Multi-sensor fusion in Kalman-filter for high performance force sensing

Sensorless force sensation by disturbance observer has been widely employed in numerous applications due to its superiority to the measurement by a force sensor. This paper introduces the development of the disturbance observer to obtain the high performance force sensing with a wideband force sensation. In this paper, a multi-sensor data fusion by Kalman-filter algorithm is exploited for velocity estimation which plays the role of an input of the disturbance observer. The combination of multi-sensor-based Kalman-filter and the disturbance observer provides the enhanced force sensing performance and the effective noise reduction. The proposed method is implemented in FPGA with the sampling period of 5 µs. Experimental results confirm the feasibility of the proposed method.