An adaptive gain based approach for event-triggered state estimation with unknown parameters and sensor...

Area of Science:

• Control Systems Engineering
• Signal Processing
• Networked Systems

Background:

• Distributed estimation in wireless sensor networks is challenging due to nonlinearities and disturbances.
• Existing methods often lack scalability and robustness to network complexity.

Purpose of the Study:

• To develop a scalable distributed state and parameter estimation architecture for discrete-time nonlinear systems.
• To address sensor nonlinearities and stochastic disturbances in wireless sensor networks.
• To ensure the boundedness of estimation error using adaptive coupling and event-triggering.

Main Methods:

• Introduced a novel distributed state estimator architecture with adaptive coupling gains.
• Implemented an event-triggering mechanism for information exchange between sensor nodes.
• Developed an algebraic connectivity-based criterion for uniformly ultimately bounded stability.
• Solved for estimator gains and coupling gains using matrix inequalities.

Main Results:

• The proposed architecture ensures scalable parameter identification irrespective of network complexity.
• Uniformly ultimately bounded stability guarantees boundedness of estimation error.
• Simulation examples demonstrate the effectiveness of the developed estimation architecture.

Conclusions:

• The novel distributed estimation architecture effectively handles nonlinear systems with sensor nonlinearities and disturbances.
• The adaptive coupling and event-triggering mechanisms enhance scalability and robustness.
• The algebraic connectivity criterion provides a theoretical foundation for stability analysis.