Adaptive UKF Method with Applications to Target Tracking
Han Chong · Acta Automatica Sinica · 2011
To improve low filtering precision and divergence caused by unknown system noise statistics in target tracking, an adaptive UKF(Unscented Kalman filter)is proposed.In the filtering process,by introducing the modified Sage-Husa noise statistic estimator,the new algorithm can estimate the statistical parameters of unknown system noises online and restrain the filtering divergence.Therefore,the filter numerical stability is effectively improved and the state estimation error is reduced. Simulation results show that compared with the standard UKF algorithm the proposed algorithm provides better accuracy and stability for target tracking.