Multi-legged robot motion control and inspection path planning method and system for curved surface environment
By employing a multi-legged robot motion control method, combined with depth camera fusion edge detection and cubic Bézier curve planning, the robustness and control strategy safety issues of visual navigation and positioning on wind turbine blade surfaces were resolved. Stable inspection path planning and closed-loop control were achieved, adapting to complex curved surface environments.
Patent Information
- Application Number
- CN202610784600.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Filing Date
- 2026-06-02
- Publication Date
- 2026-07-17
AI Technical Summary
Existing technologies for visual navigation and positioning on wind turbine blade surfaces lack robustness, adaptability to unknown environments, safety of control strategies, and closed-loop feedback capabilities. This leads to problems such as pose estimation drift, path drift, accumulated deviation, collisions, and missteps that can easily occur during robot inspections.
A multi-legged robot motion control method is adopted, which acquires RGB images and depth images through a depth camera, extracts surface edge points by fusing edge detection, plans the foot trajectory using cubic Bézier curves, and performs closed-loop control by combining visual deviation closed-loop feedback and contact torque feedback to achieve stable inspection.
It improves the robot's motion adaptability and safety in complex curved surface environments, reduces collision risks, achieves stable path tracking and navigation functions, and enables autonomous mapping and path planning in unknown environments.
Smart Images
Figure CN122411488A_ABST