Enhanced Data-Driven Optimal Terminal ILC Using Current Iteration Control Knowledge

Area of Science:

• Control Systems Engineering
• Nonlinear System Dynamics
• Data-Driven Control

Background:

• Iterative learning control (ILC) is crucial for repetitive tasks in discrete-time systems.
• Nonlinear and nonaffine systems present significant control challenges.
• Existing ILC methods often require precise system models.

Purpose of the Study:

• To propose an enhanced data-driven optimal terminal iterative learning control (E-DDOTILC) strategy.
• To address control challenges in nonlinear and nonaffine discrete-time systems.
• To develop a controller that relies solely on input-output data.

Main Methods:

• A dynamical linearization approach is employed to transform system dynamics.
• An ILC law with nonlinear learning gain is designed.
• A parameter updating law iteratively estimates unknown system partial derivatives.
• Input signals are updated using terminal tracking errors and previous input signals.

Main Results:

• The proposed E-DDOTILC method effectively controls nonlinear and nonaffine discrete-time systems.
• The approach demonstrates monotonic convergence of tracking errors.
• Simulation results validate the controller's effectiveness.

Conclusions:

• The E-DDOTILC is a robust, data-driven control strategy for nonlinear discrete-time systems.
• The method eliminates the need for explicit system models, relying only on I/O data.
• The proposed approach offers a viable solution for complex control problems.