Robust Trajectory Tracking Control for Underactuated Autonomous Underwater Vehicles in Uncertain Environments

This article addresses the tracking control problemof 3-D trajectories for underactuated underwater robotic vehiclesoperating in a constrained workspace including obstacles. Morespecifically, a robust nonlinear model predictive control (NMPC)scheme is presented for the case of underactuated autonomousunderwater vehicles (AUVs) (i.e., unicycle-like vehicles actuatedonly in the surge, heave, and yaw). The purpose of the controlleris to steer the unicycle-like AUV to the desired trajectory withguaranteed input and state constraints (e.g., obstacles, predefinedvehicle velocity bounds, and thruster saturations) inside a par-tially known and dynamic environment where the knowledge ofthe operating workspace is constantly updated via the vehicle’sonboard sensors. In particular, considering the sensing rangeof the vehicle, obstacle avoidance with any of the detectedobstacles is guaranteed by the online generation of a collision-freetrajectory tracking path, despite the model dynamic uncertaintiesand the presence of external disturbances representing oceancurrents and waves. Finally, realistic simulation studies verifythe performance and efficiency of the proposed framework.QC 20201116