Design and implementation of autonomous surface vehicle using visual target tracking

Abstract

Recently, due to the increasing interest in autonomously exploring the vast ocean environment, the ocean robots are being developed for exploring the oceans. The research is actively conducted about the Autonomous Surface Vehicle (ASV) and the Autonomous Underwater Vehicle (AUV). Such robots can be widely used to monitor the ocean environment or to militarily reconnoiter. The ASV is especially gaining attention that enable to explore the surface of the oceans. To perform their tasks, an accurate localization and a navigation system are essential for the ASV to move safely in harsh environment. This thesis describes design and implementation of an autonomous navigation system for capturing task. In this thesis, the ASV is designed and implemented to capture the target using images from the camera module. The implemented system is composed of the localization module, the vision module, and the control module. The measured current location and angle data of the ASV are transmitted to the control module on the vehicle and it calculates the control signal to the thrusters. After the surface vehicle reaches near the target point, it controls its thrusters using the visual algorithm to capture the targets. In order to verify the feasibility of the proposed algorithm, the experiments were conducted.