179 results about "Multirobot systems" patented technology

The invention discloses a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA, and belongs to the technical field of path planning. The invention discloses a distributed multi-mobile robot path planning and obstacle avoidance method based on ACO-DWA. The method comprises the following steps: constructing an environment model; performing global path planning by using an ant colony algorithm ACO; establishing a motion model of the robot, and selecting an optimal speed group; and designing a conflict judgment rule, introducing a priority strategy, and solving conflicts according to the importance and urgency of robot tasks. According to the distributed multi-mobile-robot path planning and obstacle avoidance method based on ACO-DWA, an improved dynamic window method (DWA) and an ant colony optimization (ACO) algorithm are utilized, effective local path planning and global path optimization are achieved in the environment with high-density obstacles and unknown static obstacles, and safe obstacle avoidance of the robots is ensured. Meanwhile, a motion conflict judgment rule and a priority strategy in the multi-robot system are provided to coordinate the motion between the robots and between the robots and obstacles, and safe and efficient navigation is ensured.

The invention provides an industrial multi-robot intelligent collaborative planning method based on deep learning. The industrial multi-robot intelligent collaborative planning method comprises six parts including environment modeling, feature extraction, task allocation, trajectory planning, control instruction generation and rule distillation. The method comprises the following steps: acquiring multi-robot environment information by constructing a probability grid map and a topological structure, and extracting state and task features to form a comprehensive feature matrix; training an optimal task allocation strategy by adopting deep reinforcement learning, and combining CVAE and CEM joint modeling to optimize trajectory generation; an adaptive impedance controller based on MADDPG is further designed, and dynamic adjustment of interaction parameters is achieved; and finally, the control strategy is converted into a decision rule set through knowledge distillation, the control interpretability is improved, and the deployment complexity is reduced. According to the invention, the task cooperation efficiency and the control stability of the multi-robot system in a complex industrial environment can be effectively improved.

The invention relates to the technical field of robot control, particularly discloses a path planning and dynamic obstacle avoidance control method for multi-task collaborative operation of an industrial robot, and aims to solve the problems of task scheduling conflict, dynamic obstacle response lag and low collaborative efficiency in a multi-robot system. The method comprises the following steps: constructing a task-resource joint scheduling model and generating initial task allocation; planning a conflict-free collaborative path based on an improved space-time A star algorithm; predicting a dynamic obstacle trajectory by using an LSTM network and generating a space-time envelope; constructing a second-order safety barrier function fusing task priorities; local obstacle avoidance re-planning is realized through rolling horizon model predictive control; and the global rescheduling is triggered when the task delay exceeds the limit or the deadlock risk occurs. According to the technical scheme, closed-loop linkage of global task collaboration and local dynamic obstacle avoidance is realized, and the system collaboration efficiency, the obstacle avoidance success rate and the operation robustness are remarkably improved.

The invention discloses a safety cooperative control method of a multi-robot system, and relates to the technical field of robot control. The method comprises the following steps: constructing a multi-mechanical-arm multi-robot system model; setting an end effector of the follower robot in a Cartesian space, and determining a dynamic model of the multi-mechanical-arm multi-robot system; according to the input-driven fuzzy state observer and the kinetic model of the multi-mechanical-arm multi-robot system, a distributed fuzzy observer error is obtained; according to the symbol graph network topology, constructing a bidirectional grouping cooperation error, a bidirectional grouping tracking error and a fuzzy estimation error; a backstepping method is applied to obtain a time-varying Tan type Lyapunov obstacle function; and then determining a distributed formation controller and an adaptive law, and determining a distributed bidirectional cooperative controller of the multi-mechanical-arm multi-robot system. The operation safety of the multi-mechanical-arm multi-robot system in a specific task space range is improved.

The invention relates to a multi-robot system global flexible preset performance control method under an unknown state and a DoS attack, and belongs to the technical field of multi-robot system control. Aiming at the problems in the prior art that the communication topology is unstable due to attack, the initial error constraint is strict, and the input saturation conflicts with the performance, the following scheme is provided: dynamically reconstructing the attacked communication topology based on a bipartite consensus layering algorithm; designing a global error constraint mechanism fusing the scale function and the tunnel type performance function; constructing a balance control input and performance boundary of the input saturation self-adaptive auxiliary system; and estimating an unknown state and compensating an observation error by adopting a state observer. The method has the technical effects that the DoS attack resistance of the system is enhanced, the global preset performance control of any initial error is realized, the contradiction between input saturation and tracking performance is dynamically coordinated, and the multi-robot collaborative operation under the unbalanced topology is supported.

