Dynamic characteristics and deep reinforcement learning of proportional-integral-differential controller for quadruped stator-based ultrasonic linear motor

• Published in: Japanese Journal of Applied Physics Vol. 63; no. 4; pp. 4 - 14
• Main Authors: Jiang, Yukun, Wang, Fangyi, Sasamura, Tatsuki, Mustafa, Abdullah, Morita, Takeshi
• Format: Journal Article
• Language: English
• Published: Tokyo IOP Publishing 01.04.2024; Japanese Journal of Applied Physics
• Subjects: Algorithms; Controllers; Deep learning; deep reinforcement learning; DRL-PID control; dynamic characteristic; Dynamic characteristics; Electric motors; Electric potential; Electromagnetic interference; Motor stators; Proportional integral derivative; Speed control; Stators; ultrasonic linear motors; Voltage; Weight reduction
• ISSN: 0021-4922; 1347-4065
• DOI: 10.35848/1347-4065/ad2f18

Abstract Ultrasonic linear motors have many advantages, such as small size, light weight, and a lack of electromagnetic interference. In our previous study, a quadruped stator-based ultrasonic linear motor that mimics the motion of an inchworm based on longitudinal and bending modes was proposed. However, our previous studies focused on voltage control, overlooking the influence of phase, which made accurate and sustainable operation difficult. This study measures the dynamic characteristics of this motor using polymer-based rods and combines deep reinforcement learning (DRL) with proportional-integral-differential (PID) algorithms for both position and speed control through phase manipulation. The proposed DRL-PID controller outperforms voltage control in both simulation and real-world experiments.