一种人工智能辅助的智能控制决策方法
By constructing a motion intention prediction model for obstacle groups using deep reinforcement learning and graph neural networks, the uncertainty problem of path planning in dynamic obstacle environments is solved, enabling real-time optimization of robot paths and improved safety.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- GUANGDONG JIONG TECH CO LTD
- Filing Date
- 2025-08-25
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies struggle to accurately predict the movement intentions of obstacle groups in dynamic obstacle environments, leading to uncertainty and inefficiency in path planning. In particular, the effectiveness of robot path selection is affected when faced with sudden changes in the movement patterns of obstacle groups and their mutual influence.
A deep reinforcement learning algorithm is used to construct a motion intention prediction model for obstacle groups. Combined with graph neural network analysis, the interaction relationship between obstacles is analyzed. By generating a set of candidate paths and using the A* algorithm to determine the optimal path, the robot path can be optimized in real time.
It enables accurate prediction of the movement of obstacle groups in complex dynamic environments and real-time optimization of robot paths, improving the safety and efficiency of robot navigation in dynamic environments.
Smart Images
Figure CN121209570B_ABST