The invention relates to the technical field of information, and discloses a multi-machine collaborative operation control method and system based on a common inductance calculation control technology, and the method comprises the steps: obtaining task granularity distribution and a resource demand matrix, and carrying out the classification of the task granularity distribution and the resource demand matrix through a classification algorithm, and obtaining an initial task segmentation feature set; based on the optimal segmentation point set, a preliminary task allocation scheme is generated in combination with node calculation heterogeneity, and a task segmentation feature set is dynamically updated through a real-time sensing technology; a real-time task state set is generated through a group information sharing mechanism, a task allocation scheme is optimized through a dynamic programming algorithm, and it is ensured that tasks are reasonably allocated according to priorities and node resources; and the cooperation state matrix is updated through a real-time communication protocol, so that the accuracy and timeliness of a multi-node cooperation execution result are ensured. According to the invention, task allocation can be dynamically adjusted in a complex environment, resource utilization is optimized, the cooperation efficiency of a multi-robot system is improved, and the method can be widely applied to the fields of industrial production, warehouse logistics, disaster rescue and the like.

The invention discloses a multi-robot probability trajectory generation method based on distributed model predictive control, and relates to the technical field of multi-robot systems, and the method constructs collision avoidance constraints based on a safety corridor with time perception. Uncertainty introduced by state estimation noise and motion interference is fully considered in construction of collision avoidance constraint, probability collision avoidance constraint is converted into deterministic constraint of mean value and covariance of robot states, and robustness of collision avoidance is ensured; meanwhile, in order to solve the problem of deadlock possibly occurring in a multi-robot system, a warning tape mechanism is introduced into probability constraint for avoiding collision and is combined with a right hand rule, and robot trajectory planning is dynamically adjusted; finally, the probabilistic collision constraint avoiding and deadlock preventing mechanisms are integrated into a distributed model predictive control framework, so that a locally optimal collision-free trajectory is generated.

The invention provides a multi-robot collaborative exploration method based on intention reasoning and related equipment, and the method comprises the steps: obtaining the sensing data of each robot, constructing a local observation map of each robot according to the sensing data of each robot, and estimating the pose information of each robot, obtaining a global observation map of each robot, and fusing the plurality of global observation maps to obtain a global fusion map; each robot extracts a spatial feature map from the global observation map, pre-estimates own detection intention information according to exploration information shared by other robots, and obtains a global exploration target according to fusion of the exploration intention information and the spatial feature map; and each robot generates a target exploration path for the corresponding global exploration target based on the global fusion map, and explores according to the target exploration path, so that information sharing among multiple robots is realized, and the exploration capability of a multi-robot system in a complex environment is improved.

The invention discloses a multi-machine system number-real consistency error collaborative prediction method and system based on multi-agent reinforcement learning, and belongs to the technical field of multi-robot systems and artificial intelligence. According to the method, each robot is modeled as an intelligent agent, a centralized training distributed execution framework is adopted, and collaborative prediction of errors is achieved. Each agent collects own, neighbor and environment state information in real time, multi-source information integration is carried out through feature embedding and gating fusion, the influence of different information sources on errors is dynamically evaluated by adopting a layered attention mechanism, and propagation and coupling rules of the errors in a cluster are effectively modeled. In the training stage, the centralized commentator network is utilized to evaluate the global prediction performance, the cooperative optimization of each executor network is guided, and the prediction precision is improved. According to the method, active and collaborative prediction of the multi-machine system number-real consistency error is realized, and a key technical support is provided for error feed-forward compensation and system reliability guarantee.

