巡轨机器人动态避障控制方法
By integrating data from multiple sensors and active physical probing, the robot achieves accurate identification and control of obstacle states, solving the problem of high misjudgment rates in existing technologies and improving the safety and efficiency of inspections.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- HANGZHOU GONGSHU DISTRICT EDGE INTELLIGENCE INNOVATION RESEARCH INSTITUTE
- Filing Date
- 2026-06-17
- Publication Date
- 2026-07-17
AI Technical Summary
Existing track-inspecting robots cannot effectively distinguish the actual presence of obstacles under complex track surface conditions, resulting in a high misjudgment rate and affecting inspection efficiency and safety.
By fusing laser point cloud data, visual image data, and thermal distribution data, an obstacle fusion feature set is constructed. Through a pre-trained target recognition model and active physical probing, the physical state classification of obstacles is obtained, and matching local control trajectories and execution strategies are generated.
Accurately determining the actual state of obstacles avoids misjudgment, improves the safety and efficiency of inspections, and enhances the accuracy and applicability of obstacle handling and control.
Smart Images
Figure CN122411518A_ABST