Decoupled Adaptive Tracking Algorithm for Multi-Sensor Measurement Fusion

Abstract

A decoupled adaptive tracking filter is developed for centralized measurement fusion to track the same maneuvering target to improve the tracking accuracy. The proposed approach consists of a dual-band Kalman filter and a two-category Bayesian classifier. Based upon data compression and decoupling techniques, two parallel decoupled filters are obtained for lessening computation. The Bayesian classification scheme is employed which involves switching between high-level-band filter and low-level-band filter to continuously resist different target maneuver turns. The simulation results are presented which demonstrate the effectiveness of the proposed method.