Robotic sorting method based on machine vision and improved genetic algorithm

A robotic sorting method based on machine vision and beetle antennae search algorithm improved genetic algorithm is proposed for the fast-sorting demand of mechanical parts. The sorting method starts with pre-processing of the part image, then extracts the part images using an image recognition algorithm with Sift feature matching and localizes the target parts using affine transformation. Then, a mathematical model is established for the obtained part positions, and the BAS-GA algorithm is used to solve the mathematical model and obtain the grasping path of the robot to achieve fast sorting of the robot. The experiments show that the BAS-GA algorithm achieves a better finding effect compared with the SA algorithm、the GA algorithm and the PSO-ACO algorithm. And compared with the initial path, the optimized path is shortened by 11%, which indicates that the method can effectively improve the robotic sorting speed.