Multi-agent motion sequential control method and system

Abstract

The invention relates to the technical field of motion time sequence control, in particular to a multi-agent motion time sequence control method and system. Comprising a path sensing module, a spatial-temporal feature extraction module, a dynamic road right distribution module, an anomaly detection and recovery module, a distributed cooperative communication module, a real-time path updating module, a performance evaluation and optimization module and a central coordination module. The right of way is distributed according to the passing sequence to ensure that the intelligent agent orderly passes through the intersection. Meanwhile, motion abnormity can be monitored, a recovery mechanism can be triggered, local information sharing and negotiation are achieved, paths are dynamically adjusted according to road right changes, system performance is continuously evaluated and optimized, and all modules are planned as a whole to process global constraints. The problem that an existing multi-agent motion time sequence control system cannot efficiently cope with the motion speed difference of agents and dynamic obstacles while precise time synchronization is guaranteed is solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of motion timing control, in particular to a multi-agent motion timing control method and system. BACKGROUND

[0002] In the application field of multi-agent systems, such as logistics robot scheduling, unmanned aerial vehicle swarm control, etc., how to achieve efficient and reliable motion timing control is a key technical challenge. Traditional methods often rely on precise time synchronization, requiring all agents to run strictly according to a pre-set schedule, but this approach has many problems.

[0003] Firstly, achieving global precise time synchronization requires high-precision clock devices and complex synchronization algorithms, which not only increases the hardware cost of the system, but also puts extremely high requirements on the reliability of the communication network. Secondly, since agents are affected by various unpredictable factors in actual operation, such as battery power fluctuations, mechanical component wear, environmental obstacles, etc., it is difficult to maintain synchronization at all times. Once a certain agent is delayed or advanced, it may trigger a series of chain reactions, leading to a decline in the performance of the entire system or even dangerous situations such as collisions.

[0004] In the prior art, although some methods attempt to improve the performance of multi-agent motion timing control by optimizing path planning or increasing communication frequency, these methods mostly can only perform local optimization for specific scenarios, lacking adaptability to complex dynamic environments and effective management strategies for large-scale agent clusters. SUMMARY

[0005] The purpose of the present application is to provide a multi-agent motion timing control method and system, aiming to solve the problem that existing multi-agent motion timing control systems cannot efficiently cope with agent motion speed differences and dynamic obstacles while ensuring precise time synchronization.

[0006] To achieve the above purpose, in a first aspect, the present application provides a multi-agent motion timing control method, comprising a path perception module, a space-time feature extraction module, a dynamic road right allocation module, an anomaly detection and recovery module, a distributed cooperative communication module, a real-time path update module, a performance evaluation and optimization module, and a central coordination module. The path perception module, the space-time feature extraction module, the dynamic road right allocation module, the real-time path update module, the anomaly detection and recovery module, the distributed cooperative communication module, the performance evaluation and optimization module, and the central coordination module are connected in sequence. The path perception module is used to obtain the path planning scene of the agent group and the motion path of all agents in real time. The space-time feature extraction module is used to identify the intersection of the motion path and calculate the agent passing time feature. The dynamic right-of-way allocation module is configured to allocate right-of-way at path intersection points based on the order of the intelligent agents; The anomaly detection and recovery module is configured to monitor abnormal motion of the intelligent agents and trigger a recovery mechanism; The distributed cooperative communication module is configured to enable local information sharing and negotiation among the intelligent agents; The real-time path updating module is configured to dynamically adjust the motion paths of the intelligent agents according to changes in the right-of-way; The performance evaluation and optimization module is configured to continuously evaluate the performance of the system and optimize control parameters; The central coordination module is configured to coordinate the various functional modules and handle global constraints.

[0007] The path perception module includes a data acquisition unit, a data fusion unit, a path recognition unit, and a scenario update unit; The data acquisition unit is configured to acquire real-time data related to the motion paths of the intelligent agents in the path planning scenario; The data fusion unit is configured to fuse the collected multi-source data; The path recognition unit is configured to recognize the motion paths of the intelligent agents from the fused data; The scenario update unit is configured to update the path planning scenario in real time based on the motion of the intelligent agents.

