Fifth "Coffee with EUMR"

This talk discusses the problem of using single or multiple cooperative marine autonomous vehicles (trackers) to localize and pursue an unknown underwater moving target using measurement of the ranges between the vehicles and the target. We first show how this problem can be solved using a combination of model predictive control (MPC) and estimation theory, based on the computation of the Bayesian Fisher information matrix (FIM). In this context, the Bayesian FIM is used to predict range-information along admissible vehicle trajectories, while MPC aims to control the vehicles moving along optimal trajectories that maximize range-related information while meeting energy and inter-vehicle collision avoidance constraints.  For the case of multiple vehicles, we propose an efficient distributed estimation and control (DEC) strategy for the vehicles that takes into account explicitly the constraints imposed by the inter-vehicle communication network topology. To this end, a distributed extended Kalman filter (DEKF) and a distributed control strategy are developed for the vehicles to cooperatively pursue and localize the target. Using this set-up, all vehicles are guaranteed converge to a specified vicinity of the target while keeping an optimal vehicle-target relative geometry that maximizes the range information acquired to estimate the target’s state.  Simulations illustrate the performance of the systems developed.

In the recent past, intervention-class autonomous underwater vehicles (AUVs) have gained the ocean/marine research community’s attention due to their multidisciplinary operational capabilities. The presentation focuses on a detailed comparative-performance study of an underwater vehicle (UV) with two different actuator configurations based on fixed-thrusters and vectored thrusters. The fixed-thruster configuration consists of eight thrusters. Four thrusters generate thrust in the horizontal plane and the remaining thrusters produce thrust in a heave direction. The vectored thruster configuration has four thrusters placed at diagonal corners of a rectangular vehicle, and four independent internal servomotors rotate these thrusters. The classification terminology in this article identifies the underwater vehicle without vectoring capability as Fixed-Thruster Underwater Vehicle (FTUV) and the one with vectoring capability as Vectored-Thruster Underwater Vehicle (VTUV). Furthermore, a nonlinear controller is designed, based on a backstepping scheme for tracking a given spatial profile in a three-dimensional space. An underwater robot’s control for a trajectory tracking problem is useful for any real-time multidisciplinary autonomous applications. The comparative simulation results are summarized to validate both the vehicle variants’ performance differences.