Nonlinear and Machine-Learning-Based Station-Keeping Control of an Unmanned Surface Vehicle

Abstract This study is part of ongoing work on situational awareness and autonomy of a 16’ WAM-V USV. The objective of this work is to determine the potential and merits of application of two different station-keeping controllers for a fixed-pose motion control of the USV. The assessment includes performance and power consumption metrics tested under harsh environmental disturbances to evaluate the robustness of the control methods. The first is a nonlinear trajectory-tracking control method based on the sliding-mode control technique, while the second method uses a machine-learning approach based on Deep Reinforcement Learning. Results from both the approaches are compared for various case studies.