Model Predictive Visual Servoing of Fully-Actuated Underwater Vehicles with a Sliding Mode Disturbance Observer

Abstract

This paper presents a sliding-mode observer-based model predictive control (SMO-MPC) strategy for image-based visual servoing (IBVS) of fully-actuated underwater vehicles subject to field of view and actuator constraints and model uncertainties. In the proposed SMO-MPC controller, the visual system model and the approximate underwater vehicle model are used to predict the future trajectories from the current states driven by input candidates over a certain horizon. With the consideration of system uncertainties, including external disturbances and unknown dynamic parameters, a sliding-mode observer is designed to estimate the modeling mismatch, which is feedforward to the dynamic model in MPC. The actual control signals are generated at each step by minimizing a cost function of predicted trajectories under system constraints. The effectiveness of the proposed SMO-MPC IBVS controller is verified by comparative simulations using a fully-actuated underwater vehicle with different control configurations.