The invention discloses a multi-robot collaborative job scheduling optimization method based on a graph neural network. The method comprises the following steps: S1, constructing a collaborative scheduling graph comprising robot nodes, task nodes and meeting point nodes; s2, initializing a feature vector of each node; s3, node feature aggregation and scheduling representation vector generation are executed through the graph neural network; s4, an improved Co-DPSIPP algorithm is adopted to calculate a handover score of a meeting point node, and optimization is carried out based on task completion time, energy consumption change and communication load; s5, selecting an optimal handover mode according to the handover score, and updating the scheduling graph; s6, executing task redistribution and path planning based on the updated scheduling graph; and S7, repeatedly executing reasoning and scheduling control until all tasks are completed, and outputting a scheduling scheme and a path sequence. The scheduling efficiency and path planning of the multi-robot system are remarkably improved, and the method is widely applied to automatic production and robot cooperation tasks.

The invention discloses a multi-robot fixed time cooperative hunting method and system based on a distributed time-varying optimization algorithm. The method comprises the following steps: establishing a dynamic model and a communication topological graph of a multi-robot system; the method comprises the following steps: converting a fixed time distributed hunting problem of a multi-robot system into a fixed time distributed time-varying optimization problem, and constructing a global objective function; designing a fixed time distributed gradient estimator, and enabling each robot to estimate gradient information of a system global objective function within fixed time in a distributed mode; designing a self-adaptive zero-order neural network which is used for approximating an inverse matrix of a Hessian matrix; and designing a fixed-time distributed time-varying optimization algorithm for the multi-robot system by adopting the gradient information of the global objective function estimated by the distributed gradient estimator in the step 3 and the inverse matrix of the Hessian matrix approximated by the adaptive zero-order neural network in the step 4, so that each robot encircles the dynamic target in the fixed time in a distributed mode. According to the invention, the multi-robot system is ensured to surround the moving target in a formation form within a fixed time.

The invention discloses a multi-robot collaborative surrounding and sinking method and system based on reinforcement learning, and relates to the technical field of robot collaborative control. The method comprises the steps that a multi-robot surrounding and sinking task model is established according to a target surrounding and sinking task, a chasing robot carries out surrounding and sinking on a target robot, and the limited degree of the target robot is quantified by utilizing a reachable safe area; based on a clustering algorithm, self-adaptive dynamic grouping is carried out on the pursuing robots in the enclosing and sinking task model; and based on the grouping condition of the pursuing robots, mining the environment observation information by using the multi-robot surrounding and sinking decision network, and optimizing the action decision of the pursuing robots according to the environment observation information. According to the method, technical means such as a self-adaptive grouping mechanism and strategy network optimization design are combined, efficient, stable and intelligent surrounding and sinking decision construction is achieved, and the task execution capacity of a multi-robot system in a complex environment can be remarkably improved.

The invention relates to the technical field of multi-robot systems, and discloses a distributed command filtering fuzzy control method for a multi-robot system, which comprises the following steps that: each robot reads own hardware configuration and software parameters, ensures that all sensors and actuators work normally, and sends a command to the robot after initialization is completed; the robot sends a state report to a central controller or an adjacent robot through a distributed communication protocol; loading a predefined fuzzy rule base, and setting an initial fuzzy rule; environmental data are collected in real time through a sensor, and a fuzzy rule is dynamically adjusted according to real-time data by using an online learning mechanism. According to the distributed command filtering fuzzy control method of the multi-robot system, by reading hardware configuration and software parameters, comprehensive understanding of robot states is achieved, normal work of all sensors and actuators is ensured, the system can ensure that each robot is in the optimal working state when started, and the system is suitable for being used in a large scale. And the reliability and the stability of the system are improved.

The invention discloses a distributed task scheduling system and method for multi-robot collaborative drawing, and the system employs a distributed architecture of a master control server and a plurality of robot clients, and achieves the decoupling of drawing task generation and execution through a task queue. The master control server receives the task request, asynchronously generates a robot control instruction from the image data through a built-in task calculation module, and maintains a robot state management module; and the robot client registers the unique identifier and keeps communication with the server. During scheduling, the server actively queries the state management module, selects an available robot to issue an instruction and atomically updates the state of the available robot to be unavailable; and after the task is completed, the client feeds back a receipt, and the server updates the state to be available. According to the method, the problems of high concurrency, repetition prevention and loss prevention of task scheduling in a multi-robot system are solved, and efficient and reliable parallel execution of non-standardized artistic creation tasks and elastic expansion of the system are realized.

The present invention discloses a safe cooperative control method for a multi-robot system, and relates to the field of robot control technology. The method comprises: constructing a multi-robot system model of a multi-robot manipulator; setting the end effector of the follower robot in the Cartesian space, and determining the dynamic model of the multi-robot system of a multi-robot manipulator; obtaining a distributed fuzzy observer error according to the input-driven fuzzy state observer and the dynamic model of the multi-robot system of a multi-robot manipulator; constructing a two-way grouping cooperative error, a two-way grouping tracking error, and a fuzzy estimation error according to the symbolic graph network topology; applying the backstepping method, obtaining a time-varying Tan-type Lyapunov obstacle function; and then determining a distributed formation controller and an adaptive law, and determining a distributed two-way cooperative controller of a multi-robot system of a multi-robot manipulator. The present invention improves the safety of a multi-robot system of a multi-robot manipulator operating within a specific task space.

The invention relates to the technical field of multi-robot tasks, and discloses a hybrid multi-robot cooperation method and system driven by a large language model. The method comprises the steps that a natural language task instruction input by a user is received, semantic analysis is conducted through a task decomposition large language model, and a task set containing a plurality of structured sub-tasks is generated; a plurality of robots adopt a decentralized negotiation mechanism to respond and compete for each subtask, and a task allocation result when a single robot is used as an execution main body is determined; when the skill demand of the subtask exceeds the skill space of any single robot, triggering a collaborative task team construction process based on a decentralized language negotiation mechanism; and each execution main body executes the allocated sub-task according to the negotiation result. According to the method, the understanding ability and execution flexibility of the multi-robot system for the natural language task can be remarkably enhanced, and the cooperation efficiency and task completion quality of the multi-robot system in a complex task scene are improved.

The invention discloses a distributed dynamic task allocation method and device, equipment and a storage medium, and relates to the technical field of multi-robot control, the method is applied to a multi-robot system, and the method comprises the following steps: obtaining communication state information of each mobile robot, and determining minimum communication topological graph information by adopting a preset distributed minimum spanning tree algorithm; inputting the system parameters and the minimum communication topological graph information into a preset k-WTA network calculation formula to obtain activation signals corresponding to the mobile robots; and obtaining position information of the to-be-tracked target equipment and the mobile robots, and performing task allocation on the mobile robots according to the position information and the activation signals. According to the method, the complex communication topology in the multi-robot system is simplified through the minimum spanning tree algorithm, and calculation is performed based on the minimum communication topological graph information, so that the consumption of calculation resources is reduced, and the task allocation efficiency of each mobile robot is improved.

The invention discloses a multi-robot cooperative positioning and elastic formation control method based on a distributed Kalman consistency filtering (DKCF) algorithm. Aiming at the problems of insufficient sensor noise suppression, poor robustness under network attack and the like in the prior art, a dynamic model containing process noise and a sensor model of measurement noise are constructed, and a consistency state estimation method of weighted fusion observation data is provided. According to the method, multi-modal redundant sensor information is fused, and an improved DKCF is combined, so that cooperative state estimation between robots is realized, and the positioning precision in a noise environment is effectively improved; an elastic positioning framework under denial of service attack is established, and the robustness to network attack is enhanced while the formation precision is ensured through the joint design of distributed cooperative position estimation and a formation controller. According to the method, high positioning precision and a stable formation form can still be kept in the face of network attacks of denial of service, and the safety cooperation performance of a multi-robot system in a complex scene is remarkably improved.

The invention relates to the technical field of industrial automation, in particular to a multi-robot control method and system based on the industrial Internet of Things and a medium. The method comprises the following steps: acquiring industrial warehouse channel data; generating a warehouse digital twin model based on the industrial warehouse channel data; performing multi-robot path conflict detection according to the warehouse digital twin model to obtain path conflict data; determining a robot number according to the path conflict data; performing motor winding damage detection based on the robot number to obtain motor winding damage data; performing iron core loosening analysis based on the motor winding damage data to obtain iron core loosening data; performing vibration suppression control optimization according to the iron core loosening data to obtain vibration suppression control data; and robot scheduling is carried out according to the path conflict data to obtain robot scheduling data. According to the invention, the path planning precision and fault response efficiency of the multi-robot system are improved based on the industrial automation technology, and the operation efficiency of industrial warehouse operation is improved.

