Reduced-Order Observer-Based Dynamic Event-Triggered Adaptive NN Control for Stochastic Nonlinear Systems Subject to...

Area of Science:

• Control Systems Engineering
• Nonlinear Dynamics
• Stochastic Systems Analysis

Background:

• Stochastic nonlinear systems present significant control challenges due to unknown dynamics, input saturation, and unmeasured states.
• Existing event-triggered control mechanisms often lack adaptability and efficiency in resource-constrained environments.
• Addressing asymmetric input saturation and partial state unavailability requires advanced control strategies.

Purpose of the Study:

• To develop a dynamic event-triggered control (DETC) scheme for stochastic nonlinear systems.
• To improve resource efficiency by minimizing unnecessary data transmissions.
• To handle unknown nonlinearities, input saturation, and partially unmeasured states.

Main Methods:

• Design of a dynamic event-triggered mechanism (DEM) with dynamically adjusted thresholds.
• Utilization of an improved neural network for approximating unknown nonlinear terms, incorporating reconstruction error.
• Construction of an auxiliary system to manage asymmetric input saturation.
• Implementation of a reduced-order observer for estimating unmeasured states.

Main Results:

• The proposed dynamic event-triggered control scheme effectively manages stochastic nonlinear systems.
• The DEM significantly reduces data transmission, enhancing resource efficiency.
• The method successfully compensates for unknown nonlinearities, input saturation, and unmeasured states.
• Theoretical proofs confirm the achievement of desired control objectives.

Conclusions:

• The developed dynamic event-triggered control scheme offers a robust and efficient solution for complex stochastic nonlinear systems.
• The innovative approach demonstrates superior performance in handling system uncertainties and constraints.
• The method's effectiveness is validated through simulations on a one-link manipulator and a ship maneuvering system.