文章基本信息

标题：Cost-Effective Stereo Vision System for Mobile Robot Navigation and 3D Map Reconstruction
本地全文：下载
作者：Arjun B Krishnan ; Jayaram Kollipara
期刊名称：Computer Science & Information Technology
电子版ISSN：2231-5403
出版年度：2014
卷号：4
期号：7b
页码：75-86
DOI：10.5121/csit.2014.4716
出版社：Academy & Industry Research Collaboration Center (AIRCC)
摘要：The key component of a mobile robot system is the ability to localize itself accurately in anunknown environment and simultaneously build the map of the environment. Majority of theexisting navigation systems are based on laser range finders, sonar sensors or artificiallandmarks. Navigation systems using stereo vision are rapidly developing technique in the fieldof autonomous mobile robots. But they are less advisable in replacing the conventionalapproaches to build small scale autonomous robot because of their high implementation cost.This paper describes an experimental approach to build a cost- effective stereo vision system forautonomous mobile robots that avoid obstacles and navigate through indoor environments. Themechanical as well as the programming aspects of stereo vision system are documented in thispaper. Stereo vision system adjunctively with ultrasound sensors was implemented on themobile robot, which successfully navigated through different types of cluttered environmentswith static and dynamic obstacles. The robot was able to create two dimensional topologicalmaps of unknown environments using the sensor data and three dimensional model of the sameusing stereo vision system.
关键词：Arduino; Disparity maps; Point clouds; Stereo vision; Triangulation