[0008] The spatiotemporal feature extraction module includes an intersection point identification unit, a time feature calculation unit, and a feature compression unit; The intersection point identification unit is configured to identify intersection points on the motion paths; The time feature calculation unit is configured to calculate the time features of the intelligent agents passing through the intersection points; The feature compression unit is configured to map the three-dimensional spatiotemporal trajectory into a two-dimensional spatiotemporal feature map using a spatiotemporal compression algorithm.

[0009] The dynamic right-of-way allocation module includes a right-of-way evaluation unit, an allocation strategy unit, and a conflict resolution unit; The right-of-way evaluation unit is configured to evaluate the right-of-way priority of the intelligent agents at path intersection points; The allocation strategy unit is configured to develop a right-of-way allocation strategy and allocate right-of-way based on the evaluation results; The conflict resolution unit is configured to handle conflicts during the right-of-way allocation process.

[0010] The anomaly detection and recovery module includes an anomaly monitoring unit, a recovery strategy unit, and a multi-level fault tolerance unit; The anomaly monitoring unit is configured to monitor abnormal behavior during the motion of the intelligent agents in real time; The recovery strategy unit is configured to formulate a recovery strategy in an abnormal situation. The multi-level fault-tolerant unit is configured to implement a multi-level fault-tolerant mechanism.

[0011] The distributed cooperative communication module includes a communication management unit, an information sharing unit, and a negotiation processing unit. The communication management unit is configured to manage the communication link between the intelligent agents. The information sharing unit is configured to realize local information sharing between intelligent agents. The negotiation processing unit is configured to process the negotiation process between intelligent agents.

[0012] In a second aspect, a multi-agent motion timing control method is provided for the multi-agent motion timing control system of the first aspect, including the following steps: Real-time acquisition of intelligent agent group path planning scene and related motion data, identification of motion path after fusion processing and update of scene; Based on the intersection point identified by path perception, the intelligent agent passing time characteristics are calculated, and the time-space compression algorithm is used to reduce the computational complexity; Evaluate the intelligent agent's right-of-way priority at the intersection, develop a distribution strategy to handle conflicts and determine the intelligent agent's passing sequence; Monitor the intelligent agent's motion abnormalities and develop a recovery strategy to ensure stable operation of the system with the help of a multi-level fault-tolerant mechanism; Manage the intelligent agent's communication link and share local information, process the negotiation process between intelligent agents, and support collaboration and path adjustment; Evaluate the path state, perform update operations on the affected segment with incremental optimization, and guide the intelligent agent to move according to the new path; Monitor system performance indicators and evaluate and analyze the optimization and adjustment of the system; Manage resources, global scheduling and coordination, and predictively control resource allocation in advance.

[0013] The multi-agent motion timing control system of the present application can real-time acquire intelligent agent motion paths and update scenes, accurately identify intersection points and calculate time characteristics, assign right-of-way according to passing sequence, and ensure that intelligent agents pass through intersection points in an orderly manner. At the same time, it can monitor motion abnormalities and trigger a recovery mechanism, realize local information sharing and negotiation, dynamically adjust paths according to changes in right-of-way, continuously evaluate and optimize system performance, and coordinate the processing of global constraints by various modules. The existing multi-agent motion timing control system cannot guarantee accurate time synchronization while efficiently dealing with intelligent agent motion speed differences and dynamic obstacles. BRIEF DESCRIPTION OF DRAWINGS

[0014] In order to make the technical solutions of the embodiments of the present application or the prior art clearer, the accompanying drawings needed in the embodiments or prior art description will be briefly introduced below. Obviously, the accompanying drawings in the following description only show some embodiments of the present application, and for those skilled in the art, other drawings can also be obtained from these drawings without creative effort.

[0015] Figure 1 is a schematic diagram of a multi-agent motion timing control system provided by the present application.

[0016] Figure 2 is a flowchart of a path perception module.

[0017] Figure 3 is a flowchart of a space-time feature extraction module.

[0018] Figure 4 is a flowchart of a dynamic right-of-way allocation module.

[0019] Figure 5 is a flowchart of an anomaly detection and recovery module.

[0020] Figure 6 is a flowchart of a distributed cooperative communication module.

[0021] Figure 7 is a flowchart of a real-time path update module.

[0022] Figure 8 is a flowchart of a performance evaluation and optimization module.

[0023] Figure 9 is a flowchart of a central coordination module.

