A stereo vision localization method for autonomous recovery of autonomous underwater vehicle

Abstract

This paper proposes a stereo visual guidance method to address the inability of the conventional light visual guidance method to meet accuracy and robustness requirements when an autonomous underwater vehicle (AUV) is docked to an underwater base station in short range. The proposed method is mainly divided into three parts:light source extraction, light source matching, and stereoscopic measurement. Firstly, an adaptive image binarization algorithm and a pseudo-light source elimination algorithm are used to obtain the pixel coordinates of the centers of the light source logos from the left and right images. In the light source matching part, each light source feature is extracted and matched to complete the accurate pairing of the light sources in the left and right images. Finally, binocular stereo ranging is conducted on the light source array to obtain the three-dimensional position of the base station in the camera coordinate system. Experimental results show that the proposed method has good robustness to invalid light sources and that the relative position and angle of the base station are obtained with high accuracy.