论文部分内容阅读
为准确快速地定位高压线路上的障碍物,实现有效越障,提出了一种用于巡线机器人视觉导航的单目测距算法。采用Hough变换在视频帧图像中检测出机器人行走导线的两侧边缘;预先测取导线上可成像的最近点到镜头的距离且机器人移动过程中保持不变,由此处导线两侧边缘直线纵坐标差和障碍物处的两直线纵坐标差之比,得到障碍物到镜头的距离并对其进行线性修正,最终可得到障碍物沿导线到镜头的距离。试验结果表明,提出的方法在5 000mm内测量误差小于5%,可测距离至少能达7.4m;且算法具有精度较高、速度快、鲁棒性强、需测量参数较少和易于实现等优点。
In order to locate the obstacle on the high voltage line accurately and quickly and realize the effective obstacle avoidance, a monocular range finding algorithm for visual navigation of line inspection robot is proposed. The Hough transform is used to detect the edges of the robot walking wire in the video frame image. The distance between the nearest point on the wire that can be imaged to the lens and the distance between the lens and the robot are kept unchanged. The edge of the wire Coordinate difference and the obstacle at the two lines of the vertical axis of the ratio of the difference between Obtain the distance to the lens and its linear correction, and finally get the obstacle along the wire to the lens distance. The experimental results show that the proposed method has the measurement error of less than 5% within 5 000 mm and the measurable distance of at least 7.4 m. The proposed algorithm has the advantages of high accuracy, fast speed, robustness, less measurement parameters and easy implementation advantage.