基于计算机视觉的排爆机器人三维坐标计算（英文）

针对目前排爆机器人手动控制方式的不足,研究和开发了一种基于计算机双目视觉的排爆机器人三维坐标计算方法,并依此设计相应的控制系统,以实现排爆机器人的自动抓取。首先介绍排爆机器人机械结构设计,然后介绍计算机双目视觉的原理及其软硬件,最后抓取精度实验和分析,机器人自动抓取目标物时,误差小于2 cm,满足抓取任务的精度要求。抓取中,当手爪在机器人正前方时,误差最小,反之,当手爪在机器人正前方的上面或下面、左边或右边时,误差变大;同时目标位置愈近愈好,但由于机械手的视觉范围决定不能太近,否则可能抓不到,或无法使目标物在摄像头的视觉范围之内。

Obtain 3D coordinates of EOD robots based on computer vision

Aiming to solve the current shortage of manual control methods of EOD robots, a 3d coordinate calculation method of EOD robots based on computer binocular vision was studied and developed, and a corresponding control system was designed accordingly to achieve the automatic grasping of EOD robots. This paper first introduces the mechanical structure design of EOD robots, then introduces the principles of computer binocular vision and its software and hardware. Finally, the experiment and analysis of grasping precision are carried out. In grasping, when the hand claw is in front of the robot, the error is minimized. Conversely, when the claw is above or below, to the left or to the right of the front of the robot, the error becomes larger. Meanwhile, the closer the target position, the lower the error. But the visual range of the robot is not very close, otherwise the target may not be caught, or the target may not be able to be in the visual range of the camera.