Adaptive Control of Nonlinear MIMO System With Orthogonal Endocrine Intelligent Controller.

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Nonlinear Dynamics

Background:

• Traditional controllers struggle with nonlinear, complex dynamical systems.
• Intelligent control offers adaptive and self-regulating capabilities.
• Hybrid approaches can leverage the strengths of multiple AI techniques.

Purpose of the Study:

• To propose and evaluate a new intelligent hybrid controller.
• To enhance self-regulation and nonlinear approximation capabilities.
• To improve tracking performance in complex dynamical systems.

Main Methods:

• Developed an intelligent hybrid controller integrating Orthogonal Endocrine Neural Network (OENN) and Orthogonal Endocrine ANFIS (OEANFIS).
• Utilized Chebyshev orthogonal functions for nonlinear approximations.
• Implemented a hybrid learning algorithm with online and offline procedures and adaptive learning rates.
• Experimentally validated on a nonlinear multiple-input-multiple-output two-rotor aerodynamical system.

Main Results:

• The proposed controller demonstrated significantly superior online tracking performance compared to other intelligent control logics and PID controllers.
• The controller exhibited enhanced self-regulation capabilities due to artificial endocrine influence.
• Chebyshev orthogonal functions contributed to computational simplicity and accuracy.

Conclusions:

• The novel OENN-OEANFIS hybrid controller offers advanced performance for nonlinear dynamical systems.
• While computationally intensive, its tracking accuracy justifies its use when performance is paramount.
• The controller's self-regulation mechanisms effectively handle external disturbances.