Nonlinear static decoupling of six-dimension force sensor for walker dynamometer system based on artificial neural network

Abstract

The static coupling of six-dimension force sensor for walker dynamometer system is a key factor to limit its measuring precision. A new decoupling method based on artificial neural network is proposed in this paper. Relevant error check results shows that, after the calibration by using the Back Propagation neural network and Radial Basis Function neural networks, the maximal system precision error with single-direction force was 7.78% and 4.33% and the maximal crosstalk was 7.49% and 6.52%,respectively. In comparison with traditional linear calibration method, the proposed technique can effectively increase the measurement accuracy of walker loads and greatly decrease the coupling effect.