Secant iterative learning control of ultrasonic motor

Area of Science:

• Control Engineering
• Robotics
• Mechatronics

Background:

• Traditional Newton iterative learning laws face challenges in determining differential terms, impacting control performance.
• Ultrasonic motors exhibit nonlinear characteristics that complicate precise speed control.

Purpose of the Study:

• To propose a novel secant iterative learning control strategy to overcome limitations of Newton methods.
• To enhance the speed control performance of ultrasonic motors through an improved learning control approach.
• To develop practical improvements and adaptive gain adjustment for real-world applications.

Main Methods:

• The secant method from numerical analysis is employed to formulate a new iterative learning control law.
• The strategy is applied to the speed control of an ultrasonic motor.
• Practical improvements, including a method for determining learning gain based on experimental data and online adaptive adjustment, are introduced.

Main Results:

• The secant iterative learning control strategy demonstrates effectiveness in ultrasonic motor speed control.
• Control performance progressively improves with iterative learning, showing enhanced convergence speed and superior performance.
• The proposed adaptive gain adjustment effectively compensates for the nonlinear characteristics of the ultrasonic motor.

Conclusions:

• The secant iterative learning control strategy offers a viable and effective alternative to Newton iterative learning laws.
• The practical improvements and adaptive gain adjustment enhance the robustness and applicability of the control strategy.
• The developed method significantly improves the speed control performance of ultrasonic motors.