The invention discloses a multi-robot affine formation path optimization method driven by a virtual navigator, and the method comprises the steps: firstly setting a point set composed of the virtual navigator and followers, enabling the virtual navigator to form a convex hull and meet an affine localization condition, so as to guarantee that the followers are always located in the convex hull; generating a collision-free reference trajectory of the mass center of the virtual navigator based on a fast random tree algorithm; constructing a nonlinear constraint optimization problem NCO which contains linear transformation constraint and meets smoothness, obstacle avoidance, speed and angular speed constraint; and establishing a communication graph of the virtual navigator and the follower, calculating balance stress, designing a smoother follower distributed differential speed control law, and realizing formation path optimization and obstacle avoidance control. Compared with an existing method, the formation path generation method has the advantages that the smooth and stable formation path fitting the reference trajectory can be generated while obstacle avoidance and physical feasibility are guaranteed, and the cooperative motion capability and the control performance of a multi-robot system in a complex environment are improved.

The invention discloses a multi-robot safety control method and system based on a large language model, and the method comprises the steps: receiving first information through a first large language model, and generating a reference planner meeting the task requirements; if the reference planner output by the first large language model has compiling errors, the multi-robot distributed safety control system automatically captures error information and transmits the error information to the second large language model, the second large language model checks and modifies codes, it is ensured that the codes can be correctly executed, and the modified correct codes are input into the reference planner; and according to the control signal obtained by the reference planner, all the robots of the multi-robot system are controlled to work cooperatively through the control signal to complete a specified task and safely reach a specified target point. According to the invention, the stability and safety of all robots in the robot system can be ensured in a complex environment.

The invention relates to the technical field of robot fault recognition, and provides a multi-robot consistency Byzantine fault detection method. The method comprises the steps that observation motion information of each target robot in the multi-robot system at the current moment is acquired; encrypting each piece of observation motion information to obtain encrypted information of each piece of observation motion information, and uploading all the encrypted information to the block chain; decrypting and integrating all encrypted information at the current moment on the block chain to obtain a state data frame at the current moment; and when the state data frames are conflict data frames, determining the number of supporters and the number of counters of each target robot according to the conflict data frames, and determining a Byzantine fault robot from all the target robots based on the number of supporters and the number of counters of all the target robots. According to the method, the Byzantine fault identification reliability can be improved.

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.

The embodiment of the invention relates to a multi-robot map fusion algorithm without prior information. The multi-robot map fusion algorithm comprises the steps that each robot constructs acquired surrounding environment information into a local map by using an SLAM (Simultaneous Localization and Mapping) algorithm; when there is a repeatedly accessed area between the local maps, the plurality of robots estimate closed-loop measurements between the plurality of robots through appearance-based scene recognition by a single robot according to closed-loop detection; false closed-loop measurements are rejected by calculating graph topology correctness of closed-loop measurements between multiple robots and consistency between closed loops and selecting an accurate topology-correct and consistent subset, the topology-correct and consistent subset is added to a factor graph to optimize a pose graph to construct a globally consistent map. According to the embodiment of the invention, the global map is constructed according to the space-time consistency of closed-loop measurement between the robots, compared with a traditional method, the real-time performance is improved by more than 70%, and in a multi-robot SLAM system, compared with a subset formed by sensing deviation, the real closed-loop measurement has higher topology correctness and consistency.

The invention discloses a vision-based mobile robot navigation-following formation control method and system. The method comprises the steps of S1, installing vision cameras on other mobile robots except a first mobile robot in a multi-robot system; s2, establishing a dipole vector field, and generating and maintaining a required formation shape; s3, converting the visibility constraint of the visual camera into an angular velocity input constraint and a linear velocity input constraint, respectively constructing an angular velocity control scheme and a linear velocity control scheme, and then respectively designing an angular velocity auxiliary system and a linear velocity auxiliary system by using the angular velocity control scheme and the linear velocity control scheme; and S4, estimating the pose of the navigator, and realizing formation control under the condition of no communication according to the pose of the navigator in combination with the angular velocity, the state of the linear velocity auxiliary system and the dipole vector field. According to the method, the communication-free autonomous operation capability of the navigation-following formation mode is promoted, and the collaboration and stability of the formation are ensured by designing the dipole vector field.

