Algorithm of object localization applied on high-voltage power transmission lines based on line stereo matching

Abstract

The algorithm designing of object localization is one of the key technologies in the research field of high-voltage power transmission line inspection robot. We present an obstacle localization method based on line stereo matching to solve the problems remaining in the current obstacle localization algorithms such as the weak capacity of resisting distraction, the low localization precision, and the extra prior conditions. The proposed algorithm regards line as the element unit for stereo matching and adopts Earth mover’s distance as similarity measurement criterion. In addition, it can remove false matching points through the condition of distance constraint and the distribution information of depth. Finally, we give all sampling points on obstacle different weights according to the distance between the points and obstacle center, through which the weighted mean depth of obstacle is obtained. The experimental result shows that the mean relative localization error is 2.697% and the fluctuating range is 1.11% to 3.58%, which means the localization precision is high and the error fluctuation is small. So this algorithm can meet the demand of obstacle localization for high-voltage power transmission lines inspection robot efficiently.