视觉机器人的目标识别与跟踪定位研究  

Research on Target Recognition and Tracking of Vision Robot

作　　者：蔡云[1] CAI Yun(Service Robot Application Technology Collaborative Innovation Center of Anhui Sanlian University,Hefei 230601,China)

机构地区：[1]安徽三联学院服务机器人应用技术协同创新中心,合肥230601

出　　处：《重庆科技学院学报（自然科学版）》2020年第2期88-92,共5页Journal of Chongqing University of Science and Technology：Natural Sciences Edition

基　　金：国家自然科学基金一般项目“移动机器人目标识别与跟踪技术研究”(YJQR19005)。

摘　　要：根据国际机器人足球协会(FIRA)的全自动中型机器人组比赛规则,结合视觉影响、实时图像处理、动态目标追踪、无线通信、自动定位、机器人运动控制、路径规划以及控制策略等技术,实现机器人的自动定位。运用现有的定位算法,如双圆定位法、三点定位法实现机器人的自动定位时,其结果易受到比赛场地变动与环境光线的影响,产生定位误差。为了减少环境影响、改善定位误差,采用交比定位法实现机器人自动定位。应用交比定位法的误差相对较小,可通过算法自动计算场地标线交点,实现实时定位。Based on the competition rules of FIRA's RoboSot League,the competition aims are achieved by combining visual technology,real-time image processing,dynamic target tracking,wireless communication technology,automatic positioning,robot motion control,path planning and control strategy.In the competition,after getting the environmental information through omnidirectional mirror,the robot's automatic positioning function is realized.However,in the existing positioning algorithms,whether it is double circle or three-point positioning,the color model is used to calculate the goal image to achieve the automatic positioning function.Because of the change of competition venue and the influence of environmental light,the color mode easily leads to positioning error.In order to reduce the environmental impact and improve the positioning error,the intersection point of the landmark is obtained by omnidirectional mirror,and the automatic positioning target of the robot is realized by applying the cross ratio theorem.The results show that the error of the cross-ratio positioning method is less than that of the double circle positioning method and the three-point positioning method.

关 键 词：机器人 自动定位 全方位镜 交比定理 

分 类 号：TP242[自动化与计算机技术—检测技术与自动化装置]