A Novel USV Cooperative Positioning Algorithm Based on Adaptive Filter

To address the issue of decreased positioning accuracy caused by the heavy-tailed distribution of underwater acoustic ranging in the Unmanned Surface Vessels (USVs) system, which is related to cooperative positioning, this paper first analyzes the kinematic and measurement models of USVs based on underwater acoustic ranging. Based on the foregoing, a cooperative positioning algorithm for USVs using Variational Bayesian Kalman Filtering was developed. The positions of USVs were determined via an iterative method, and offline data from actual ship experiments were utilized to confirm the effectiveness of the proposed algorithm. The outcomes indicate that the positioning errors of the algorithm are greatly reduced: compared with the Unscented Kalman Filter (UKF), Hybrid Interacting Multiple Model Filter (HIMM Filter, corrected from “HIDDF”), and Federated Kalman Filter (FKF, corrected from “FGMC”) algorithms, its Root Mean Square Error (RMSE) is reduced by 58%, 14%, and 15% respectively, which greatly improves the positioning performance of USVs.