Guidance for time efficient path following of under-actuated differential thrust AUVs

Abstract

Autonomous Underwater Vehicles (AUVs) are relatively new underwater devices developed to execute missions in water without human operators. The advancements made in AUV technology have significant implications for a wide range of underwater applications. Guidance and control of AUVs plays an important role in many applications, and it is a challenging research topic, not only because of the significant nonlinearities and couplings in the AUV’s dynamics, but the under-actuation found in typical AUVs. This thesis presents new work contributing to time efficient path following of under-actuated AUVs. Different from the conventional fin-manoeuvred AUVs, the prototype vehicle considered in this thesis is a differential thrust manoeuvred AUV devoid of fins or rudders. Such a manoeuvring feature makes the vehicle very agile, but brings challenges in guidance and control. A model of the prototype AUV is constructed based on the vehicle dynamics and manoeuvring features. In order to achieve time efficient path following, the AUV should operate at its motion limits. To derive the motion limits, a Monte Carlo analysis is conducted using the AUV model, which provides a numerical solution to derive the maximum admissible motion of the vehicle with respect to the curvatures along given paths. Thus, a curvature-based guidance system is developed. The strategy alters the AUV path following speed according to the path curvature, hence increasing the overall time efficiency. The effectiveness of the proposed method is demonstrated through simulations of the AUV following a range of different paths.