基于OpenCV与NAOqi库的NAO机器人点球决策  

Penalty Decisions by NAO Robot Based on OpenCV and NAOqi Library

作　　者：冯萍[1] 陈泽宇 FENG Ping;CHEN Zeyu(College of Computer Science and Technology, Changchun University, Changchun 130022, China)

机构地区：[1]长春大学计算机科学与技术学院,长春130022

出　　处：《长春大学学报》2021年第10期1-5,共5页Journal of Changchun University

基　　金：吉林省教育厅课题(SJYB19-05)。

摘　　要：针对NAO机器人在点球比赛中,目标识别率过低的问题,本文提出一种基于OpenCV与NAOqi库的NAO机器人点球决策。使用NAOqi库获取NAO机器人摄像头捕获到的视频流,利用掩膜+霍夫圆检测算法实现对足球的高精准定位,并使用ALMotion模块控制机器人完成点球决策,结合使用Choregraphe程序完成点球动作。实验结果表明,该决策降低了点球比赛中识别球失败、踢球角度出现误差的情况,实现了快速稳定的目标识别。Focusing on the issue of the low accuracy of target recognition by NAO robot in the penalty match,a penalty decision by NAO robot based on OpenCV and NAOqi library is proposed.The video stream captured by NAO robot camera is obtained by using NAO library,and the high precision positioning of the football is realized by using the mask and Hough circle detection algorithm,the AlMotion module is used to control the robot to complete the penalty decision,and Choregraphe program is used to complete the penalty action.Experimental results show that the decision can reduce the failure of ball recognition and the error of kick angle,realizing the fast and stable target recognition.

关 键 词：NAO机器人 视觉传感 点球策略 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]