Pseudo-linear estimator for bearings-only passive target tracking

The pseudo-linear extended Kalman tracking filter previously developed for passive target tracking using bearings-only measurements required the use of a null target state vector as an initial estimate to obtain convergence for all types of scenario. The pseudo-linear estimator is projected in such a way that it does not require any initial estimate at all and at the same time offers all the features of the extended Kalman filter based pseudo-linear filter; namely sequential processing, flexibility to adopt the variance of each measurement, etc. The algorithm is tested in Monte Carlo simulations and its results are presented for two typical scenarios at various noise levels.