In this paper, a new multi-target trackingalgorithm based on fuzzy logic for tracking in clutter isdeveloped, it is called directional fuzzy data association(DFDA) filter. The new algorithm incorporates thedirectional information of the targets for data associationwith the Mahalanobis distance. Firstly, the directionalinformation, called pseudo-direction, is defined; the methodof how to calculate the pseudo-direction has been introduced.Then the state incorporating with the pseudo-direction isupdated using the cubature Kalman filter (CKF). At last thefuzzy logic inference method is used for data association.Simulation results are used to evaluate the performance ofthis new algorithm comparing with the nearest neighborstandard filter (NNSF) and joint probability dataassociation filter (JPDAF), the final results show that theproposed DFDA filter an efficient and effective approach forreal application.
展开▼