Nonfragile Output Feedback Tracking Control for Markov Jump Fuzzy Systems Based on Integral Reinforcement Learning...

Area of Science:

• Control Systems Engineering
• Artificial Intelligence
• Nonlinear Dynamics

Background:

• Uncertain Markov jump nonlinear systems pose significant control challenges.
• Takagi-Sugeno fuzzy models are often used to represent such complex systems.
• Designing nonfragile controllers for these systems requires advanced techniques.

Purpose of the Study:

• To propose a novel integral reinforcement learning (RL)-based nonfragile output feedback tracking control algorithm.
• To address uncertain Markov jump nonlinear systems modeled by Takagi-Sugeno fuzzy systems.
• To ensure stochastic asymptotic stability and H∞ performance.

Main Methods:

• Formulating nonfragile control as a zero-sum game between control and disturbance inputs.
• Developing an offline parallel output feedback tracking learning algorithm using RL.
• Designing an online parallel integral RL-based algorithm to relax system information requirements.
• Utilizing Lyapunov stability theory and stochastic analysis for stability and performance guarantees.

Main Results:

• Successfully designed an integral RL-based nonfragile output feedback tracking control algorithm.
• Achieved tracking objectives for the uncertain Markov jump nonlinear systems.
• Ensured stochastic asymptotic stability and expected H∞ performance.
• Demonstrated the algorithm's effectiveness through simulation on a robot arm system.

Conclusions:

• The proposed integral RL-based control algorithm effectively handles uncertain Markov jump nonlinear systems.
• The method guarantees system stability and achieves desired performance metrics.
• The approach is validated by its successful application to a practical robotic system.