Identification of roll dynamics of an autonomous underwater vehicle

• Published in: Proceedings of the Institution of Mechanical Engineers. Part M, Journal of engineering for the maritime environment Journal of engineering for the maritime environment, 2019-11, Vol.233 (4), p.1056-1067
• Main Authors: Abtahi, Seid Farhad, Alishahi, Mohammad Mehdi, Azadi Yazdi, Ehsan
• Format: Article
• Language: English
• Subjects: Approximation; Autonomous underwater vehicles; Coefficients; Control stability; Control systems design; Controllers; Empirical analysis; Hydrodynamic coefficients; Hydrodynamics; Identification; Parameter identification; Parameters; Particle swarm optimization; Proportional integral derivative; Robustness (mathematics); Stability analysis; Torque; Underwater exploration; Underwater vehicles
• ISSN: 1475-0902; 2041-3084
• DOI: 10.1177/1475090218810248

This article aims to identify the roll channel parameters of an autonomous underwater vehicle. These parameters include hydrodynamic coefficients, motor torque, eccentricity, misalignments and mounting imperfections. In the proposed method, an approximation of the hydrodynamic coefficients is made at first via semi-empirical methods. In the next step, a proportional–integral–derivative controller is designed with respect to the approximated coefficients. Since the approximations can be very uncertain, the robustness of stability and performance of the proportional–integral–derivative controller is evaluated throughout µ-analysis. Finally, the unknown parameters are identified using the recorded data of on-board sensors during motion of the vehicle. The identification is based on minimization of the one-step prediction error. The minimization problem is nonlinear in unknown parameters, and particle swarm optimization is used to find an optimal solution. The performance of the proposed method is exhibited through a 6-degrees-of-freedom simulation of an autonomous underwater vehicle.