| Welding robot is a very complex multi-input and multi-output nonlinear systemwith time-varying and strong coupling characteristics. In practice, the position of thetorch often deviates from the theoretical position, causing the welding error, affectingthe quality of welding. This paper mainly deals with the UPR100arc welding robottrajectory tracking intelligent control. The main contents are as follows:First, the link coordinate system of UPR100arc welding robot is establishedaccording to the D-H method, and kinematics equation is derived by homogeneoustransformation matrix and using the Matlab language programming for solving theforward and inverse kinematics equations. The trajectory planning in Cartesian space isbased on forward and inverse kinematics.Secondly, a kinematical simulation model is developed by Matlab/SimMechinics,and then comparing the data of theoretical data with the motion data of trajectoryplanning simulation, the simulation results verify the validity of the simulation modeland the rationality of trajectory planning.Then, the dynamics model of the arc welding robot is established based on theLagrange method, utilizing the powerful parametric modeling capabilities of Pro/E toestablish the three-dimensional modeling of the welding robot, and the parameters likequality, centroid coordinates and moment of inertia of each link are achieved throughquality analysis for each joint. The desired trajectory of each joint is known, the desiredtorque curve of the first three joints also are known, and the simulation results verify thecorrectness of the calculated kinetic parameters which providing a theory basis for theseam tracking.Finally, nonlinear, coupling, uncertainties and other factors have bad influence onthe robot control system. The PD control，sliding mode control and feedforward fuzzyPD control are applied to welding robot trajectory tracking control, the simulationresults of the proposed algorithms are studied and good tracking results are achievedultimately. |
PDF Full Text Request