3D Reconstruction for the Operating Environment of the Robots in Distribution Hot-line Working Based on Mixed Reality

Abstract

The hot-line work robot for distribution network, an important branch of intelligent robot, has been given more and more attention. In the operation of the hot-line work robot for distribution network, environmental inference such as weather, light and temperature could have a great impact on the robot’s performance. Therefore, it is of great significance to improve the anti-inference performance of robot in distribution-network operation. Firstly, this article studies about 3D vision recognition and reconstruction in terms of distribution network operation, and explores the principle of binocular stereo vision in measuring sparse parallax of conducting wires in live working environment of distribution network and the operator-for -contour-matching algorithm based on RGB image data. Secondly, this article builds up a system, where based on binocular vision and laser point cloud, the distance between wires and the robot is measured, parameters of the operating environment are identified, the target’s depth is estimated. Thus, the anti-disturbance performance and operating accuracy are effectively improved.