The Fractional Kalman Filter Based Asynchronous Multi-Rate Sensors Information Fusion

Abstract

A fractional Kalman filter based multi-rate sensors fusion algorithm is presented to fuse asynchronous multirate measurements of robotic systems. The fractional Kalman filter is introduced to improve approximation effect of system state. Based on the characteristics of multi-rate and delay measurement, the state is re-estimated at the time when the delayed measurement occurs by using weighted fractional Kalman filter, and then the state estimation is updated at the current time when the delayed measurement arrives following the similar pattern of Kalman filter. This method has high computation efficiency and estimation accuracy. The simulation examples are given to illustrate the effectiveness of the proposed fusion method.