A 7-degree-of-freedom humanoid arm adaptive parameter identification and compliant control method
By combining adaptive particle swarm optimization and extended Kalman filter state estimator, the problems of insufficient accuracy and anti-interference ability of the 7-DOF humanoid arm dynamics model are solved, achieving high-precision external force estimation and stable control, and reducing equipment costs.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- GUANGZHOU HUARUI ZHIPU INTELLIGENT ROBOT CO LTD
- Filing Date
- 2026-05-25
- Publication Date
- 2026-07-21
AI Technical Summary
Existing sensorless force estimation methods based on dynamic residuals suffer from insufficient accuracy of the dynamic model, poor anti-interference ability, and module fragmentation in 7-DOF humanoid arms, making it difficult to meet the requirements of high accuracy and stability.
The excitation trajectory is designed using Fourier series with adaptive particle swarm optimization. Combined with extended Kalman filter state estimator and closed-loop cooperative compliant control, the real-time estimation and stable control of the end force are achieved through adaptive dynamic identification and Jacobian pseudo-inverse solution.
It achieves high-precision external force sensing without the need for additional sensors, reduces equipment costs, improves system adaptability and control accuracy, and meets the needs of industrial applications.
Smart Images

Figure CN122425702A_ABST