Title :
PixelLaser: Evaluating monocular range from texture
Author :
Leece, Michael ; Lesperance, Nicole ; Matsumoto, Steve ; Korbel, Max ; Lei, Kenny ; Dodds, Zachary
Author_Institution :
Comput. Sci. Dept., Harvey Mudd Coll., Claremont, CA, USA
Abstract :
This work presents and evaluates the PixelLaser system, designed to estimate range-to-obstacle scans from single images. Its visual pipeline uses nearest-neighbor texture-matching to segment groundplane (traversable) texture from non-groundplane (obstacle) texture. Using the known pose of a webcamera, the system then transforms these segmentations into range-scans. This paper presents a thorough evaluation of the range accuracy obtained in the resulting scans. In addition, it presents three practical robot applications that use PixelLaser´s monocular ranging: (a) a safe-wandering agent, (b) a Monte-Carlo localizer, and (c) a mapping system using CoreSLAM. In each case, algorithms that usually depend on laser scans have succeed with PixelLaser data instead.
Keywords :
cameras; collision avoidance; image segmentation; image texture; laser ranging; optical scanners; robot vision; CoreSLAM; Monte-Carlo localizer; PixelLaser; image texture; mapping system using; monocular range; monocular ranging; nearest neighbor texture matching; nongroundplane texture; range-to-obstacle scan; robot application; safe wandering agent; visual pipeline; webcamera; Accuracy; Classification algorithms; Image segmentation; Pipelines; Pixel; Robot sensing systems; image segmentation; range estimation; visual control;
Conference_Titel :
Technologies for Practical Robot Applications (TePRA), 2011 IEEE Conference on
Conference_Location :
Woburn, MA
Print_ISBN :
978-1-61284-482-4
DOI :
10.1109/TEPRA.2011.5753499
