Path Planning and T racking of an Autonomous Underwater Vehicle using Virtual Way-points

Abstract

In this paper, a novel method for the path planning and tracking of the Autonomous Underwater Vehicles (AUVs) using cameras is proposed. As the navigation of the AUVs is one of the emerging research areas in oceanic engineering, the importance of the path planning and tracking has been emphasized. Use of the visual data from cameras is one of attractive methods for underwater sensing and it is especially effective in the close range detections. In the proposed algorithm, using the vision as the primary sensor, a method for the robust path planning and smooth tracking has been implemented by generating virtual way-points based on the visually detected landmar ks. The feasibility of the algorithm has been demonstrated by the experiments using an AUV platform, K AURO, where the artificial path mar kers are used as landmar ks.