The invention discloses a large-model-driven intelligent multi-robot area search method and system, and the method comprises the steps: constructing a coarse and fine dual-resolution area search grid map, and constructing a layered coupling decision information model based on the coarse and fine dual-resolution area search grid map; a decision information model with rich levels is constructed for the multi-robot system from the global macroscopic and microscopic angles, and the multi-robot system can be helped to make efficient and intelligent decisions; a high-level intelligent decision-making module is constructed to perform local optimal state judgment and guide the robot to quickly approach an unsearched area by utilizing the strong thrust capability of the large model, so that the overall search efficiency of the multi-robot system is improved; the designed low-level motion decision-making module makes full use of guidance information provided by a large model in the high-level intelligent decision-making module, and combines the advantages that a biological inspiration neural network model does not depend on environmental prior information and does not need training, so that the multi-robot system is efficiently driven to quickly complete a given area search task.

The invention relates to the technical field of data processing, and discloses an AMR transmission path data comprehensive processing method and system. The method comprises the following steps: processing environment sensing data through a data fusion algorithm to obtain an environment state matrix for recording information of an obstacle, a dynamic target and a passable area; robot state information and the environment state matrix are combined to obtain a joint state space integrating global and local information; training the joint state space through a multi-agent reinforcement learning algorithm to obtain a collaborative decision model composed of an independent strategy network and a shared evaluation network; processing the control instruction through a trajectory prediction model to obtain a collaborative path set storing robot path information; and analyzing the collaborative path set through a conflict detection algorithm, and triggering a scheduling strategy to obtain a path optimization result when a conflict is detected. According to the invention, the cooperative efficiency of the multi-robot system and the intelligent level of path planning are obviously improved.

The invention relates to the field of intelligent control, in particular to a robot motion control method and system suitable for multiple scenes, and the method comprises the steps: carrying out the modeling of a task scene, and initializing the parameters of a robot; according to the task priority table, the tasks are dynamically allocated to all the robots; a local strategy of each robot is generated by using MARL, and path planning is performed by using an SAC algorithm; the robot extracts an experience optimization strategy shared by other robots from the global shared pool, and stores key experience into the global shared pool after completing a single-step action; detecting path conflicts, performing path optimization through a game theory algorithm, and updating the task priority table in real time until the task is completed. According to the motion control method, MARL and a single-robot SAC algorithm are combined, multi-scene application is achieved through experience sharing among intelligent agents, and the cooperation efficiency and the strategy convergence speed of a multi-robot system can be remarkably improved.

The invention discloses a fixed-time robust encircling control method for a multi-robot system under a switching topology, and the method comprises the steps: building a dynamic model of the multi-robot system under a multi-leader mode, defining the related concept and control target of encircling control, building a multi-robot system encircling control model under the switching topology, and carrying out the fixed-time robust encircling control of the multi-robot system under the switching topology; according to the method, switching topology attributes are defined, conditions which need to be met by switching topology are provided, setting time of fixed time convergence constraint under the switching topology conditions is defined, then an estimator based on an RBF neural network is designed, compensation is carried out through an unknown interference estimation item obtained by substituting the estimator, and a robust encircling controller under the switching topology is designed. And the controller is substituted into the multi-robot system dynamic model to obtain a closed-loop error system, and fixed-time encircling control under the switching topology is realized. According to the method, adverse effects caused by topology switching and external disturbance in the multi-robot system can be effectively suppressed, so that the robustness and effectiveness of a multi-robot encircling control scheme are improved.

In order to solve the problem, the invention provides an intention-aware multi-robot scheduling optimization method based on a graph neural network, and the method comprises the steps: constructing an explicit intention vector of a multi-tuple structure, fusing a robot state, task information and a resource use condition, constructing an intention graph representing a task conflict and a cooperative relationship, and introducing a spectrum sparsification mechanism; then, a graph attention network and a strategy updating module are adopted to jointly train a strategy model, strategy optimization of each robot node on a graph structure is achieved, and therefore the overall task income is maximized, path conflicts and resource competition are minimized, and a universal path planning algorithm is used for dynamically constructing a conflict avoidance path; compared with the prior art, the method has higher real-time performance, expandability and environment adaptability, and the scheduling efficiency and the task completion rate of the multi-robot system in dynamic scenes such as disaster response and the like are remarkably improved.