the current operating state of each robot node; the second determining unit is configured to determine the future resource change state of each robot node when executing the initial resource scheduling strategy based on the future state prediction result; and the adjustment unit is configured to dynamically adjust the initial resource scheduling strategy corresponding to each robot node based on the future resource change state to obtain the target resource scheduling strategy corresponding to each robot node.

[0123] In some embodiments, the adjustment unit is specifically used to: determine the energy change trend corresponding to each robot node based on the future resource change state; determine the computing power adjustment method corresponding to each robot node based on the energy change trend; and adjust the computing power allocation content in the initial resource scheduling strategy based on the computing power adjustment method to obtain the target resource scheduling strategy.

[0124] In some embodiments, the adjustment unit is specifically used to: determine the task change trend and path change trend corresponding to each robot node based on the future resource change state; determine the task path collaborative adjustment method corresponding to each robot node based on the task change trend and the path change trend; and adjust the task execution content and path execution content in the initial resource scheduling strategy based on the task path collaborative adjustment method to obtain the target resource scheduling strategy.

[0125] In some embodiments, the first control module 250 is specifically configured to: generate a scheduling control instruction for each robot node based on the target resource scheduling strategy; send the scheduling control instruction to the corresponding robot node to control the corresponding robot node to perform resource scheduling behavior; obtain execution feedback information after the corresponding robot node performs the resource scheduling behavior; and update the parameters in the unified scheduling model based on the deviation information between the execution feedback information and the expected execution result corresponding to the target resource scheduling strategy.

[0126] In some embodiments, the resource scheduling device 200 further includes: a second determining module, a correcting module, and a second controlling module, wherein: the second determining module is used to determine the task coordination relationship and resource coordination relationship among multiple robot nodes based on the target resource scheduling strategy corresponding to each robot node; the correcting module is used to correct the target resource scheduling strategy corresponding to at least one robot node based on the task coordination relationship and the resource coordination relationship to obtain a corrected target resource scheduling strategy; and the second controlling module is used to control the corresponding robot node to perform resource scheduling behavior based on the corrected target resource scheduling strategy.

[0127] Each functional module in the resource scheduling device described in the foregoing one or more embodiments can be logically regarded as a virtual functional unit undertaking a specific processing task. Its function can be implemented by software program instructions, processor execution logic or dedicated hardware circuit. The specific implementation method can be flexibly selected according to the actual deployment requirements, and is not specifically limited here.

[0128] Figure 11 The diagram shows a structural schematic of a multi-robot system according to an embodiment of this application. This embodiment does not limit the specific implementation of the multi-robot system.

[0129] like Figure 11 As shown, the multi-robot system 400 may include: multiple robot nodes 410, a processor 420, a communication interface 430, a memory 440, and a communication bus 450.

[0130] In this system, multiple robot nodes 410, processor 420, communication interface 430, and memory 440 communicate with each other via communication bus 450. Communication interface 430 is used for communication connections with other external devices. Processor 420 executes program 460 to implement the methods described in one or more of the above embodiments. Communication bus 450 includes wired and / or wireless connections.

[0131] Specifically, program 460 may include program code that includes computer operation instructions. When program 460 is invoked, processor 420 executes the computer operation instructions to implement the steps in one or more embodiments of the method.

[0132] Depending on the actual application scenario, the processor 420 can be of the appropriate type, including but not limited to mainstream embedded processors such as microcontrollers (MCUs) and ARM architecture processors, as well as other types of processors such as digital signal processors (DSPs), application-specific integrated circuits (ASICs), and field-programmable gate arrays (FPGAs), as long as they can provide the computing and control capabilities required by the actual application scenario.

[0133] Memory 440 is used to store program 460. It includes a program storage area and a data storage area. The program storage area is used to store firmware programs, embedded applications, and various functional modules; the data storage area is used to store data and calculation results during program execution. For example, memory 440 may include: on-chip RAM (for temporary data storage during program execution); on-chip Flash memory (for storing program code and configuration data); and EEPROM or other types of non-volatile memory (for storing parameters that need to be retained when power is off).

[0134] This application also provides a computer-readable storage medium. This computer-readable storage medium can be a non-volatile computer-readable storage medium. This computer-readable storage medium stores a computer program.

[0135] When executed by a processor, the computer program implements one or more steps of the method disclosed in the embodiments of this application. A complete computer program product is embodied on one or more computer-readable storage media containing the computer program disclosed in the embodiments of this application.

[0136] The above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Under the concept of this application, the technical features of the above embodiments or different embodiments can also be combined, the steps can be implemented in any order, and there are many other variations of different aspects of this application as described above. For the sake of brevity, they are not provided in detail. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. These modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

Claims

1. A resource scheduling method applied to a multi-robot system, characterized in that, The method includes: A unified scheduling model is constructed based on the current operating state of each robot node in the multi-robot system. Based on the unified scheduling model, the resource scheduling behavior of the multi-robot system is uniformly scheduled to obtain the global resource collaborative scheduling result. Based on the global resource collaborative scheduling results and the current operating status of each robot node, the initial resource scheduling strategy corresponding to each robot node is determined. Based on the future state prediction results of each robot node, the initial resource scheduling strategy corresponding to each robot node is dynamically adjusted to obtain the target resource scheduling strategy corresponding to each robot node. Based on the target resource scheduling strategy, the corresponding robot nodes are controlled to perform resource scheduling behavior.

