Optimization of Capacitive Six-Axis Force/Torque Sensors Based on Error Analytical Model

The measurement accuracy of capacitive six-axis force/torque (F/T) sensors is difficult to predict and determine during the design process, which leads to the uncertainty of the actual measurement accuracy of the sensor. In this article, a sensor accuracy prediction method based on an error analysis model is proposed. A mathematical model for mapping the six-axis F/T to changes in capacitance is established by computing the overall flexibility matrix of the elastic structure. The influence mechanism of various errors on the measurement accuracy of the six-axis F/T sensor system is deeply analyzed, and a mathematical model for error prediction and a comprehensive performance evaluation index function is established. The comprehensive performance evaluation index function is proposed, coupled with applying a nonlinear programming genetic algorithm for parameter optimization. A prototype was fabricated using the optimization results and calibration experiments were conducted to verify its accuracy. The experimental results demonstrate that the nonlinearity error of the prototype is 0.280% full scale (FS), and the calibration measurement accuracy is 0.601%FS. The design method proposed in this article effectively enables the development of high-precision capacitive six-axis F/T sensors.