Calibration and optimization of non-linear 6D force/torque cells

Multi-dimensional force/torque sensors have a wide range of applications in the field of automation technology and in the steadily growing field of robotics. Through the development of new applications, the demands on multi-axis force/torque sensors and their calibrations are growing. Therefore, the focus of our research target lies in the optimization of the sensor’s measuring cell geometries and strain gauge positioning. Further our intention is to create a calibration that is resistant to ageing and temperature with maximized application accuracy using FEM-Models and Machine Learning. In this paper, we investigate the requirements for AI calibration procedures of six-dimensional force/torque sensors, also against the background of generating an optimal calibration model.