Event-triggered state estimator design for unknown input and noise-correlated random system

Area of Science:

• Control Systems Engineering
• Stochastic Systems Analysis
• Signal Processing

Background:

• Stochastic systems often involve unknown inputs and correlated noise, complicating accurate state estimation.
• Traditional state estimators can be resource-intensive, leading to high data transmission rates and reduced sensor longevity.

Purpose of the Study:

• To design an event-driven state estimator for stochastic systems with unknown inputs and correlated measurement noise.
• To develop a transmission strategy that balances estimation accuracy with resource efficiency and sensor battery life.

Main Methods:

• Utilized random stability theory and Lyapunov functions to derive the event-triggered state estimator's gain.
• Employed output errors for inhibiting unknown inputs and a quadratic performance index for the transmission strategy.
• Analyzed mean square convergence of state estimation errors.

Main Results:

• The event-driven state estimator effectively estimates system states in the presence of uncertainties.
• Achieved a significant reduction in data transmission, extending sensor battery life by approximately 48%.
• Demonstrated reduced utilization of network resources.

Conclusions:

• The proposed event-driven state estimator offers a robust solution for complex stochastic systems.
• The integrated transmission strategy optimizes performance, conserves energy, and minimizes network load.
• Validated through numerical simulations, the approach provides practical benefits for real-world applications.