Posture Optimization in Robotic Flat-End Milling Based on Sequential Quadratic Programming

Abstract

Robotic flat-end milling of complex surfaces offers advantages such as high flexibility and high machining efficiency. In the process of planning the toolpath based on the cutter contact path, the robot functional redundancy and the tool orientation need to be solved carefully. This paper presents a posture optimization method for robotic flat-end milling. Taking the weighted sum of the machining width and the toolpath smoothness performance criterion as the objective function, an optimization model considering the joint limits and gouging avoidance is established. An efficient algorithm based on sequential quadratic programming is proposed to solve this nonconvex problem. During the execution of the algorithm, the machining width is efficiently calculated by an iterative method based on conformal geometric algebra, while its derivatives are approximated analytically. Simulations and experiments demonstrate that the presented technique can resolve the tool axis direction and the robot redundancy effectively to increase the machining width and improve the toolpath smoothness, thus reducing the time for machining and improving the surface quality.