Abstract: Power line inspection robot must plan its behavior to negotiate obstacles according to their types when it is crawling along the power transmission line.For this purpose，a visual navigation system is designed to recognize the obstacles and locate their positions by means of stereo cameras.The algorithm of visual navigation consists of two modules，obstacle recognition and obstacle locating.We propose a structure-constrained obstacle recognition algorithm to recognize obstacles，after the obstacle is recognized，a region feature based on stereo matching algorithm is used to search the correspondence points in the stereo images，and the position of the obstacle relative to the robot is calculated by 3D reconstruction.To improve the matching speed，a dynamic search window is applied.Experiments with simulation and real transmission line show that the algorithm can recognize and locate obstacles such as counterweight，strain clamp，and suspension clamp from complex background reliably.

Key words: power transmission line inspection robot, visual navigation, obstacle recognition, stereo matching

摘要： 巡线机器人沿相线行走时必须探测识别和定位各种障碍物，并根据障碍类型规划越障行为。针对220 kV架空输电线路的结构特点，利用视觉传感器，设计了基于结构约束的障碍识别算法，完成了障碍识别和分类。根据障碍物的结构特点，设计了一种自适应多窗口区域立体匹配算法，实现了障碍物的双目视觉定位。模拟线路实验结果表明，算法能可靠地从复杂背景中识别并定位出防振锤、悬垂线夹和耐张线夹等障碍物，满足了巡线机器人导航要求。

关键词: 巡线机器人, 视觉导航, 障碍识别, 立体匹配