관성 항법 센서와 도킹 유도 센서를 융합한 무인잠수정의 도킹 유도 항법 알고리즘에 대한 연구

Abstract

This paper describes a docking guidance navigation algorithm for autonomous underwater vehicles (AUVs). To dock an AUV to a docking station installed underwater, a set of inertial navigation sensors used for conventional AUV navigation and a set of docking guidance sensors for docking missions are used. A docking guidance sensor typically measures the relative distance and attitude between the AUV and the docking station. As many different sensors are used, a precise position and attitude estimation technique, considering both types of sensors, is needed. Therefore, we designed a serial filter chain for estimating the orientation, position, and relative distance. Considering the non-holonomic characteristics of an AUV in the underwater environment, we first estimated the attitude by separating the position and the attitude. Then, the position of the AUV was estimated on the basis of the estimated attitude. Finally, we estimated the relative distance from the docking station to the estimated AUV position. This paper contributes to the following topics: i) position and attitude estimation considering non-holonomic characteristics of AUVs: and ii) precision estimation algorithm for docking guidance using docking guidance sensors.