A New Information Fusion Method for Underwater Vehicles with Acoustics Communication Delays and Packet Losses

In this paper, we mainly investigate the information fusion problem for underwater vehicles in the presence of acoustic communication delays and packet losses. Due to time delays and packet losses in acoustic communication in underwater environments, different sensors on underwater vehicles may have varying measurement information, making the design of an information fusion strategy challenging. To solve this problem, we first develop a local optimal estimator for each sensor on the underwater vehicle. Furthermore, the estimation error cross-covariance matrices between any two local estimators are derived, and the global weighted fusion filter is given to minimize the variance of the overall state estimation. Moreover, the stability conditions of the global weighted fusion filter are proposed. Finally, a numerical example is presented to demonstrate the effectiveness of the information fusion algorithm.