Skip to Main Content
An intelligent filter with fuzzy gain is proposed, which calculates a manoeuver input into the time-varying variance of the overall process noise and compensates an estimation error by using a fuzzy gain. The proposed technique not necessary to predesign the acceleration levels according to the target manoeuver properties. The manoeuver input is esitmated using the difference between the non-manoeuvering filter and the manoeuvering one. To compensate the estimation error, the fuzzy gain is used to describe the relationship between the filter residual and its variation. A fuzzy system is utilised for an universal approximator. It is optimised by using a genetic algorithm. A computer simulations show that the effectiveness of the proposed method is compared with multiple filters.