[0024] Figure 10 is a flowchart of a multi-agent motion timing control method provided by the present application.

[0025] In the figure: 1-path perception module, 2-spatiotemporal feature extraction module, 3-dynamic right-of-way allocation module, 4-anomaly detection recovery module, 5-distributed cooperative communication module, 6-real-time path update module, 7-performance evaluation optimization module, 8-central coordination module, 11-data acquisition unit, 12-data fusion unit, 13-path identification unit, 14-scene update unit, 21-intersection identification unit, 22-time feature calculation unit, 23-feature compression unit, 31-right-of-way evaluation unit, 32-allocation strategy unit, 33-conflict resolution unit, 41-anomaly monitoring unit, 42-recovery strategy unit, 43-multistage fault-tolerant unit, 51-communication management unit, 52-information sharing unit, 53-negotiation processing unit, 61-path evaluation unit, 62-incremental optimization unit, 63-update execution unit, 71-performance monitoring unit, 72-index evaluation unit, 73-optimization adjustment unit, 81-resource management unit, 82-global scheduling unit, 83-predictive control unit. DETAILED DESCRIPTION

[0026] Embodiments of the present application are described below in detail, examples of which are shown in the drawings, wherein the same or similar reference signs represent the same or similar elements or elements having the same or similar functions throughout. The embodiments described below by reference to the drawings are exemplary and are intended to explain the present application, and cannot be understood as a limitation of the present application.

[0027] Please refer to Figures 1 to 9 , in a first aspect, the present application provides a multi-agent motion timing control system, comprising a path perception module 1, a spatiotemporal feature extraction module 2, a dynamic right-of-way allocation module 3, an anomaly detection recovery module 4, a distributed cooperative communication module 5, a real-time path update module 6, a performance evaluation optimization module 7 and a central coordination module 8, the path perception module 1, the spatiotemporal feature extraction module 2, the dynamic right-of-way allocation module 3, the real-time path update module 6, the anomaly detection recovery module 4, the distributed cooperative communication module 5, the performance evaluation optimization module 7 and the central coordination module 8 are connected in turn; The path perception module 1 is used for real-time acquisition of agent group path planning scene and all agent motion paths; The spatiotemporal feature extraction module 2 is used for identifying motion path intersections and calculating agent passing time features; The dynamic right-of-way allocation module 3 is used for allocating agent path intersection right-of-way based on passing sequence; The anomaly detection recovery module 4 is used for monitoring agent motion anomalies and triggering recovery mechanisms; The distributed cooperative communication module 5 is used for realizing local information sharing and negotiation among agents; The real-time path updating module 6 is configured to dynamically adjust the motion path of the agent according to the change of the right-to-use; The performance evaluation optimization module 7 is configured to continuously evaluate the system performance and optimize the control parameters. The central coordination module 8 is configured to coordinate the functions of the modules and handle the global constraints.

[0028] In the embodiment, Further, the path perception module 1 comprises a data acquisition unit 11, a data fusion unit 12, a path recognition unit 13, and a scene updating unit 14. The data acquisition unit 11 is configured to acquire the data related to the motion path of the agent group and the path planning scene in real time. The data fusion unit 12 is configured to fuse the collected multi-source data. The path recognition unit 13 is configured to recognize the motion path of the agent from the fused data. The scene updating unit 14 is configured to update the path planning scene in real time according to the motion of the agent.

[0029] In the embodiment, the data acquisition unit 11 observes the position, speed, direction, and other information of the agent in real time; the data fusion unit 12 fuses these multi-source data, removes noise and errors, and improves data quality; the path recognition unit 13 accurately recognizes the motion path of the agent from the processed data; and the scene updating unit 14 updates the path planning scene in real time according to the motion of the agent, ensuring that the scene information is consistent with the actual situation.

[0030] Further, the spatio-temporal feature extraction module 2 comprises an intersection identification unit 21, a time feature calculation unit 22, and a feature compression unit 23. The intersection identification unit 21 is configured to identify the intersection points on the motion path. The time feature calculation unit 22 is configured to calculate the time features of the agent passing through the intersection points. The feature compression unit 23 is configured to map the three-dimensional spatio-temporal trajectory to a two-dimensional spatio-temporal feature map using a spatio-temporal compression algorithm.

