Abstract: Target Motion Analysis(TMA) is one of the most popular techniques for estimation of time-varying state of underwater targets.Two types of nonlinear filters,Extended Kalman Filter(EKF) and Unscented Kalman Filter(UKF),are applied to the monostatic and bistatic bearing-only tracking.The tracking performance of the two filters is demonstrated via Monte Carlo simulations.The results show that the UKF outperforms in most cases in the underwater passive tracking scenarios,especially if the target makes maneuvers frequently.Compared with EKF,UKF shows better numerical stability and higher estimation accuracy with little cost of computational complexity.