Stiffness Parameter Identification and Cutting-Force-Induced Error Compensation of an Adsorption Machining Robot

Owing to the advantages of excellent flexibility and accessibility, robots have attracted extensive attention in the field of machining. However, due to their relatively low rigidity, the cutting-force-induced error is the main obstacle to their application. To compensate for the error, an accurate stiffness model is a premise, so it is required to identify the stiffness parameter through experiments, which remains a challenging issue for parallel robots because their component compliance has a complex effect on the robot stiffness due to their complex multiclosed-loop architecture. In this article, an adsorption machining robot with a parallel configuration is presented. An experiment-based stiffness parameter identification method is proposed to obtain an accurate stiffness model of the robot through experiment. To predict the external load acting on the robot end-effector when machining, an analytical cutting force model is established. With the stiffness model and cutting force model, by modifying the NC program offline based on the mirror compensation method, the cutting-force-induced error is compensated. Finally, machining comparison experiments are conducted on an S-shaped workpiece to verify the effectiveness of the proposed method. The results demonstrate that the dimensional accuracy of the surface is improved significantly with the proposed method.