Trajectory Planning And Tracking Control Of Mobile Welding Robots

With the promulgation of industry 4.0 and the "made in China 2025",our country's indu-strial production towards the development of automation,intelligence,information technolog y has become an inevitable trend.Welding robot integrates many technologies,such as identification,welding,control and so on,which reflects the development level of the national science and technology.In this paper,the mobile welding robot is taken as the research object.the algorithm of welding seam recognition,the kinematics and dynamics model of the mobile welding robot,the trajectory planning and tracking control were studied.1.Algorithm of welding seam recognitionThe median and mean filter method were used to filter the weld and the surrounding image to weaken the interference,Then Sobel edge detection algorithm was used to extract the weld edge curve,so as to achieve the purpose of weld identification.2.Kinematics and dynamics modelThe kinematics model of mobile platform was established by analyzing its non-integrity and using odd order transformation.The kinematics model of the manipulator was established by using Denavit-Hartenberg modeling method.Then the kinematics model of the welding robot was established based on these two kinematics model.Finally,the dynamic equation of the mobile welding robot was established by using the Lagrange Mechanics.3.Trajectory planningFirstly,assuming the mobile platform moving in a straight line,the trajectory planning of manipulator was studied.The joint trajectory of manipulator was optimized by gradient descent method and dichotomy algorithm,and the effectiveness of the algorithm was verified by simulation.Then the synchronous planning of mobile platform and manipulator for large curve welding seam were studied.For synchronous planning,firstly studies the Point-to-point planning of manipulator end.By studying point-to-point kinematic characteristics of manipulator end,the essence of planning to choose the center which the mobile platforms moved around was summed.Then the joint motion of the mobile platform and the manipulator were planned by the optimization algorithm based on gradient descent method and bisection method.The simulation results showed that the end of the manipulator was better to track the points on the weld track.Then the velocity planning problem of the manipulator end was studied.Then the joint velocity of the left and right wheels of the mobile platform and the joint velocity of the manipulator were planned by the genetic algorithm,and the velocity direction of the manipulator end was consistent with the tangent direction of the weld.,so as to realize the tracking of welding seam.4.Research on mobile robot controller based on sliding mode variable structure controlFirstly,the linear sliding mode surface of mobile welding robot was designed.Then based on the dynamic equations of the mobile welding robot,the control law based on approach law was designed.The chattering of sliding mode variable structure control was weakened by saturation function.The effectiveness of the sliding mode controller was verified by simulation.It not only weakened the chattering phenomenon,but also improved the robustness of the system and the accuracy of trajectory tracking.