摘要
为了让NAO机器人能准确行走至目标位置,针对NAO机器人在目标识别中存在的坐标精度问题,搭建了基于NAO机器人的目标识别行走系统。在NAO机器人目标定位与行走实验中,采用单目视觉定位算法来确定目标位置,并在改变NAO机器人与目标的相对位置后进行了多次测量;通过坐标的误差补偿,使目标位置的坐标误差缩小至2 cm左右,满足了精度要求。
In order to enable the NAO robot to accurately walk to the target position,a target recognition walking system based on the NAO robot is built to solve the problem of coordinate accuracy in the target recognition of the NAO robot.In the experiment of NAO robot’s target positioning and walking,the monocular vision positioning algorithm is used to determine the target position,and several measurements are made after changing the relative position between NAO robot and the target.Through the error compensation of the coordinates,the coordinate error of the target position is reduced to 2 cm,which meets the accuracy requirements.
作者
温盛军
孙宇昂
WEN Shengjun;SUN Yuang(Zhongyuan-Petersburg Aviation College,Zhongyuan University of Technology,Zhengzhou 450007,China;School of Electronic and Information,Zhongyuan University of Technology,Zhengzhou 450007,China)
出处
《中原工学院学报》
CAS
2022年第2期45-50,共6页
Journal of Zhongyuan University of Technology
基金
河南省科技攻关项目(182102410056)
NSFC-深圳机器人基础研究中心项目重点支持项目(U1813201)。
关键词
NAO机器人
目标识别
单目视觉定位
误差补偿
NAO robot
target recognition
monocular vision positioning algorithm
error compensation
