Control of an Autonomous Underwater Vehicle attached with a Manipulator

Abstract

The thesis addresses the modelling and control problem of an Autonomous Underwater Vehicle (AUV) that is attached with a 2-link manipulator. Firstly, an INFANTE AUV and 2-link manipulator are modeled. A system comprising of both, an AUV and a manipulator is formed and the effects of different forces acting on them are studied in detail.

Further a Model Reference Adaptive Control (MRAC) algorithm is proposed for the Autonomous Underwater Vehicle Manipulator System (AUVMS) for trajectory tracking. Since the number of states of the system increases due to addition of manipulator, a resolution method is applied to remove the redundant states from the system. It may be noted that due to the hydrodynamic effects, the AUVMS experiences a dissipative natured drag force that has no
contribution to the motion but consumes much energy thereby reducing the efficiency of the system. Therefore, Steepest Descent method have been employed to reduce the drag on the system. The positive gains for position and velocity are varied by developing an adjustment using MIT Rule. It optimizes the error and subsequently generates the required thruster force to traverse the desired reference path. The manipulator begins to function when the AUV gets stable. The links of the manipulator moves from an initial angular position to a desired angular position based on the control input torque generated by the respective joints of the manipulator. The effectiveness of the simulated result is described in MATLAB and SIMULINK environment.

Subsequently, a Model Predictive Controller (MPC) is designed for the position control of
the AUV and the manipulator. Force generated by a thruster for an AUV has a practical limit and cannot exceed beyond its defined range.This aspect of imposing constraint over the control input has not been considered in the MRAC algorithm which can be resolved by designing an MPC. A linear model is derived for both the AUV and the 2-link manipulator and further an MPC is developed for controlling the AUVMS. The constraints are set over the thruster force and torque to control the AUV and the manipulator respectively, to maintain a desired fixed position. The performance of the controller is analyzed from the MATLAB and SIMULINK simulation results.