Grinding marks and traces, as well as the over- and under-cutting phenomenon are the severe challenges in robotic abrasive belt grinding of turbine blades and it greatly limits the further application of robotic machining technology in the thin-walled blade fields. In the paper, an active force control method consisting of force/positon and PI/PD controller based on six-dimensional force/torque sensor is introduced to eliminate the grinding marks and traces, and a passive force control method including PID controller based on one-dimensional force sensor is proposed to reduce the over- and under-cutting phenomenon in robotic machining system. Then the Kalman filter information fusion methodology is adopted to combine the active and passive force control methods which could improve the controlled force accuracy and efficiency, as well as avoid the control interference. Finally both the test workpiece and turbine blade are employed to examine and verify the reliability and practicality of the proposed hybrid force control method by achieving the desired surface quality and higher profile precision.

磨痕和痕迹以及过切和欠切现象是涡轮叶片的机器人砂带研磨的严峻挑战，它极大地限制了机器人加工技术在薄壁叶片领域的进一步应用。本文介绍了一种基于力/正位和基于六维力/转矩传感器的PI / PD控制器组成的主动力控制方法，以消除磨削痕迹和痕迹，并提出了一种基于PID控制器的被动力控制方法。提出了三维力传感器，以减少机器人加工系统中的过切和欠切现象。然后采用卡尔曼滤波信息融合方法，将主动和被动力控制方法相结合，可以提高控制力的精度和效率，并避免控制干扰。最后，通过获得期望的表面质量和更高的轮廓精度，测试工件和涡轮叶片都被用来检验和验证所提出的混合力控制方法的可靠性和实用性。