[0031] In the embodiment, first, the intersection identification unit 21 finds out the intersection points on different motion paths, i.e., the key positions where the agents may meet or conflict. Then, the time feature calculation unit 22 calculates the time features (estimated arrival time, passing time, etc.) of the agents passing through these intersection points. Finally, the feature compression unit 23 maps the three-dimensional spatio-temporal trajectory of the agent to a two-dimensional spatio-temporal feature map using a spatio-temporal compression algorithm, greatly reducing the computational complexity.

[0032] Further, the dynamic right-of-way allocation module 3 includes a right-of-way evaluation unit 31, an allocation strategy unit 32, and a conflict resolution unit 33. The right-of-way evaluation unit 31 is configured to evaluate the right-of-way priority of the intelligent agents at the path intersection. The allocation strategy unit 32 is configured to formulate a right-of-way allocation strategy based on the evaluation results. The conflict resolution unit 33 is configured to handle conflicts in the right-of-way allocation process.

[0033] In this embodiment, the right-of-way evaluation unit 31 considers factors such as the motion state of the intelligent agents and the urgency of the task to evaluate the right-of-way priority of each intelligent agent at the intersection. The allocation strategy unit 32 formulates a reasonable right-of-way allocation strategy based on the evaluation results to ensure that the intelligent agents pass through the intersection in an orderly manner. The conflict resolution unit 33 is responsible for handling possible right-of-way conflicts, such as when two or more intelligent agents have the same priority and arrive at the intersection at the same time. Through mechanisms such as random allocation and intelligent agent number priority, the uniqueness and feasibility of the allocation are ensured.

[0034] Further, the anomaly detection and recovery module 4 includes an anomaly monitoring unit 41, a recovery strategy unit 42, and a multi-level fault tolerance unit 43. The anomaly monitoring unit 41 is configured to monitor abnormal behavior in real time during the motion of the intelligent agents. The recovery strategy unit 42 is configured to formulate a recovery strategy in abnormal situations. The multi-level fault tolerance unit 43 is configured to implement a multi-level fault tolerance mechanism.

[0035] In this embodiment, the anomaly monitoring unit 41 uses sensors and other devices to sense information such as the speed and motion trajectory of the intelligent agents in real time, and promptly detects abnormal behavior (such as sudden changes in speed and path deviation). Once an anomaly is detected, the recovery strategy unit 42 immediately formulates a corresponding recovery strategy, such as re-planning the path or adjusting the speed. At the same time, the multi-level fault tolerance unit 43 starts the fault tolerance mechanism to ensure that the system can still operate normally when some intelligent agents are abnormal.

[0036] Further, the distributed cooperative communication module 5 includes a communication management unit 51, an information sharing unit 52, and a negotiation processing unit 53. The communication management unit 51 is configured to manage the communication links between intelligent agents. The information sharing unit 52 is configured to enable local information sharing between intelligent agents. The negotiation processing unit 53 is configured to handle the negotiation process between intelligent agents.

[0037] In the embodiment, the communication management unit 51, the information sharing unit 52 and the negotiation processing unit 53 are responsible for managing the communication links between agents, realizing local information sharing and negotiation processing between agents.

[0038] Further, the real-time path updating module 6 includes a path evaluation unit 61, an incremental optimization unit 62 and an updating execution unit 63; The path evaluation unit 61 is configured to evaluate the state and feasibility of the current path; The incremental optimization unit 62 is configured to perform incremental optimization on the affected path segments; The updating execution unit 63 is configured to perform path updating operations.

[0039] In the embodiment, the path evaluation unit 61 evaluates the state and feasibility of the current path, the incremental optimization unit 62 performs incremental optimization on the affected path segments, and the updating execution unit 63 sends new path planning information to agents to guide them to move according to the updated path. The performance evaluation and optimization module 7 continuously monitors the performance indicators of the system (total path length, maximum completion time, energy consumption, number of conflicts, etc.).

[0040] Further, the performance evaluation and optimization module 7 includes a performance monitoring unit 71, an indicator evaluation unit 72 and an optimization adjustment unit 73; The performance monitoring unit 71 is configured to monitor the performance indicators of the system; The indicator evaluation unit 72 is configured to evaluate and analyze the performance indicators; The optimization adjustment unit 73 is configured to optimize and adjust the system according to the evaluation results.

