A six-dimensional force sensor decoupling method based on humanoid robots
By constructing static decoupling and compensation matrices to eliminate interdimensional cross-interference and dynamic interference of the six-dimensional force sensor of the humanoid robot, and combining recursive filtering technology, the problem of signal distortion of the six-dimensional force sensor is solved, realizing high-precision force signal decoupling and accurate recognition by the motion controller, thus improving the control performance of the humanoid robot.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN LIGENT SENSOR TECH CO LTD
- Filing Date
- 2026-03-16
- Publication Date
- 2026-05-26
AI Technical Summary
The six-dimensional force sensor of the humanoid robot is subject to cross-interdimensional interference and dynamic interference when collecting three-dimensional force and three-dimensional torque signals, which leads to signal distortion and affects the motion controller's accurate identification of external forces and torques, thus failing to meet the requirements of high-precision motion control.
By constructing a static decoupling matrix to eliminate interdimensional cross-interference, and combining the kinematic model of the humanoid robot and the data from the inertial measurement unit, a compensation matrix is generated and recursive filtering is performed on each channel to suppress high-frequency vibration noise and motion inertial coupling residuals, and output a high-precision decoupling force signal.
It achieves high-precision decoupling of force signals from humanoid robots, ensuring that the motion controller can accurately identify external forces and torques, improving the robot's high-precision control performance and adapting to the force perception requirements in complex motion scenarios.
Smart Images

Figure CN122077633A_ABST