Path planning for an identification mission of an Autonomous Underwater Vehicle in a lemniscate form

摘要

This paper describes the algorithms for guidance and control of an Autonomous Underwater Vehicle (AUV) for a specific identification mission. The goal is to traverse several times an object of interest on the sea floor in a given geometric path in order to maximize the amount of information that can be obtained using the optical and acoustical sensors. Existing ocean currents that can be estimated using a Doppler Velocity Log (DVL) have to be considered when planning the path. We will describe the overall mission scenario and suggest a possible solution, based on an adequate planning of the path in form of a lemniscate and use the "chase the rabbit" approach for path following. This approach yields stable mission execution even in the presence of a considerable ocean current. The approach is validated with a state-of-the-art AUV using live tests in a special lake that contains a current.