Neural network-based robust integral error sign control for servo motor systems with enhanced disturbance rejection...

Area of Science:

• Control Systems Engineering
• Robotics
• Artificial Intelligence

Background:

• Servo systems face challenges with uncertain dynamics and time-varying disturbances, leading to reduced tracking accuracy.
• Existing control methods struggle to effectively mitigate these performance degradations.

Purpose of the Study:

• To develop a novel adaptive robust control scheme for enhancing servo system tracking accuracy.
• To address uncertain dynamics and unknown time-varying disturbances in servo systems.

Main Methods:

• A new neural network compensator combining reference-driven and error-driven networks was developed.
• A robust integral of the sign of the error (RISE)-based controller was designed for feedback control.
• Lyapunov theory was used to guarantee asymptotic tracking control.

Main Results:

• The proposed method demonstrated superior tracking performance in simulations and experiments.
• The control scheme showed a powerful ability to reject external disturbances.
• Comparative experiments validated the effectiveness across various conditions.

Conclusions:

• The novel adaptive robust control scheme effectively enhances servo system performance.
• The integration of neural networks and RISE provides robust compensation for system uncertainties and disturbances.
• The method offers a promising solution for achieving high-precision tracking in challenging servo applications.