Robust control for a tracked mobile robot based on a finite-time convergence zeroing neural network.

Area of Science:

• Robotics
• Control Systems
• Artificial Intelligence

Background:

• Tracked mobile robots are complex non-linear systems, posing significant challenges for accurate trajectory tracking.
• Existing control methods often struggle with the inherent non-linear dynamics and potential disturbances affecting these robots.

Purpose of the Study:

• To develop and validate a novel neural network control strategy for enhancing trajectory tracking in tracked mobile robots.
• To design a robust control system capable of handling non-linear dynamics and external disturbances.

Main Methods:

• A new fractional exponential activation function was designed for a zeroing neural network.
• An implicit derivative dynamic model was developed, termed the finite-time convergence zeroing neural network.
• Lyapunov stability theory was employed for model analysis, determining convergence time bounds and investigating robustness.

Main Results:

• Successful trajectory tracking of an eight-shaped path was achieved in numerical experiments.
• The proposed finite-time convergence zeroing neural network demonstrated effectiveness and superiority compared to existing methods.
• The model's robustness was validated under various error disturbance conditions.

Conclusions:

• The finite-time convergence zeroing neural network provides an effective solution for the trajectory tracking problem in tracked mobile robots.
• The proposed model offers enhanced performance and robustness, making it suitable for real-world applications, including disturbed environments.