Answers are grounded in this paper and cite their sources.
Ultrasonic motor (USM) has many advantages such as small, light, high torque at low speed, and so on. However, USM contains non-linear elements caused by various frictions, changes of temperature, fluctuation of load-mass, and so on. Therefore, it is difficult to accomplish satisfactory control performance in USM by using conventional control method such as PI control. In this paper, we propose a practical design method of neural networks (NN) controller for USM. In this design method, learning of NN is started from using input-output data of PI control for USM to construct NN controller. Once NN controller has been constructed, self-learning of NN is adopted to improve the control performance of USM. The effectiveness of proposed design method is confirmed by experiments using the existent USM.
No takes yet. Share an insight, caveat, or question.
Oka et al. (2001) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: