Facing the demands of high precision,high efficiency and low cost in the field of modern assembly and manufacturing,the research of industrial laser 3D projection system in the field of assembly and manufacturing has attracted much attention."Parameter calibration" is the key technology of the system,which has a direct impact on the accuracy of projection.In order to obtain more accurate rotation parameters and position parameters and improve the accuracy of industrial laser 3D projection system,the research on the calibration technology of industrial laser 3D projection system is particularly important.In this paper,a calibration method of industrial laser 3D projection system based on stereo vision is proposed,and a calibration model of rotation parameters and position parameters of industrial laser 3D projection system is established to verify the feasibility of the calibration model,and a galvanometer rotation based on BP neural network is proposed.The angle compensation method completes the precision calibration and compensation of the industrial laser 3D projection system.The main research contents of this paper are summarized as follows:1)The basic composition and working principle of industrial laser 3D projection system are studied,the geometric model of 3D projection system of industrial laser is constructed,and the relationship between the parameters of the projection system and the coordinates of projection points is analyzed theoretically.Thus,the vector model of industrial laser 3D projection system is established,which provides theoretical support for the parameter calibration of the subsequent projection system.2)The calibration method of industrial laser 3D projection system based on stereo vision measurement system was proposed,and a model for solving the calibration parameters of industrial laser 3D projection system is established.A “two–step” parameter calibration method is proposed,that is the system rotation parameter calibration model using the Least Square method,Newton iterative algorithm,Levenberg-Marquardt algorithm combined with Quaternion method,Rodrigue Matrix method,and the system position parameter calibration model based on Particle Swarm Optimization algorithm,Breed Particle Swarm Optimization algorithm.A calibration error evaluation model was established to evaluate the calibration accuracy of the system.3)The calibration model of rotation parameters and position parameters is simulated.The simulation results show that the rotation parameter calibration model based on Levenberg-Marquardt algorithm combined with Quaternion method and the position parameter calibration model based on Particle Swarm Optimization algorithm have higher precision,the accuracy and feasibility of the calibration model are verified.4)The rotation angle deviation of biaxial scanning galvanometer in industrial laser 3D projection system is compensated.A compensation model of the rotation angle deviation of the mirror based on back propagation neural network was established,and the precision evaluation method of the model was given.The calibration experiment site of industrial laser 3D projection system was set up;the calibration experiment of industrial laser 3D projection system parameters assisted by stereo vision measurement system was carried out;and the rotation angle deviation of the mirror was compensated.Experimental results show that the calibration method,calibration model and compensation method are feasible. |