Synthetic aperture sonar movement estimation - the adaptive Kalman filter approach

Inertial navigation systems include IMU, DVL, DGPS, and deep sensors. The movement of underwater towed fish can be considered to stochastically disturb the known heading and velocity of the surface ship, and to be influenced by sea flow, so we must consider estimation of the unknown maneuvering input and sea flow. For this reason, we present a new estimation algorithm, which is adequately applied to estimate the abrupt change of input. The approach consists of a Kalman filter without an input term, and the other is the adaptive forgetting factor RLSE. From numerical simulation, we can conclude that as the unknown varies over a long time interval, the proposed method is robust.