Using Bearing-Only Measurements of an Underwater Observer Robot for Passive Tracking of a Highly Maneuvering Target Using Interactive Multiple-Model Cubature Kalman Filter
Mohsen Ebrahimi, Seyed Mohammad Mehdi Dehghan Bonadaki · 2023
Tracking of underwater moving targets by underwater robots requires the continuous use of passive observation of the target angle over time. In this case, the improvement of target position estimation depends on the movement of the submarine robot to increase the field of view. In order to protect the observer bot from being seen by the other party, passive BOT measurements are used in this paper. Since the motion of a target may have several models, one of the best choices is the interactive multiple model (IMM) filter. On the other hand, due to the non-linearity of the measurement equations and the target motion equations, the Cubature Kalman filter (CKF) can be used to improve the tracking accuracy. By combining these filters, in this paper, the IMM- CKF filter is used to track a highly maneuverable target in a situation where there is only one measurement from the observing robot at any time.