Welding Seam Recognition Technology of Welding Robot Based on A Novel Multi-Path Neural Network Algorithm

Abstract

Robot welding technology includes independent planning, welding seam position detection, automatic welding seam tracking, etc. Welding seam recognition is a very important link. Traditional algorithms are far inferior to artificial intelligence algorithms in the welding seam recognition. This paper proposes a novel multi-path neural network algorithm, which performs well in the self-collected welding seam recognition data set called WL_HIST. The accuracy of welding seam recognition is as high as 95.3%, which is much higher than 65.3% of the traditional HOG manual feature extraction algorithm. The results show that the deep learning algorithm has a significant and outstanding performance in the welding robot recognition technology.