Abstract
High-voltage inspection robot can detect and identify various obstacles while it is walking along the high-voltage line and plan specific bypassing strategy for different types of obstacles. In this paper vision-based obstacle recognition algorithms are studied based on visual sensors specifically for the unique features of 500kv high-voltage wire. Considering the texture features owned by high-voltage wire in a complex background, the algorithms combining grey co-occurrence matrix (GLCM) and fuzzy C- means clustering algorithms (FCM) are used to extract the texture features and segment the high-voltage wire. Then the image HSV features are used to achieve recognition of the driving wheels for independent hanging-on-line operation by the inspection robot. Also SVM classification is combined with additional structural constraints and type of obstacle can be accurately identified. The experimental results show that the algorithms can reliably identify the torsional damper, spacers and suspension clamp. The robot can be guided to bypass obstacles.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have