Experiments on vision guided docking of an autonomous underwater vehicle using one camera
Experiments on vision guided docking of an autonomous underwater vehicle using one camera
复制标题
DOI：
10.1016/j.oceaneng.2008.10.001
复制
发表时间：
2009-01-01
影响因子：
5
通讯作者：
Oh, Junho
中科院分区：
文献类型：
--
作者：
Park, Jin-Yeong;Jun, Bong-huan;Oh, Junho
介绍了一种基于单CCD相机的自主式水下航行器（AUV）试验平台ISiMI的水下对接过程。AUV由安装在对接站入口周围的灯和由CCD相机和AUV中的帧捕获器组成的视觉系统进行光学引导。本文提出了一种基于光图像识别的码头识别方法，并提出了一种基于视觉引导的码头识别算法。介绍了一种消除光捕获图像噪声的信号处理方法，并提出了一种两阶段的终端稳定对接方法。针对垂直面和水平面的视觉导引控制，采用传统的PID控制器设计了视觉导引控制器。通过实验验证了视觉引导对接系统的有效性。(C)2008爱思唯尔有限公司保留所有权利。
This paper introduces an underwater docking procedure for the test-bed autonomous underwater vehicle (AUV) platform called ISiMI using one charge-coupled device (CCD) camera. The AUV is optically guided by lights mounted around the entrance of a docking station and a vision system consisting of a CCD camera and a frame grabber in the AUV. This paper presents an image processing procedure to identify the dock by discriminating between light images, and proposes a final approach algorithm based on the vision guidance. A signal processing technique to remove noise on the defused grabbed light images is introduced, and a two-stage final approach for stable docking at the terminal instant is suggested. A vision-guidance controller was designed with conventional PID controllers for the vertical plane and the horizontal plane. Experiments were conducted to demonstrate the effectiveness of the vision-guided docking system of the AUV. (C) 2008 Elsevier Ltd. All rights reserved.