基于计算机视觉的运动目标轨迹识别预测方法
By generating a passable path and modulating the sampling weight using a confusion parameter, the uncertainty in trajectory prediction of autonomous driving systems under complex road conditions is resolved, stable trajectory output is achieved, and the robustness and accuracy of the system are improved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- BEIJING ZHONGHAIJICHUANG SCI TECH DEV
- Filing Date
- 2026-01-20
- Publication Date
- 2026-07-17
AI Technical Summary
Existing autonomous driving systems face challenges when dealing with complex road conditions, including systemic interference caused by map building errors, endpoint drift disrupting the stability of topological connections, difficulties in path identification caused by shared guide segments, and output uncertainty issues brought about by random sampling mechanisms. These challenges make it difficult to accurately predict the future trajectories of traffic participants.
By acquiring the observation trajectory points of the moving target and road data containing map error information, multiple passable paths are generated. The spatial tolerance range is defined by combining the map error information, the shared guiding road segments between passable paths are identified, and the sampling weights are modulated using the confusion parameter. A sampling consistency fitting operation is performed, and the optimal motion model instance is selected for prediction.
It effectively solves the path identification problem caused by endpoint drift and overlapping guide lines at complex intersections, improves the robustness of the prediction system, eliminates the output jitter caused by traditional random sampling, and achieves stable trajectory output that meets the requirements of mass production-level control systems.
Smart Images
Figure CN121963140B_ABST