On strong tracking Kalman filter based on forgetting factor dynamic optimization

The fixed forgetting factor of state estimation error variance matrix in the strong tracking Kalman filter algorithm with suboptimal multi-fading factors is very important. When it takes a value too small, the role of current information will be emphasized excessively. As a result, it is possible to cause the time variant fading factors to regulate state estimation overly. Conversely, the role of current information will be weakened relatively. In fact, the optimum filter tracking performance can not be achieved in either case. This paper proposed an estimation method of error variance matrix on the basis of fuzzy forgetting factor. This method regulates fuzzy forgetting factor according to fuzzy rules, by which the fuzzy logic controller monitoring fuzzy similarity coefficient and state estimation variance. And then adjusts suboptimal multiple fading factors to improve the tracking precision of the filter in the strong tracking Kalman filter algorithm. The effectiveness of the algorithm is proved by the simulation results.