Development of an adaptive radial basis function neural network estimator-based continuous sliding mode control for...

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Robotics

Background:

• Trajectory tracking of uncertain nonlinear systems presents significant control challenges.
• Existing control methods often suffer from chattering and difficulty in approximating unknown system dynamics.

Purpose of the Study:

• To propose an adaptive neural network-based nonlinear controller for robust trajectory tracking.
• To extend second-order sliding mode control (SOSMC) for nonlinear uncertain systems.
• To eliminate the chattering phenomenon inherent in traditional sliding mode control.

Main Methods:

• Integration of a continuous second-order sliding mode control (CSOSMC) scheme with a radial basis function neural network (RBFNN).
• Development of an adaptive RBFNN estimator to approximate unknown system parameters and disturbances.
• Application of Lyapunov stability theory to prove system convergence and global stability.

Main Results:

• The proposed CSOSMC-ANNE controller effectively approximates unknown nonlinear system dynamics.
• The controller demonstrates superior performance in trajectory tracking compared to conventional methods.
• Chattering is completely eliminated due to the smooth continuous control action.

Conclusions:

• The adaptive neural network-based nonlinear controller offers a robust and effective solution for trajectory tracking.
• The CSOSMC-ANNE methodology provides enhanced performance and stability for uncertain nonlinear systems.
• Validated through simulations on a nonlinear inverted pendulum system.