States Estimation for Unmanned Surface Vehicles (USVs) Using Dual Unscented Kalman Filters
Yifang Wang, Xiao Yang He, Donghang Liu, Jiankai Qin, Peng Li · 2021
For future Unmanned Surface Vehicles (USVs), the accuracy of state estimation and speed of tracking are vital in complex environments. However, modelling the USV contains a large number of parameters, some of which are difficult to estimate. This brings a challenging task for USV applications. In this paper, we consider the USV modelling with unknown parameters, and propose a novel Dual Unscented Kalman Filter (DUKF) scheme. To this end, the two parallel UKFs are used for simultaneous estimation of both states and parameters. Simulation results show the superiorities of the DUKF method for USV state estimation, with higher accuracy and faster speed of convergence compared to traditional filter schemes.