Nonlinear modeling and parameter identification for AUV dynamics.

• Ano de defesa: 2019
• Autor(a) principal: Cárdenas Vivanco, Persing Junior
• Tipo de documento: Tese
• Tipo de acesso: Acesso aberto
• Idioma: eng
• Instituição de defesa: Biblioteca Digitais de Teses e Dissertações da USP
• Palavras-chave em Português: AUV; Coeficientes hidrodinâmicos; Dynamics of underwater vehicles; Hidrodinâmica; Hydrodynamic coefficients; Identificação de parâmetros; Model parameter identification; Nonlinear dynamic systems; Sistemas dinâmicos; Submersíveis não tripulados
• Link de acesso: http://www.teses.usp.br/teses/disponiveis/3/3152/tde-13022020-092242/

The dynamics of an underwater vehicle is intrinsically nonlinear. Hence, when it is desirable to reproduce various types of maneuvers of an underwater vehicle, such as straight line, zigzag and turning in circles; a nonlinear mathematical model is required. Usually, for vehicles such as an autonomous underwater vehicle (AUV), the dynamics is modeled as a decoupled-nonlinear system which considers that the vehicle is maneuvering in the horizontal plane or in the vertical plane. This work proposes a new method for identification of the AUV Dynamics combining an analytical and semi-empirical (ASE) approach and a system identification approach to estimate the hydrodynamic coefficients related to the horizontal maneuvering of the vehicle. The ASE method is used to initialize the parameter estimation process, assuming a simplified geometry of the AUV. The system identification phase adopted nonlinear estimators, such as the Extended Kalman Filter (EKF) and the Unscented Kalman Filter (UKF), to estimate the hydrodynamic coefficients according to the type of maneuver in which they are relevant. The experimental data were obtained with the sensors of the Pirajuba AUV during sea-trials. The identified model is used to simulate the vehicle maneuver and the movement variables are compared to the experimental data, thus validating the identified model. The method is applied to the estimation of the complete set of hydrodynamic coefficients included in the equations of motion.