Tool Positioning Error Minimization during Robotic Milling Based on the Genetic Algorithm Technique

Abstract

Compared to CNC machines, robotic milling has performance limitations such as accuracy and quality. The main source of the robot’s inaccuracy during machining is the flexibility of its parts (body or joints). This error disturbs the movement of the end effector, affecting the part’s surface finish. In order to improve the robot’s accuracy and minimize the positioning error of the end effector during the milling operation, this paper presents, first, a method based on the elasto-static model to predict the Cartesian deflection of the end effector of a three DOF redundant planar robot, and second, optimization techniques with original objective functions based on the single and multi-objective genetic algorithm, which will be presented and compared. The programming of the two methods and the results of the study will be done using MATLAB software. The analysis of simulation results of the two optimization techniques GA and MOGA revealed that the tool configuration and cutting parameters used for robotic milling have a direct influence on the robot's path accuracy and milling performance. Whereas for a φ0=69.6, φf=72.43 the maximum tool deviation in its path is Δxmax ≈ |0.125| mm with a maximum roughness profile height Ra = 1600 μm. While the positioning error is said to be minimal Δxmin ≈ |0.025| when φ0= -38.67, φf = -35.92, and the roughness Ra= 25 μm.