Research on Maintenance Process Technology for Rail Vehicle Carbody Surface Quality

Abstract

Objective After manually polishing the surface of railway vehicle carbodies and subsequently repairing paint defects, because of uncertainties in the vehicle operating environment and varying degrees of paint deterioration, the surface maintenance process becomes difficult to control, leading to instability in the process. Therefore, it is necessary to study the maintenance technology for the surface quality of railway vehicle carbodies. Method A rail vehicle carbody grinding robot process system is established, employing visual sensors to detect surface paint defects such as yellow spots, scratches, and peeling along the grinding path. The defects are processed through image analysis, and converted into central coordinates of the paint defect areas. The 'nine-point method' is applied for hand-eye calibration of the robot to calculate the transformation matrix. The central coordinates obtained from image acquisition are transformed into the robot working space coordinates. Using an improved adaptive genetic algorithm based on the shortest path principle, the robot coordinates are optimized for path planning to form the most efficient route for defect grinding on vehicle carbody surface. Result & Conclusion The system automatically extracts defect area central coordinates from captured images and converts them into end-effector movement coordinates in robot workspace. Through the optimization of robot defect grinding paths using the above algorithm, the system improves the efficiency of manual operation while enhancing the stability of surface quality maintenance technology for vehicle carbodies.

Keywords

• rail vehicle
• image processing
• genetic algorithm
• path optimization