Calibration method of DVL based on position observation information

The accuracy of underwater navigation is significantly impacted by the scale factor error and misalignment angle of the Doppler Velocity Log (DVL). A DVL calibration method based on position observation information is proposed as a solution to the integrated navigation system composed of compass, DVL, ultra-short baseline acoustic positioning system, and depth sensor. A transfer model from DVL error parameters to the positioning error of Dead Reckoning (DR) is established. Based on this model, dual-loop Kalman filter (KF)-based and particle swarm optimization (PSO)-based DVL error parameter estimation algorithms are proposed. Compared with classical methods, the proposed methods achieve high-precision DVL calibration in deep-water environments at a lower cost. Simulation and sea trials show that both proposed algorithms exhibit excellent stability, consistency, and accuracy. After calibration, the positioning errors of DR have decreased from 1.56% to 0.28% and 0.23% of the voyage, respectively.