Profile error-oriented optimization of the feed direction and posture of the end-effector in robotic free-form milling

In robotic milling of free-form surfaces, most existing studies tried to reduce the profile errors by optimizing the robot stiffness. However, the stiffness could not directly and rigorously reflect the milling performance to some content, especially, when the significant influence of feed direction on the profile error was ignored for robotic milling of free-form surfaces with large curvatures. In order to solve these problems, direct optimization of the feed directions and end-effector postures is theoretically formulated to seek the solution of a new profile error -oriented optimization model. This model characterizes the profile error in relation to the robot deformation caused by cutting forces, called force-induced profile error. The force-induced profile error is calculated and reduced at each cutter contact point on the free-form surface by comprehensively considering robot stiffness, free-form surface features, feed directions and cutting forces for generating feed direction and posture of the end-effector. The surface is partitioned into multiple sub-regions, in each of which the principle for determining the initial feed direction is proposed to ensure the smooth milling without abrupt change of feed direction. Robotic milling process of the workpiece with saddle contour is experimented. Feed direction and posture of the end-effector are generated by the proposed method and the existing method for comparative studies. Measured profile errors and photographs of machined surface indicate that the developed method can greatly improve the milling performance.

Automated summary

In the field of robotic milling of free-form surfaces, most studies have focused on optimizing robot stiffness to reduce profile errors. However, this method does not accurately reflect the milling performance, especially when the influence of feed direction on profile error is ignored. To address this issue, a new profile error-oriented optimization model is proposed, which directly optimizes the feed directions and end-effector postures. The model considers robot stiffness, free-form surface features, feed directions, and cutting forces to calculate and reduce the force-induced profile error at each cutter contact point. Experimental results show that this method significantly improves milling performance.