Contents I. Introduction
Nowadays, the servo motor drive system is widely utilized in industries, such as robotic arms, high-precision metal processing systems, and fast response manufacturing systems [1]–[5]. Compared to other transmission systems, servo motors stand out for their ability to provide precise position control, adaptability to various mechanical structures, high-speed performance, and rapid response capabilities [6]. Additionally, the inherent stability of servo motors helps minimize oscillations, promoting reliable operation. By reacting quickly and accurately to control signals, servo motors reduce response times and enhance overall accuracy in the control process [7]–[8]. The notable challenge lies in accurately controlling the position in the servo motor drive system amidst uncertainties like friction, dynamic loads, or external forces. Numerous research works have been conducted on the design of servo motor control systems, ranging from simple to complex structures [8]–[19]. In documents [9]–[10], pure proportional-integral control was employed with a straightforward and computationally friendly algorithm; however, the system was susceptible to noise and suitable for specific objects. In works [11]–[13], model predictive control was applied to servo systems to enhance control performance; yet, determining complex coefficients in the design required a deep understanding of the object. In [14]–[16], optimal control and neural networks capable of learning and processing nonlinear data were developed. The challenge in this design process was the lengthy processing time and computational resource consumption. [17]–[22] implemented PID controller design combined with a fuzzy controller. Each of these control systems had its own advantages, particularly relying solely on simulation results without real-world system testing to assess control quality.
