A vision-based obstacle detection system for Unmanned Surface Vehicle

Author

Wang, Han ; Wei, Zhuo ; Wang, Sisong ; Ow, Chek Seng ; Ho, Kah Tong ; Feng, Benjamin

Author_Institution

Sch. of Electr. & Electron. Eng., Nanyang Technol. Univ., Singapore, Singapore

fYear

2011

fDate

17-19 Sept. 2011

Firstpage

364

Lastpage

369

Abstract

This paper describes a vision-based obstacle detection system for Unmanned Surface Vehicle (USV) towards the aim of real-time and high performance obstacle detection on the sea surface. By using both the monocular and stereo vision methods, the system offers the capacity of detecting and locating multiple obstacles in the range from 30 to 100 meters for high speed USV which runs at speeds up to 12 knots. Field tests in the real scenes have been taken and the obstacle detection system for USV is proven to provide stable and satisfactory performance.

Keywords

collision avoidance; marine vehicles; mobile robots; remotely operated vehicles; robot vision; stereo image processing; monocular vision method; stereo vision method; unmanned surface vehicle; vision-based obstacle detection system; Cameras; Computer vision; Feature extraction; Image motion analysis; Optical imaging; Sea surface; Vectors; Unmanned Surface Vehicle (USV); computer vision; obstacle detection;

fLanguage

English

Publisher

ieee

Conference_Titel

Robotics, Automation and Mechatronics (RAM), 2011 IEEE Conference on

Conference_Location

Qingdao

ISSN

2158-2181

Print_ISBN

978-1-61284-252-3

Type

conf

DOI

10.1109/RAMECH.2011.6070512

Filename

6070512