Proceedings of the Korean Institute of Intelligent Systems Conference (한국지능시스템학회:학술대회논문집)
- 2008.04a
- /
- Pages.233-236
- /
- 2008
Terrain Information Extraction for Traversability Analysis of Unmaned Robots
무인로봇의 주행성 분석을 위한 지형정보 추출
-
Jin, Gang-Gyoo
(Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University)
;
- Lee, Hyun-Sik (Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University) ;
- Lee, Yun-Hyung (Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University) ;
-
So, Myung-Ok
(Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University)
;
- Chae, Jeong-Sook (Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University) ;
- Lee, Young-Il (Div. of Computer, Control & Electronic Communication Eng., Korea Maritime University)
-
진강규
(한국해양대학교 컴퓨터 제어 전자통신공학부)
;
- 이현식 (국방기술품질원 기동화력센터) ;
- 이윤형 (한국항만연수원) ;
-
소명옥
(한국해양대학교 선박전자기계공학부)
;
- 채정숙 (국방과학연구소) ;
- 이영일 (국방과학연구소)
- Published : 2008.04.25
Abstract
Recently, the development and application of unmaned robots are increasing in various fields including surveillance and reconnaissance, planet exploration and disaster relief. Unmaned robots are usually controlled from distance using radio communications but they should be equipped with autonomous travelling function to cope with unexpected terrains and obstacles. This means that unmanned robots should be able to evaluate terrain's characteristics quantitatively using mounted sensors so as to traverse harsh natural terrains autonomously. For this purpose, this paper presents an algorithm for extracting terrain information from elevation maps as an early step of traversability analysis. Slope and roughness information are extracted from a world terrain map based on least squares method and fractal theory, respectively. The effectiveness of the proposed algorithm is verified on both fractal and real terrain maps.