Dual Event-Triggered Remote Information-Based State Estimation With Measurement Outliers

Le résumé fourni par la source

This paper examines event-triggered remote state estimation in scenarios with measurement outliers. To address the issue of trigger failures caused by outliers, a dual event-triggered criterion is proposed to ensure the transmission of informative and reliable data. Considering that feedback values in closed-loop structures may be corrupted by outliers, the trigger vector is constructed using local sensor information, rather than feedback from the estimator. For state estimation, a minimum-mean-square-error (MMSE) estimation method is first deduced within the information filter framework. To address the exponential growth in posterior Gaussian components induced by the dual trigger mechanism, a mode dependent time-varying transition matrix is introduced, enabling an interacting multiple model (IMM) approximation approach based on conditional transition probabilities. Moreover, the boundedness of error covariance is theoretically established. Simulations validate that the proposed algorithm exhibits strong robustness under measurement outliers and achieves good performance.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.