An optimal guidance scheme for cross-track control of underactuated underwater vehicles

This paper considers the development of an optimal guidance scheme for cross-track control of underactuated underwater vehicles. We propose an optimal guidance algorithm giving a set of desired course angles for the underwater vehicle. The computed angles are then used as reference inputs to an underlying controller, which guarantees tracking of the desired course angles. The proposed guidance algorithm seeks to minimize the cross-track error, i.e. the distance from the vehicle to the desired path, and maximize the rate of convergence, while keeping the desired course angles and the corresponding desired rates of turn of the vehicle within feasible bounds. Moreover, damping is injected in the guidance algorithm to guarantee a well-damped final approach. The proposed guidance scheme allows for easy adjustment of the tradeoff between high convergence rate and a well-damped final approach. The performance of the proposed guidance algorithm is compared to the performance of traditional line-of-sight (LOS) guidance in a case study with the HUGIN AUV