变电站机器人视觉伺服系统研究  

Research on Visual Servo System of Robot for Substation

作　　者：许湘明[1] 宋晖[1] 

机构地区：[1]西南科技大学计算机学院,四川绵阳621010

出　　处：《西南科技大学学报》2011年第4期61-64,70,共5页Journal of Southwest University of Science and Technology

基　　金：科技部中小企业创新基金(09C26215105358)

摘　　要：变电站巡检机器人替代人工巡检变电站设备,主要是机器人携带CCD摄像仪和红外成像仪,在固定点对目标设备的状态进行检测。由于机器人的定位误差和机器人本体自由度的误差,会造成机器人到固定点取图像时巡检目标偏移甚至丢失。采用Adaboost图像识别分类方法和光学成像原理组成视觉伺服控制系统,对机器人进行调整,使巡检目标保持在图像的中心位置。实验结果表明,机器人视觉伺服系统能有效避免机器人定位和自身误差带来巡检目标偏移和丢失的影响。Robot replace human inspecting substation equipment, it carries a CCD camera and an infrared imaging device to detect the status of the target device at fixed points. As errors of location and body degrees of freedom cause that the robot take images with target offset or even lost at fixed points. Using an adaboost classification image recognition and optical imaging theory composite a visual servo control system, it can adjust the robot gesture so that inspection goals stay at the center of the image. Experimental results show that the robot visual servo system can effectively avoid the target offset and lost which caused by the errors of location and the robot degrees of freedom.

关 键 词：巡检机器人 视觉伺服系统 图像识别 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]