Research On Path Planning Of BIW Welding Robot Based On Secondary Development Of ROBCAD

Welding is one of four automobile production processes, the quality of which directlyaffects the performance of the vehicle. There are3000～5000welding points of BIW,while most of them are welded by robots. Nowadays domestic enterprises mainly relyon planners to plan paths for welding robots, which is low efficient and often affects thebeat because of robot interference. Thus it is necessary to arrange robots’ welding pathsreasonably at welding line simulation planning stage to give full play to the advantagesof manufacturing resources, ensure the welding quality and increase productivity.This article provided a reliable solution for single or multiple welding robots pathplanning and ROBCAD application. First, the structure of BIW and welding lineprocess were researched, three objects which affect the results of the welding robot pathplanning were analyzed in detail. On this basis, the math model (traveling salesmanproblem) for single robot welding path planning was created. According to thecharacteristics of single robot path planning, genetic algorithm was selected as thesolution algorithm and each parameter of it was determined. A single robot’s optimalwelding path of a mainline station was calculated by means of genetic algorithm on theplatform of MATLAB.Second, multiple robots’ welding path planning was studied, the main task andtechnical routes of which were determined, two math models of welding tasksassignment were established and solved by genetic algorithm. Two robots’ weldingtasks of a mainline station were optimized by means of genetic algorithm on theplatform of MATLAB.Finally, secondary development on ROBCAD was done and the correspondingROBUIL programs were analyzed and made. ROBCAD application was developed byVS2008, which include controlling ROBCAD objects by ROBAPI functions andintegrating genetic algorithm to ROBCAD. One robot was taken example to verify therationality of the program.