Output-feedback stochastic nonlinear adaptive control against Markovian jump actuator failures

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Stochastic Processes

Background:

• Designing controllers for stochastic uncertain nonlinear systems with Markovian jump actuator failures is challenging due to nonlinearities, actuator failures, and complex stability analysis.
• Existing methods struggle with mismatched uncertainties, non-vanishing disturbances, and the interconnected terms arising from multiple actuator failures.

Purpose of the Study:

• To investigate and develop an output-feedback adaptive fault-tolerant control strategy for stochastic nonlinear systems experiencing Markovian jump actuator failures.
• To address challenges posed by system uncertainties, stochastic disturbances, and actuator failures in both single-loop and large-scale systems.

Main Methods:

• Re-expressing the system into an output-feedback canonical form by combining plant and failure parameters.
• Utilizing a high-gain K-filter for state reconstruction and employing stochastic Lyapunov design with backstepping for controller synthesis.
• Developing a decentralized control strategy for large-scale systems with uncertain interactions.

Main Results:

• An adaptive fault-tolerant controller is proposed for single-loop systems, effectively handling Markovian jump actuator failures.
• The control strategy is extended to large-scale systems, ensuring global ultimate boundedness of closed-loop signals in probability.
• Tracking errors are shown to be limited to an arbitrarily small residual set in the mean quartic sense.

Conclusions:

• The proposed output-feedback adaptive control method effectively addresses stochastic nonlinear systems with Markovian jump actuator failures.
• The developed decentralized strategy enhances stability and performance in large-scale systems under uncertain interactions.
• Simulation results validate the theoretical findings, confirming the controller's efficacy.