Adaptive underwater target tracking using passive multipath time-delay measurements

This paper examines the problem of adaptively tracking in range and velocity an underwater maneuvering target using passive time delay measurements. The target can make large scale random velocity and depth changes at times which are unknown to the observer. Tracking is accomplished by making use of the basic linearized polar model of target and observer motion previously developed [1]. Now, however, a nonlinear system block has been added to the tracking system [2], [3], which leads to two major benefits. First, the need for extended Kalman filters is eliminated making the passive tracking system more robust than it was previously [3]. The second benefit is a partial decoupling of depth estimation from the polar range estimator, which considerably reduces the computational level of the adaptive tracking system.