[0041] In the embodiment, the indicator evaluation unit 72 evaluates and analyzes these indicators to find out the strengths and weaknesses of the system. The optimization adjustment unit 73 optimizes and adjusts the path planning algorithm, road right allocation strategy, communication mechanism, etc. of the system according to the evaluation results, to improve the overall performance of the system.

[0042] Further, the central coordination module 8 includes a resource management unit 81, a global scheduling unit 82 and a predictive control unit 83; The resource management unit 81 is configured to manage system resources and ensure reasonable allocation and utilization of resources; The global scheduling unit 82 is configured to perform global scheduling and coordination; The predictive control unit 83 is configured to implement predictive control functions and adjust resource allocation in advance In this embodiment, the global scheduling unit 82 coordinates the motion timing and path planning of the agents to achieve overall optimization. The predictive control unit 83 adjusts resource allocation and path planning in advance according to the motion trend and historical data of the agents to avoid potential conflicts and congestion, improving the foresight and adaptability of the system.

[0043] Please refer to Figure 10 , in a second aspect, a multi-agent motion timing control method for the multi-agent motion timing control system of the first aspect, comprising the following steps: S1 Real-time acquisition of agent group path planning scene and related motion data, identification of motion path after fusion processing and update of scene; Specifically, the data acquisition unit 11 of the path perception module 1 collects the position, speed, direction and other information of the agents through the sensor network. These data may come from different types of sensors such as laser radar, camera, IMU, etc. The data fusion unit 12 fuses these multi-source data and uses Kalman filtering algorithm to remove noise and error, improving the accuracy and reliability of the data. Then, the path recognition unit 13 uses path recognition algorithm (A algorithm, Dijkstra algorithm, etc.) to accurately identify the motion path of the agent. Finally, the scene update unit 14 updates the path planning scene according to the real-time motion of the agent to ensure the timeliness of the scene information.

[0044] S2 Based on the intersection point identified by path perception, calculate the time characteristics of the agent passing through, and use the space-time compression algorithm to reduce the computational complexity; Specifically, the system extracts the space-time features based on the data obtained by the path perception module 1. First, the intersection point identification unit 21 finds out the intersection points on the motion paths of different agents through path analysis algorithm. These intersection points are the key positions where the agents may meet or conflict. Then, the time feature calculation unit 22 calculates the time features of the agents passing through the intersection points according to the motion speed of the agents and the position of the intersection points, such as the estimated arrival time, passing time, etc. These time features provide an important basis for the subsequent dynamic right-of-way allocation. Finally, the feature compression unit 23 uses the space-time compression algorithm to map the three-dimensional space-time trajectory of the agent to a two-dimensional space-time feature map, reducing the computational complexity and improving the processing efficiency of the system.

[0045] S3 Evaluate the right-of-way priority of the agent at the intersection point, develop allocation strategies to handle conflicts and determine the passing order of the agent; Specifically, according to the time feature of the agent passing through the intersection obtained by the space-time feature extraction module 2, dynamic road right allocation is performed. The road right evaluation unit 31 comprehensively considers the motion state of the agent, the urgency of the task, and other factors to evaluate the road right priority of the agent at the intersection. For example, for an agent performing an emergency task, such as a medical rescue robot, a higher road right priority is given. The allocation strategy unit 32 formulates a reasonable road right allocation strategy according to the evaluation results, and uses priority sorting and other methods to ensure that the agents pass through the intersection in an orderly manner. The conflict resolution unit 33 handles possible road right conflicts through random allocation, agent number priority, and other mechanisms to ensure the uniqueness and feasibility of the allocation.

[0046] S4 monitors agent motion abnormalities and formulates recovery strategies to ensure stable operation of the system through a multi-level fault-tolerant mechanism; Specifically, the system monitors the motion process of the agent in real time. The abnormality monitoring unit 41 uses sensors and other devices to sense the speed, motion trajectory, and other information of the agent in real time, and discovers abnormal behaviors such as sudden speed changes and path deviations in a timely manner. Once an abnormality is discovered, the recovery strategy unit 42 immediately formulates a corresponding recovery strategy, such as re-planning the path or adjusting the speed. At the same time, the multi-level fault-tolerant unit 43 starts the fault-tolerant mechanism to ensure that the system can still operate normally when some agents are abnormal.

