Recurrent RBFN-based fuzzy neural network control for X-Y-theta motion control stage using linear ultrasonic motors

Area of Science:

• Robotics and Control Systems
• Artificial Intelligence
• Neural Networks

Background:

• Precise contour tracking is crucial for advanced manufacturing and robotics.
• Linear ultrasonic motors (LUSMs) offer high precision but require sophisticated control.
• Existing control systems may struggle with uncertainties and dynamic contour variations.

Purpose of the Study:

• To develop a robust and adaptive control system for X-Y-theta motion stages.
• To enhance contour tracking accuracy using linear ultrasonic motors.
• To integrate self-constructing fuzzy neural networks, recurrent neural networks, and radial basis function networks.

Main Methods:

• A recurrent radial basis function network (RBFN) based fuzzy neural network (FNN) was designed.
• The control system combined self-constructing fuzzy neural network (SCFNN), recurrent neural network (RNN), and RBFN.
• Concurrent and on-line structure and parameter learning were implemented.
• Structure learning utilized input space partitioning; parameter learning employed supervised gradient descent with a delta adaptation law.

Main Results:

• The proposed recurrent RBFN-based FNN control system successfully controlled the position of an X-Y-theta motion control stage.
• The system demonstrated effective tracking of various contours.
• Experimental results confirmed robust dynamic behaviors of the control system in the presence of uncertainties.

Conclusions:

• The developed recurrent RBFN-based FNN control system provides a robust solution for contour tracking with LUSMs.
• The concurrent on-line learning mechanism enhances adaptability and performance.
• This approach offers significant potential for high-precision motion control applications.