黑龙江科技大学学报2017，Vol.27Issue(3)：293-296,302,5.DOI:10.3969/j.issn.2095-7262.2017.03.018
基于折反射成像原理的机器人自定位
Robot independent localization based on central catadioptric imaging
摘要
Abstract
This paper is motivated by an attempt to realize the autonomous walking of disaster detec-tion and relief robots designed for coal mine environment and introduces an algorithm for robot self locali-zation using single view catadioptric imaging system, an algorithm capable of obtaining 360 degree images of objects around the robot by applying a panoramic imaging system and combined with the known artifi-cial landmark. The algorithm works by processing the collected images, recognizing artificial landmarks around robots, and obtaining the distance of the landmarks from the center of the panoramic image;and then using the principle behind the single view catadioptric imaging to calculate the relative positions of robot and road sign and thereby realizing the navigation and collision avoidance of underground robots. The experiment indicates that the algorithm could work better for the self-location for mobile robots.关键词
煤矿救灾机器人/全景视觉/机器人自定位/单视点折反射/目标识别Key words
coal mine rescue robot/omni-directional vision/robot independent localization/single view point central catadioptric/target recognition分类
信息技术与安全科学引用本文复制引用
梁燕华,蔡成涛,周裕..基于折反射成像原理的机器人自定位[J].黑龙江科技大学学报,2017,27(3):293-296,302,5.基金项目
黑龙江省教育厅科学技术研究项目(12541725) （12541725）
