Efficient algorithms of clustering adaptive nonlinear filters

This paper proposes a new class of efficient adaptive nonlinear filters whose estimation error performance (in a minimum mean square sense) is superior to that of competing approximate nonlinear filters, e.g., the well-known extended Kalman filter (EKF). The proposed filters include as special cases both the EKF and previously proposed partitioning filters. The new methodology performs an adaptive selection of appropriate reference points for linearization from an ensemble of generated trajectories that have been processed and clustered accordingly to span the whole state space of the desired signal. Through a series of simulation examples, the approach is shown significantly superior to the classical EKF with comparable computational burden.