[0047] S5 manages the sharing of local information by the communication link of the agent, handles the negotiation process between agents, and supports collaboration and path adjustment; Specifically, the communication management unit 51 of the distributed collaboration communication module 5 establishes and maintains the communication link between agents to ensure that the agents can effectively exchange information. The information sharing unit 52 realizes the sharing of local information between agents, enabling the agents to understand the motion state and path planning information of surrounding agents. The negotiation processing unit 53 handles the negotiation process between agents. When conflicts or coordination are needed between agents, an agreed solution is reached through the negotiation mechanism.

[0048] S6 evaluates the path state, performs update operations on the affected segments through incremental optimization, and guides the agents to move according to the new path; Specifically, according to the road right changes of the dynamic road right allocation module 3 and the feedback information of the abnormality detection and recovery module 4, the motion path of the agent is updated in real time. The path evaluation unit 61 evaluates the state and feasibility of the current path, checks whether the path is unobstructed, whether there are obstacles, etc. The incremental optimization unit 62 performs incremental optimization on the affected path segments, and uses a local path adjustment algorithm to optimize and adjust the path. The update execution unit 63 sends the new path planning information to the agents to guide them to move according to the updated path.

[0049] Specifically, the performance monitoring unit 71 collects system performance data (total path length, maximum completion time, energy consumption, collision frequency, etc.) through a sensor network and data acquisition equipment by continuously monitoring its own performance indicators. The indicator evaluation unit 72 evaluates and analyzes these performance indicators, uses data analysis algorithms to identify the strengths and weaknesses of the system. The optimization adjustment unit 73 optimizes and adjusts the path planning algorithm, road right allocation strategy, and communication mechanism of the system based on the evaluation results to improve the overall performance of the system.

[0050] S8 manages resources, global scheduling coordination, and predictive control to adjust resource allocation in advance.

[0051] Specifically, the resource management unit 81 manages computing resources, storage resources, and communication resources uniformly and allocates resources according to the system's needs. The global scheduling unit 82 coordinates the motion timing and path planning of the agent and schedules and coordinates the motion of the agent from a global perspective to achieve overall optimization. The predictive control unit 83 adjusts resource allocation and path planning in advance based on the motion trend and historical data of the agent, uses a prediction algorithm (Kalman filter prediction algorithm) to predict the future motion state of the agent, makes adjustments in advance to avoid potential conflicts and congestion, and improves the foresight and adaptability of the system.

[0052] Advantages: 1. The modules work together to form a complete motion timing control link, ensuring the efficiency and accuracy of agent motion control. 2. The path perception module obtains the agent's motion path in real time and updates the scene. The real-time path update module dynamically adjusts the path based on road right changes, allowing the agent to respond to environmental changes and unexpected situations in a timely manner. 3. The spatiotemporal feature extraction module accurately identifies motion path intersection points and calculates the agent's passing time characteristics, providing a reliable basis for subsequent road right allocation. 4. The dynamic road right allocation module allocates road rights based on the order in which agents pass through intersection points, ensuring that agents pass through intersection points in an orderly manner and avoiding conflicts and congestion. 5. The anomaly detection and recovery module monitors agent motion anomalies in real time and triggers the recovery mechanism, improving the reliability and stability of the system. 6. The distributed cooperative communication module enables local information sharing and negotiation between agents, enhancing their collaboration capabilities. 7. The performance evaluation and optimization module continuously evaluates system performance and optimizes control parameters, allowing the system to continuously improve motion timing control effectiveness. 8. The central coordination module coordinates all functional modules and handles global constraints to ensure consistent and efficient operation of the entire system.

[0053] The above merely discloses a preferred embodiment of the multi-agent motion timing control method and system, and of course cannot limit the scope of the present application. Those skilled in the art can understand that all or part of the processes of the above embodiment can be implemented, and equivalent changes made according to the claims of the present application still fall within the scope of the present application.

