ESO-based adaptive robust control of dual motor driving servo system

Based on extended state observer (ESO), we propose an adaptive robust control (ARC) for a dual motor driving servo system, in which there exist nonlinearities affecting control performance. To apply ESO and estimate the lumped uncertainty online, backlash and friction are analyzed and the nonlinearmodel of the plant is derived. We achieve several control objectives. First, the bias torque is considered in order to eliminate the effect of backlash. Second, the speed feedback is used to maintain the speed synchronization of motors. Then, to achieve feedforward control, finite-time ESO is designed to estimate the unknown nonlinearities online. Furthermore, the ESO-based adaptive robust controller is designed to guarantee L_∞ of tracking error by an initialization method, maintaining the transient performance of tracking behavior. Finally, extensive experimental results on a practical test rig validate the effectiveness of our proposed method.