Improving wrist force/torque sensor measurements through a data fusion technique

The authors present a real-time estimation scheme for improving wrist force/torque sensor measurements through a data fusion technique. The currents of joint servomotors are measured as a redundant data source for data fusion. The computational results of the model-based control method are used to reduce the extra amount of computation required for estimation. Robot joints are fully instrumented to obtain necessary dynamic information about robot motion. Simulation results reveal that the performance of the estimation scheme depends on the sampling period of the robot controller and the a priori knowledge of the measurement noise processes.<>