Research On Visual Navigation And Route Planning Of Substation Inspection Robot

Substation is an important part of Electric Power System. Regular inspection is essential to ensure the safe operation of the substation. In recent years, with the development of the Electric Power System, the manual inspection is gradually replaced by the robot inspection. In unattended substation, inspection robot can inspect electric equipment. The function of autonomous movement under various conditions of the inspection robot is the guarantee to fulfill the inspection task. The vision navigation and path planning of the substation inspection robot is the key technology to achieve this function. This paper focus on visual navigation and path planning technology of the substation inspection robot.Firstly, this paper analyses the function requirements of substation inspection robot and designs the overall structure of the substation inspection robot system. Through the analysis of existing vision navigation and path planning technology, a kind of vision navigation scheme based on the guide line and a kind of local path planning scheme based on the artificial potential field are introduced in this paper.Secondly, in the visual navigation, robot visual navigation can be divided into image preprocessing, guide line detection, guide line tracking. In the part of image preprocessing, this paper firstly describes the commonly used image preprocessing technologies such as image color segmentation and image filtering. Based on the analysis of existing technologies and the comparison of the experiments, a set of the image preprocessing algorithm suitable for this paper is designed. In the part of guide line detection, this paper put forward a kind of guide line detection algorithm based on least square method. The algorithm includes the analysis of guide line model, the classification of guide line model and the least squares curve fitting. The experiment of guide line detection proves that the algorithm has high rate detection accuracy. In the part of guide line tracking, this paper uses the guiding line tracking algorithm based on fuzzy control. The algorithm includes the calculation of navigation parameter, the analysis of mobile robot motion model and the robot fuzzy control. This paper introduces the robot which is used as a experiment platform and then an experiment of guide line tracking is carried out in this robot. The experiment proves that the algorithm can effectively control the robot to walk along the guide line.Thirdly, in the path planning, this paper introduces the principle of traditional artificial potential field and then analyses the problem of target unreachable and local minimum. Aiming at the problem of target unreachable, this paper adopts the method of modifying repulsive force function. And then an improved escape force method is put forward to solve the problem of local minimum which occurs when the target.the obstacle and the robot are in the straight line. An obstacles connection method is used to solve the the problem of local minimum which caused by multiple obstacles. Based on the analysis of the environment of the substation, this paper establish a artificial potential field model. The experiment of traditional artificial potential field and the improved algorithm in this model proves that the improved algorithm can effectively solve the problems existing in the artificial potential field method.