An intention game-based multi-robot cooperative scheduling optimization method and device

By constructing an intention game model and introducing reinforcement learning optimization, the problem of collaborative scheduling of multi-robot systems in dynamic environments is solved, achieving efficient collaborative decision-making and task completion, reducing conflict and competition, and improving the overall performance of the system.

CN122411567APending Publication Date: 2026-07-17BEIJING UNIV OF POSTS & TELECOMM
View PDF 0 Cites 0 Cited by

Patent Information

Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2026-04-20
Publication Date
2026-07-17

AI Technical Summary

Technical Problem

Existing multi-robot scheduling methods lack unified intent game modeling at the high-level decision-making stage, making it difficult to predict resource competition and spatiotemporal conflicts in dynamic environments, which affects the system's real-time response capability and task completion efficiency.

Method used

An intention game model is constructed by acquiring robot operation data and task distribution data to establish a multi-agent intention game model. The robot strategy is optimized using reinforcement learning algorithms to achieve a unified model of cooperative and competitive relationships. The model is then iteratively optimized through reinforcement learning algorithms until a preset equilibrium condition is reached.

Benefits of technology

It significantly improves the collaborative scheduling efficiency, task completion stability, and environmental adaptability of multi-robot systems in dynamic and complex environments, while reducing path conflicts and resource competition.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN122411567A_ABST
    Figure CN122411567A_ABST
Patent Text Reader

Abstract

The application provides a multi-robot cooperative scheduling optimization method and device based on intention game, and the application comprises the following steps: obtaining multi-robot operation and task distribution data, and constructing intention description data containing intention role, intention task and self state characteristics; establishing a multi-agent intention game model based on the intention description data, and constructing an utility function for depicting cooperative, competitive and conflict relationships; iteratively optimizing the game model through reinforcement learning, and obtaining a joint intention result satisfying an equilibrium condition; mapping the joint intention to an execution strategy to control the robot to complete the work, and updating environment data for the next period of scheduling. The application can improve the cooperative efficiency of the multi-robot, reduce task conflicts and enhance the decision robustness in a complex environment.
Need to check novelty before this filing date? Find Prior Art