An all-neighbor fuzzy association approach in multisensor-multitarget tracking systems

This paper proposes the first all-neighbor fuzzy logic data association approach in distributed multisensor-multitarget (MSMT) tracking systems. The proposed approach is developed based on fuzzy clustering means (FCM) algorithm. This fuzzy clustering algorithm determines the grade of membership of each received data point in each fuzzy cluster. Unlike all fuzzy logic data association algorithms, which assign only one observation to each track according to some association measure, the proposed all-neighbor fuzzy logic data association approach incorporates-all observations within the gate of the predicted target state to update the state estimate using a degree of membership weighted sum of innovations. To demonstrate the feasibility, efficiency, and simplicity of the proposed approach to perform data association in multisensor-multitarget environment, it is applied to an example of a four-dimensional tracking system. The performance of the proposed approach is evaluated using Monte Carlo simulations. Its performance is also compared to the performance of the nearest neighbor standard filter (NNSF) and perfect data association. The results show that the proposed approach is very efficient.