Nonlinear distributed tracking with probabilistic data association

A nonlinear distributed tracking technique using probabilistic data association to account for the uncertainty in the origin of the measurement data is presented. In this distributed tracking structure there is a coordinator which contains the model of a nonlinear random process (state of the target in track), and an arbitrary number, N , of spatially distributed sensors taking observations of the target´s state. The coordinator must reconstruct the (global) probability density of the target´s state conditioned on the observation histories of the sensors. However, the coordinator can only access the N (local) conditional densities produced by local processing of the sensor´s returns, not the observation histories themselves