一种应用于机场自动驾驶牵引车的动态边界扩展的方法
By constructing an initial point cloud map in the airport autonomous driving system and performing real-time registration and difference region extraction, combined with multi-source data fusion and confidence verification, the problem of traditional systems being unable to adapt to dynamic obstacles is solved, achieving efficient dynamic environment response and resource optimization.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHENGDU TONGGUANG NETLINK TECH CO LTD
- Filing Date
- 2025-06-26
- Publication Date
- 2026-07-17
AI Technical Summary
Traditional airport autonomous driving systems cannot adapt to dynamic obstacles in real time, have long map update cycles, insufficient perception capabilities of individual vehicles, and difficulty in distinguishing between temporary obstacles and permanent structures, leading to path planning errors.
An initial point cloud map is constructed using LiDAR SLAM, and the vehicle system performs real-time registration and difference region extraction. Combined with semantic camera verification, multi-source data fusion is performed using DS evidence theory to dynamically update the map. Temporary boundaries are verified using a sliding time window and confidence accumulation mechanism to generate dynamic obstacle boundaries.
It achieves sub-second response capability to dynamic airport environments, reduces map update latency from hours to minutes, lowers false alarm rate to 0.1%, reduces data transmission volume by 70%, and optimizes resources.
Smart Images
Figure CN120778094B_ABST