Adaptive Model Predictive Control with Online System Identification for an Unmanned Underwater Vehicle

Abstract

Unmanned underwater vehicles (UUVs) are increasingly important for various underwater tasks, and achieving autonomy for UUV s is a major focus. This is crucial for enhancing safety, flexibility, extending operational range, and reducing costs. However, designing effective and robust control algorithms for UUV s is challenging due to nonlinear dynamics, uncertainties, constraints, and environmental disturbances. Model predictive control (MPC) is a well-established technique for UUV control, while how to obtain accurate prediction models becomes the key challenge for improving controller performance. This paper proposes an online system identification method based on an extended active observer (EAOB) and the recursive least squares with variable forgetting factor (RLS- VFF) algorithm to estimate environmental disturbances and identify uncertain hydrodynamic parameters. The estimated disturbances and parameters are continuously updated in the MPC's prediction model to generate optimal control inputs based on the real-time environment and vehicle conditions. The proposed approach is validated through the Gazebo and robot operating system (ROS) simulation environment, demonstrating its effectiveness in handling uncertainties and disturbances for UUV control.