Data-Driven Approach for Parameter Estimation and Control of an Autonomous Underwater Vehicle

This paper showcases a data-driven non-linear adaptive controller design employing an unfalsification approach to attain optimal estimates for unknown parameters in an autonomous underwater vehicle (AUV). These estimates are applied to the controller to enable precise trajectory tracking. The controller design presented is capable of adapting to parametric changes and uncertainties while fulfilling the desired performance criteria using an effective parameter update method of unfalsification. The results were validated through simulations conducted using MATLAB/SIMULINK.