Smoothness-oriented path optimization for robotic milling processes

Jingfu Peng

doi:10.1007/s11431-019-1529-x

2020-05-18

PRE

DOI：10.1007/s11431-019-1529-x

原文链接

原文求助

摘要

En 中文

Industrial robots are increasingly used for five-axis machining operations, where the rotation of the end effector along the tool-axis direction is functionally redundant. This functional redundancy should be carefully resolved when planning the robot path according to the tool path generated by a computer-aided manufacturing (CAM) system. Improper planning of the redundancy may cause drastic variations of the joint motions, which could significantly decrease the machining efficiency as well as the machining accuracy. To tackle this problem, this paper presents a new optimization-based methodology to globally resolve the functional redundancy for the robotic milling process. Firstly, a global performance index concerning the smoothness of the robot path at the joint acceleration level is proposed. By minimizing the smoothness performance index while considering the avoidance of joint limits and the singularity and the constraint of the stiffness performance, the resolution of the redundancy is formulated as a constrained optimization problem. To efficiently solve the problem, the sequential linearization programming method is employed to improve the initial solution provided by the conventional graph-based method. Then, simulations for a given tool path are presented. Compared with the graph-based method, the proposed method can generate a smoother robot path in which a significant reduction of the magnitude of the maximum joint acceleration is obtained, resulting in a smoother tool-tip feedrate profile. Finally, the experiment on the robotic milling system is also presented. The results show that the optimized robot path of the proposed method obtains better surface quality and higher machining efficiency, which verifies the effectiveness of the proposed method.

Keyword:

robot milling

functional redundancy

path optimization

smoothness

sequential linear programming