Obstacle detection method for overhead transmission line inspection robot

Abstract

The invention discloses an obstacle detection method for an overhead transmission line inspection robot, and the method comprises the steps: carrying out the real-time detection of an obstacle on a walking ground wire of the robot through a video collection sensor, and employing a mode of combining a visual feature analysis algorithm with a deep neural network model. The method comprises the following steps: firstly, pre-judging whether an obstacle exists on a ground wire in front of a robot based on contour features, detecting the type and position of the obstacle by using a target detection model based on a convolutional neural network then, and finally carrying out distance measurement on the existing obstacle to assist the robot to adopt different obstacle crossing actions to autonomously cross the obstacle. According to the invention, by adopting the obstacle existence pre-judgment module and the obstacle existence judgment algorithm based on the ground wire contour features, the sudden change of the edge of the ground wire can be detected with high precision, the detection of the abnormal obstacle of an unknown type has good robustness, the long-time operation reasoning of the deep neural network is avoided, consumption of computing resources is remarkably reduced, and obstacle recognition efficiency is improved.

Description

technical field [0001] The invention relates to the technical field of machine vision, in particular to an obstacle detection method for an overhead transmission line inspection robot. Background technique [0002] With the improvement of the intelligent level of the power grid, the intelligent inspection of transmission lines has become an indispensable part of the power grid operation. The development of overhead transmission line inspection robots plays an important role in intelligent inspections. It can not only replace manual inspections, but also be unaffected by weather and the environment, and can run all-weather and with long battery life. In order to realize continuous autonomous inspection, the robot needs to identify obstacles on the walking route in advance, and perform corresponding obstacle-crossing actions according to the obstacle type and position information. Conventional obstacles on the ground include wire fittings such as anti-vibration hammers and wi...

AI Technical Summary

Problems solved by technology

[0006] Purpose of the invention: To provide an obstacle detection method for an overhead transmission line inspection robot. The present invention comprehensively combines the working principle of the visual feature analysis algorithm and the deep neural network model to solve the problems in the prior art that the obstacle identification type is single and the abnormality cannot be identified. Obstacles, as well as the technical problems of complex computing system, long computing time and low robustness using the network model

Images