An IMM-UKF Aided SINS/USBL Calibration Solution for Underwater Vehicles

Abstract

To improve the efficiency of the calibration process and enhance the adaptability of the calibration method, an interacting multiple model and unscented Kalman filter (IMM-UKF) aided strapdown inertial navigation system (SINS)/Ultra-Short Base Line (USBL) calibration solution is proposed for the common calibration procedure, where only one transponder is employed without pre-locating. Firstly, construct the SINS/USBL calibration mechanism, where the transponder's position, level-arm and misalignment angle are estimated simultaneously utilizing the slant range, inclination angles of USBL and depth information from depthometer. Second, to mitigate the decreasing calibration accuracy caused by the single filtering parameter under the complex underwater environment, an IMM-UKF algorithm is presented to support the proposed calibration mechanism. Simulation results indicate that the proposed mechanism earns faster convergence rate and better calibration results than other solutions. Besides, the proposed solution can maintain its robustness when the observation quality changes.