Performance evaluation of a fuzzy data association algorithm for multitarget tracking (MTT)

In this paper, a performance comparison of a recently developed fuzzy data association algorithm for multitarget tracking (MTT) with the well known joint probabilistic data association (JPDA) algorithm is presented. In this scheme, a fuzzy logic multiple models algorithm is constructed first. It uses different target models like constant velocity, constant acceleration etc. to describe all the states of the system. A Kalman filter is set up for each model to estimate their states. The final state estimate is a weighted average of the model conditioned estimates with the fuzzy reasoning. Based on this algorithm and after constructing the corresponding rule set, a fuzzy data association algorithm is developed, which uses full states, prior knowledge and experience. The simulation scenario considers both the fuzzy and JPDA algorithms for tracking two and four targets in a two dimensional setting. Based on the simulation results, the advantages and disadvantages of both the approaches for MTT are presented.