Robust Sequential Fusion Estimation Based on Adaptive Innovation Event-Triggered Mechanism for Uncertain Networked...

Area of Science:

• Control Engineering
• Signal Processing
• Networked Systems

Background:

• Networked systems face challenges in transmission pressure and robust performance due to uncertain noise variances and correlated noises.
• Existing filtering algorithms can be computationally intensive, especially when dealing with complex error cross-covariance matrices.

Purpose of the Study:

• To develop an adaptive innovation event-triggered mechanism for reducing communication burden in networked systems.
• To design a robust local filtering algorithm for multi-sensor systems with uncertain noise characteristics.
• To propose novel sequential fusion estimation algorithms that avoid complex matrix calculations.

Main Methods:

• An adaptive innovation event-triggered mechanism was designed and integrated into a robust local filtering algorithm.
• Sequential fusion techniques, specifically sequential covariance intersection (SCI) and sequential inverse covariance intersection (SICI), were applied.
• Robustness analysis of the proposed SCI and SICI algorithms was conducted.

Main Results:

• The adaptive innovation event-triggered mechanism effectively reduced the communication burden.
• The robust local filtering algorithm demonstrated effectiveness in handling uncertainties from unknown noise variances.
• The proposed SCI and SICI sequential fusion estimators exhibited good robustness in simulations.

Conclusions:

• The developed adaptive event-triggered mechanism is a viable solution for mitigating transmission pressure in networked systems.
• The robust local filtering approach effectively addresses uncertainties in multi-sensor systems.
• Sequential covariance intersection and sequential inverse covariance intersection provide robust estimation solutions for networked systems.