Modelado Submarino

IEEE JOURNAL OF OCEANIC ENGINEERING, VOL. 29, NO. 1, JANUARY 2004

1

Model-Based Dynamic Positioning of Underwater Robotic Vehicles: Theory and Experiment
David A. Smallwood, Member, IEEE and Louis L. Whitcomb, Senior Member, IEEE
Abstract— This paper addresses the trajectory tracking problem for the low-speed maneuvering of fully actuated underwater vehicles. It is organized as follows:First a brief review of previously reported control studies and plant models is presented. Second, an experimentally validated plant model for the Johns Hopkins University Remotely Operated Underwater Vehicle (JHUROV) is reviewed. Third, the stability of linear PD control and a family of model-based controllers is examined analytically and demonstrated with numerical simulations. Finally we reportresults from experimental trials comparing the performance of these controllers over a wide range of operating conditions. The experimental results corroborate the analytical predictions that the model-based controllers outperform PD control over a wide range of operating conditions. The exactly linearizing model-based controller is outperformed by its non exactly linearizing counterpart. Theadaptive controllers are shown to provide reasonable online plant parameter estimates, and velocity and position tracking consistent with theoretical predictions — providing good velocity tracking and, with the appropriate parameter update law, position tracking. The effects of reference trajectory, “bad” model parameters, feedback gains, adaptation gains, and thruster saturation are experimentallyevaluated. To the best of our knowledge, this is the first reported comparative experimental study of this class of model-based controllers for underwater vehicles.

mechanical systems. These tracking controllers have generally employed one of the three following general approaches: 1) Linear Control - PD and PID: The globally asymptotic regulation of a fully nonlinear mechanical system was firstaddressed for PD control of robot arms in [6], in which Lasalle’s Invariance Theorem, [43], is employed to show that PD control provides globally asymptotic setpoint regulation. The application of this approach to the control of underwater vehicles is the PD controller outlined in Section III-A. 2) Nonlinear Model-Based Exact Linearization: The case of exact model-based trajectory tracking controlof robot arms was first reported independently in [47], [46] and [25], which report controllers with modelbased feedforward terms that exactly linearize (without any approximation) the plant and result in globally asymptotically stable linear error dynamics. Working implementations of this approach were first reported in [2]. An adaptive version of this exact linearization approach was reported in[13], [12] and shown to be globally asymptotically stable in tracking error and locally stable in plant parameter error. The application of this approach to the control of underwater vehicles is the exact linearizing controller (EL) outlined in Section III-B, and the adaptive exact linearizing controller (AEL) outlined in Section III-D. 3) Nonlinear Model-Based Without Linearization: The case ofexact model-based tracking of robot arms without exact linearization was first reported in [53], [54] along with an adaptive extension that is globally asymptotically stable in tracking error and globally stable in plant parameter error. Variations of this general approach are reported in [52], [63]. The application of this approach to the control of underwater vehicles is the nonlinear controller(NL) outlined in Section III-C, the adaptive nonlinear controller (ANL) outlined in Section III-E, and the epsilon adaptive nonlinear controller (¯ANL) of Section III-F. For the special problem of trajectory tracking of underwater vehicles, a significant number of previously reported studies have employed linearized plant approximations both for the theoretically justified case of setpoint regulation...