2. The resource scheduling method according to claim 1, characterized in that, The construction of a unified scheduling model based on the current operating state of each robot node in the multi-robot system includes: Obtain the energy status, computing power status, task status, and path status of each robot node; Based on the energy status, computing power status, task status, and path status, determine the resource constraints for each robot node and the collaborative scheduling relationships among all robot nodes; Based on the resource constraints and the collaborative scheduling relationships, the unified scheduling model is constructed.

3. The resource scheduling method according to claim 1, characterized in that, The unified scheduling model is used to uniformly schedule resource scheduling behavior in the multi-robot system to obtain a global resource collaborative scheduling result, including: Based on the unified scheduling model, the task execution behavior, computing power allocation behavior, energy supply behavior and path planning behavior of the multi-robot system are correlated and scheduled to obtain resource correlation scheduling results. Based on the resource association scheduling results, determine the resource occupancy information and execution constraint information corresponding to each resource scheduling behavior; Based on the resource occupancy information and the execution constraint information, the scheduling priority corresponding to each resource scheduling behavior is determined; Based on the scheduling priority, each resource scheduling action is collaboratively sorted and node allocated to obtain the global resource collaborative scheduling result.

4. The resource scheduling method according to claim 1, characterized in that, The step of determining the initial resource scheduling strategy for each robot node based on the global resource collaborative scheduling results and the current operating status of each robot node includes: From the global resource collaborative scheduling results, determine the node scheduling content corresponding to each robot node; Based on the current running status of each robot node, determine whether the node scheduling content meets the execution conditions of the corresponding robot node; If the node scheduling content satisfies the execution conditions of the corresponding robot node, an initial resource scheduling strategy is determined for each robot node based on the node scheduling content. If the node scheduling content does not meet the execution conditions of the corresponding robot node, the node scheduling content is modified, and the initial resource scheduling strategy for each robot node is determined based on the modified node scheduling content.

5. The resource scheduling method according to claim 1, characterized in that, The initial resource scheduling strategy for each robot node is dynamically adjusted based on the future state prediction results of each robot node to obtain the target resource scheduling strategy for each robot node, including: Based on the historical and current operating states of each robot node, the future state prediction result of each robot node is determined; Based on the future state prediction results, the future resource change state of each robot node when executing the initial resource scheduling strategy is determined; Based on the future resource change status, the initial resource scheduling strategy corresponding to each robot node is dynamically adjusted to obtain the target resource scheduling strategy corresponding to each robot node.

6. The resource scheduling method according to claim 5, characterized in that, The step of dynamically adjusting the initial resource scheduling strategy for each robot node based on the future resource change status to obtain the target resource scheduling strategy for each robot node includes: Based on the future resource change status, determine the energy change trend corresponding to each robot node; Based on the energy change trend, determine the computing power adjustment method for each robot node; Based on the aforementioned computing power adjustment method, the computing power allocation content in the initial resource scheduling strategy is adjusted to obtain the target resource scheduling strategy.

7. The resource scheduling method according to claim 5, characterized in that, The step of dynamically adjusting the initial resource scheduling strategy for each robot node based on the future resource change state to obtain the target resource scheduling strategy for each robot node also includes: Based on the future resource change status, determine the task change trend and path change trend for each robot node; Based on the task change trend and the path change trend, determine the task path collaborative adjustment method for each robot node; Based on the task path collaborative adjustment method, the task execution content and path execution content in the initial resource scheduling strategy are adjusted to obtain the target resource scheduling strategy.

8. The resource scheduling method according to claim 1, characterized in that, The step of controlling the corresponding robot node to perform resource scheduling behavior based on the target resource scheduling strategy includes: Based on the target resource scheduling strategy, a scheduling control instruction is generated for each robot node; The scheduling control command is sent to the corresponding robot node to control the corresponding robot node to perform resource scheduling behavior; Obtain the execution feedback information after the corresponding robot node performs the resource scheduling behavior; Based on the deviation information between the execution feedback information and the expected execution result corresponding to the target resource scheduling strategy, the parameters in the unified scheduling model are updated.

9. The resource scheduling method according to any one of claims 1 to 8, characterized in that, The method further includes: Based on the target resource scheduling strategy corresponding to each robot node, the task coordination relationship and resource coordination relationship between multiple robot nodes are determined. Based on the task coordination relationship and the resource coordination relationship, the target resource scheduling strategy corresponding to at least one robot node is modified to obtain the modified target resource scheduling strategy. Based on the revised target resource scheduling strategy, the corresponding robot nodes are controlled to perform resource scheduling behavior.

10. A multi-robot system, characterized in that, include: The system includes multiple robot nodes, a processor, and a memory, wherein the memory stores computer program instructions that, when invoked by the processor, cause the processor to perform the method as described in any one of claims 1-9.

Images