Research on Gravity Compensation Algorithm for Six-axis Force Sensor based on Unscented Kalman Filter

Abstract

To address the problem of inaccurate measurements caused by the gravity effect on six-axis force sensors in human-robot interaction, this paper proposes a method for error correction of six-axis force sensors based on Unscented Kalman Filter (UKF) algorithm. By processing and estimating the sensor data, the proposed method achieves accurate estimation and correction of sensor errors, reducing the sensor error to between 0-0.175N. Experimental results demonstrate that the proposed method significantly improves the measurement accuracy of six-axis force sensors, providing reliable technical support for precise force control in human-robot interaction. The contribution of this study is to provide a reliable and efficient method to enhance the accuracy of six-axis force sensors.