Transformer substation inspection robot obstacle detection and recognition method integrated with depth camera

Abstract

The invention discloses a transformer substation inspection robot obstacle detection and recognition method integrated with a depth camera. The method comprises steps of using a depth camera for collecting data, preprocessing data, converting the depth map and the RGB color image into a point cloud image, eliminating ground point cloud in the point cloud image by using a defined ground plane model, then performing adaptive clustering on obstacles by using an adaptive density clustering algorithm, finally judging the types of the obstacles, and finally converting obstacle information into two-dimensional laser radar data information and sending the two-dimensional laser radar data information to a path planning algorithm. The calculated amount of the whole process is lower than that of an obstacle segmentation algorithm and an image segmentation obstacle algorithm using deep learning; meanwhile, due to the introduction of the depth camera, the transformer substation robot can easily obtain external three-dimensional information and color images, the perception capacity of the robot is greatly enhanced, and therefore the accuracy and stability of the obstacle avoidance function of the transformer substation inspection robot can be improved.

Description

technical field [0001] The invention relates to the technical field of automatic obstacle avoidance of a substation inspection robot, in particular to an obstacle detection and recognition method for a substation inspection robot integrated with a depth camera. Background technique [0002] In recent years, with the development of mobile robot technology, more and more mobile robots have entered people's lives. For example, home service robots, shopping guide robots, sweeping robots, substation inspection robots, etc. With the advent of the era of artificial intelligence (AI), robots are becoming more and more intelligent. In the existing robot technology, the automatic obstacle avoidance technology is particularly important. Most mobile robots need to realize the automatic obstacle avoidance function. The automatic obstacle avoidance function is an important symbol of intelligence for mobile robots. Therefore, it is of great significance to study the detection and recogni...

AI Technical Summary

Problems solved by technology

[0003] At present, there are many ways to realize robot obstacle detection and recognition functions, among which the most widely used is based on two-dimensional laser radar, but due to the problem of the principle of two-dimensional laser radar, only Obtain a two-dimensional cross-section of a three-dimensional space consistent with the installation height of the sensor through the sensor, so the obstacle information obtained is less, and obstacles of different heights in front of the inspection robot cannot be obtained comprehensively, which is likely to cause safety problems during the inspection process , in addition to calculating the depth of obstacles through binocular cameras and segmenting obstacles using color images, this algorithm is more complex and has a large amount of calculation, and the real-time performance is low

Images