@inproceedings{scholars8521,
           title = {Nonlinear adaptive control synthesis using U-model for multivariable underwater remotely operated vehicle},
          volume = {2018-J},
            note = {cited By 0; Conference of 7th IEEE International Conference on Underwater System Technology: Theory and Applications, USYS 2017 ; Conference Date: 18 December 2017 Through 20 December 2017; Conference Code:135172},
             doi = {10.1109/USYS.2017.8309442},
         journal = {2017 IEEE 7th International Conference on Underwater System Technology: Theory and Applications, USYS 2017},
       publisher = {Institute of Electrical and Electronics Engineers Inc.},
           pages = {1--6},
            year = {2017},
          author = {Hussain, N. A. A. and Ali, S. S. A. and Saad, M. N. M. and Ovinis, M.},
            isbn = {9781538619186},
             url = {https://www.scopus.com/inward/record.uri?eid=2-s2.0-85050595045&doi=10.1109\%2fUSYS.2017.8309442&partnerID=40&md5=a477cdad45927eff1705dc9e60218e32},
        keywords = {Adaptive control systems; Kinematics; Model predictive control; Nonlinear equations; Radial basis function networks; System stability, Control model; Control synthesis; Model synthesis; Modeling approach; Modelling and controls; Multi variables; Nonlinear adaptive control; Radial basic function; U-model; U-model \& ROV, Remotely operated vehicles},
        abstract = {This paper presents the development of ROV control modelling and control synthesis using nonlinear adaptive U-model approach. Nonlinear ROV model based on the dynamic equation using the Newtonian method and derivation towards the kinematics equations and rigid-body mass matrixes are explained. This nonlinear ROV model represents the underwater thruster dynamics, ROV dynamics and kinematics related to the earth-fixed frame. MIMO Nonlinear adaptive control synthesis using U-model approach incorporate with neural networks algorithm are developed with MATLAB{\^a}?c Simulink software and integrated together with the nonlinear ROV model using Internal Model Control structure. The controller output is based on Newton Raphson recursive algorithm with learning rate value between zero and one which improve the system stability. Radial basis function (RBF) is chosen for the neural networks activation function due to faster learning speed. Results show good control signal convergence and tracking performance between plant or system model with U-model polynomial. {\^A}{\copyright} 2017 IEEE.}
}