Title :
Adaptive long range vision in unstructured terrain
Author :
Erkan, Ayse Naz ; Hadsell, Raia ; Sermanet, Pierre ; Ben, Jan ; Muller, Urs ; LeCun, Yann
Author_Institution :
New York Univ., New York
fDate :
Oct. 29 2007-Nov. 2 2007
Abstract :
A novel probabilistic online learning framework for autonomous off-road robot navigation is proposed. The system is purely vision-based and is particularly designed for predicting traversability in unknown or rapidly changing environments. It uses self-supervised learning to quickly adapt to novel terrains after processing a small number of frames, and it can recognize terrain elements such as paths, man-made structures, and natural obstacles at ranges up to 30 meters. The system is developed on the LAGR mobile robot platform and the performance is evaluated using multiple metrics, including ground truth.
Keywords :
learning (artificial intelligence); mobile robots; path planning; robot vision; terrain mapping; LAGR mobile robot platform; adaptive long range vision; autonomous off-road robot navigation; ground truth; predicting traversability; probabilistic online learning framework; self-supervised learning; unstructured terrain; Feature extraction; Humans; Intelligent robots; Navigation; Optical computing; Roads; Rough surfaces; Sensor systems; Surface roughness; USA Councils;
Conference_Titel :
Intelligent Robots and Systems, 2007. IROS 2007. IEEE/RSJ International Conference on
Conference_Location :
San Diego, CA
Print_ISBN :
978-1-4244-0912-9
Electronic_ISBN :
978-1-4244-0912-9
DOI :
10.1109/IROS.2007.4399622