Autonomous ROV Inspections of Aquaculture Net Pens Using DVL
Herman B. Amundsen, Walter Caharija, Kristin Y. Pettersen
Abstract
This article presents a method for guiding a remotely operated vehicle (ROV) to autonomously traverse an aquaculture net pen. The method is based on measurements from a Doppler velocity log (DVL) and uses the measured length of the DVL beam vectors to approximate the geometry of a local region of the net pen in front of the ROV. The ROV position and orientation relative to this net pen approximation are used as inputs to a nonlinear guidance law. The guidance law is based upon the line-of-sight (LOS) guidance law. By utilizing that an ROV is fully actuated in the horizontal plane, the crosstrack error is minimized independently of the ROV heading. A Lyapunov analysis of the closed-loop system with this guidance law shows that the ROV is able to follow a continuous path in the presence of a constant irrotational ocean current. Finally, results from simulations and experiments demonstrating the performance of the net pen approximation and control system are presented.