Abstract: To address the speed smoothness requirement of a high-precision servo system, a design method for a self-learning variable structure controller for the high-precision servo system is presented. This method weakens the influence of periodic torque fluctuations on the system performance. The stability and robustness of the proposed method are proven. Experimental results show that the controller has good control performance and robustness, and effectively guarantees the speed smoothness requirement of the high-precision servo system. Keywords : variable structure control; self-learning The high-precision servo system requires a high degree of speed stability, and this paper analyzes and studies the speed stability of this system. For high-precision servo systems using mechanical bearings and brushless DC torque motors, the main interfering torques are frictional torque and the ripple torque of the brushless DC motor. Frictional torque primarily affects the speed stability of the system at low speeds, causing low-speed creep. Common methods to overcome frictional torque include increasing servo stiffness and resistance. The motor ripple torque, depending on its generation mechanism, mainly includes two types: cogging torque and electromagnetic torque. Cogging torque is caused by the interaction between the rotor permanent magnet and the stator cogging teeth and is independent of the stator current; electromagnetic torque is generated by the interaction between the rotor permanent magnet and the stator current. Click here to download the original paper.