动态环境下基于深度强化学习的智能体多目标导航方法
By introducing a behavior urgency prediction module and the Soft Actor-Critic algorithm, combined with Nash equilibrium game theory, the navigation decision-making of the intelligent agent is optimized, solving the problem of poor navigation performance in complex and dynamic crowd environments and improving navigation safety and efficiency.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHANGCHUN UNIV OF TECH
- Filing Date
- 2026-04-29
- Publication Date
- 2026-07-17
AI Technical Summary
Existing intelligent agent navigation methods lack foresight in predicting pedestrian behavior and decision-making processes in complex and dynamic crowd environments, resulting in poor navigation performance, insufficient robustness, and inadequate social acceptability.
By introducing a behavior urgency prediction module, combined with the Soft Actor-Critic algorithm and Nash equilibrium game theory, the navigation decision of the intelligent agent is optimized, enabling forward-looking prediction and safety constraints of pedestrian behavior. A multi-objective reward function is constructed to improve navigation efficiency and social acceptability.
It improves the navigation safety and efficiency of intelligent agents in dynamic crowd environments, enhances the ability to respond to changes in pedestrian behavior and the robustness of the system, and enables flexible decision-making in complex environments.
Smart Images
Figure CN122113690B_ABST