Application of Kalman filter to bearing-only target tracking system

With modified polar coordinates, a specific method to realize bearing-only target tracking by means of an extended Kalman filter is discussed in this paper. By using covariance square root filtering and fading memory filtering with variable coefficients, the problems of computation and filtering divergence may be overcome, and a stable and asymptotically unbiased estimation is obtained satisfactorily. A real time tracking processing is achieved using an 8098 single chip computer.