Asynchronous multi-rate sensor information fusion algorithm based on time-varying noise
Rui Li · Journal of Naval University of Engineering · 2009
An information fusion algorithm was presented based on a class of asynchronous multi-rate and multi-sensor system,in which the observation noise from each sensor is unknown and time-varying.The fused state estimate was obtained by skillful use of information fusion theory of asynchronous multi-rate sensors and the approach for fusing time-varying noise based on synchronous single-rate sensor system.This approach could be adopted in asynchronous multi-rate sensors in time-varying linear system.By theoretical analysis,the dynamic system model of asynchronous multi-rate sensors was built for the synchronized single-rate sensors.Then the noise statistics estimator and adaptive Kalman filter were used for state estimation, and the federal filter for information integration.Thus the optimal estimation was gained based on all the observations.Theoretical analysis and simulation results show the effectiveness of the proposed algorithm.