Claims

1. A multi-agent motion timing control system, characterized in that, It includes a path awareness module, a spatiotemporal feature extraction module, a dynamic right-of-way allocation module, an anomaly detection and recovery module, a distributed cooperative communication module, a real-time path update module, a performance evaluation and optimization module, and a central coordination module, wherein the path awareness module, the spatiotemporal feature extraction module, the dynamic right-of-way allocation module, the real-time path update module, the anomaly detection and recovery module, the distributed cooperative communication module, the performance evaluation and optimization module, and the central coordination module are connected in sequence; The path perception module is used to acquire the path planning scenario of the intelligent agent group and the movement paths of all intelligent agents in real time. The spatiotemporal feature extraction module is used to identify the intersection points of motion paths and calculate the time features of the agent's passage. The dynamic right-of-way allocation module is used to allocate right-of-way at the intersection of intelligent agent paths in sequence. The anomaly detection and recovery module is used to monitor abnormal movements of the intelligent agent and trigger a recovery mechanism. The distributed cooperative communication module is used to realize local information sharing and negotiation among intelligent agents; The real-time path update module is used to dynamically adjust the movement path of the intelligent agent according to changes in right-of-way. The performance evaluation and optimization module is used to continuously evaluate system performance and optimize control parameters; The central coordination module is used to coordinate all functional modules and handle global constraints.

2. The multi-agent motion timing control system as described in claim 1, characterized in that, The path awareness module includes a data acquisition unit, a data fusion unit, a path recognition unit, and a scene update unit; The data acquisition unit is used to acquire real-time data related to the path planning scenario of the intelligent agent group and the motion paths of all intelligent agents; The data fusion unit is used to fuse the collected multi-source data. The path recognition unit is used to identify the movement path of the agent from the fused data; The scene update unit is used to update the path planning scene in real time based on the movement of the agent.

3. The multi-agent motion timing control system as described in claim 1, characterized in that, The spatiotemporal feature extraction module includes an intersection point identification unit, a temporal feature calculation unit, and a feature compression unit; The intersection point identification unit is used to identify intersection points on the movement path; The time feature calculation unit is used to calculate the time features of the agent passing through the intersection point; The feature compression unit is used to map a three-dimensional spatiotemporal trajectory into a two-dimensional spatiotemporal feature map using a spatiotemporal compression algorithm.

4. The multi-agent motion timing control system as described in claim 1, characterized in that, The dynamic right-of-way allocation module includes a right-of-way evaluation unit, an allocation strategy unit, and a conflict resolution unit; The right-of-way evaluation unit is used to evaluate the right-of-way priority of the agent at the intersection of paths; The allocation strategy unit is used to formulate a right-of-way allocation strategy and allocate right-of-way based on the evaluation results; The conflict resolution unit is used to handle conflicts during the right-of-way allocation process.

5. The multi-agent motion timing control system as described in claim 1, characterized in that, The anomaly detection and recovery module includes an anomaly monitoring unit, a recovery strategy unit, and a multi-level fault tolerance unit; The anomaly monitoring unit is used to monitor abnormal behavior during the movement of the intelligent agent in real time. The recovery strategy unit is used to formulate recovery strategies under abnormal circumstances; The multi-level fault-tolerant unit is used to implement a multi-level fault-tolerant mechanism.

6. The multi-agent motion timing control system as described in claim 1, characterized in that, The distributed collaborative communication module includes a communication management unit, an information sharing unit, and a negotiation processing unit; The communication management unit is used to manage the communication links between intelligent agents; The information sharing unit is used to realize local information sharing among intelligent agents; The negotiation processing unit is used to handle the negotiation process between intelligent agents.

7. A multi-agent motion timing control method, used in the multi-agent motion timing control system according to any one of claims 1-6, characterized in that, Includes the following steps: Real-time acquisition of path planning scenarios and related motion data of intelligent agent groups; after fusion processing, identification of motion paths and updating of scenarios; Based on the intersection points identified by path perception, the agent's transit time features are calculated, and a spatiotemporal compression algorithm is used to reduce computational complexity. Assess the right-of-way priority of agents at intersections, formulate allocation strategies, handle conflicts, and determine the order in which agents pass through. Detect abnormal movements of intelligent agents to formulate recovery strategies and ensure stable system operation through multi-level fault tolerance mechanisms; Manage the sharing of local information in the communication links of intelligent agents, handle the negotiation process between intelligent agents, and support cooperation and path adjustment; Evaluate the path status, perform update operations on the incremental optimization of the affected segments, and guide the agent to move along the new path; Monitor system performance indicators and evaluate, analyze, and optimize the system; Manage resources, coordinate and schedule globally, and predict and control to adjust resource allocation in advance.

Images