Device cluster scheduling method and system based on multi-agent cooperation

CN122132609APending Publication Date: 2026-06-02UNIV OF ELECTRONIC SCI & TECH OF CHINA CHENGDU COLLEGE
View PDF 5 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
UNIV OF ELECTRONIC SCI & TECH OF CHINA CHENGDU COLLEGE
Filing Date
2026-05-06
Publication Date
2026-06-02

AI Technical Summary

Technical Problem

Existing multi-drone scheduling methods lack flexibility when facing complex urban environments and changing inspection needs, making it difficult to quickly adjust tasks, resulting in low inspection efficiency. Furthermore, the system is prone to paralysis when the central control unit fails or communication is interrupted, leading to serious task conflicts and duplicate inspections.

Method used

A device cluster scheduling method based on multi-agent collaboration is adopted. By equipping each UAV with a local scheduling agent unit, distributed task decomposition and dependency sharing are realized. The inspection path is optimized by using task topology sorting and migration negotiation protocol, reducing the computational burden of the central control unit and forming a distributed task dependency sharing network.

Benefits of technology

It improves the reliability and stability of the drone inspection system, avoids task conflicts, maximizes resource utilization, enables rapid response to emergencies, optimizes inspection paths, and improves inspection efficiency.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122132609A_ABST
    Figure CN122132609A_ABST
Patent Text Reader

Abstract

This invention provides a device cluster scheduling method and system based on multi-agent collaboration, relating to the field of UAV scheduling technology. First, the initial task requirement set for each UAV is obtained. Then, the initial task requirement set is input into a local scheduling agent unit for task decomposition, resulting in an initial atomic task set. A directed acyclic graph (DAG) structure for task execution is generated based on the atomic task dependency parameters and broadcast, forming a distributed task dependency sharing network. In this network, the local scheduling agent unit performs task overlap analysis and invokes a task migration negotiation protocol to negotiate task migrations, generating a task migration negotiation result. The DAG structure is updated based on the negotiation result, resulting in an optimized structure. Finally, a sequence of task execution instructions is sent to the UAV execution mechanisms according to the topological order of the optimized structure. This invention improves the efficiency, reliability, and flexibility of multi-UAV collaborative inspection.
Need to check novelty before this filing date? Find Prior Art