Improved Genetic Algorithm Bezier Curve Approximation In Cartesian Space Robot Path Planning Application

Cartesian robot path planning is a combination of the robot end posecontrol and curve approximation. Compared with trajectory planning in jointspace, Cartesian space robot path planning in complex curves trackingapplications does not need to pre-teach, end position control accuracy and soon. It is widely used in industry robot.This paper focus on variety studies on Cartesian space robotic pathplanning method. Compared different methods in Cartesian space robot pathcurve approach and different curve type in curve approach, then proposesimprovement ideals. Using genetic algorithms to solve the problems ofcomplex curve description the difficulties. Add Individual calibrationimproved genetic algorithms. Compare with general GA, improved geneticalgorithm convergence is much faster. Meanwhile, it can avoid approachingcurves has singular point. Through simulation experiments, compare withpolynomial interpolation, improved GA has less error under same geneticgeneration and can adjust Bezier point automatically to fit more complexcurve.Cartesian robot control for robot operation may result in ectopic shapedproblem, so form a robot platform for actual running experiments. To provethat the method has more advantage in tracking complex curve than generalrobot teaching.