A sensorless method for predicting force-induced deformation and surface waviness in robotic milling

Process monitoring is essential to enable process parameter optimization, deformation prediction, and fault diagnosis in robotic milling. However, expensive costs and installation requirements limit the use of industrial sensors in machining process. This paper proposed a sensorless method to predict force-induced deformation and surface waviness. First, the tracking errors of tooltip was calculated based on the robot joint tracking errors and the robot kinematic model. Subsequently, the idle running and cutting process signals monitored by the robot controller were used to calculate the cutting force acting on tooltip based on Kalman filter and robot static model. On this base, the force-induced deformation, considering the posture error of the robot flange coordinate system, was calculated using the estimated milling force and the flexible model. Finally, the effectiveness of the proposed method was verified by a series of cutting experiments.