Distributed H ∞-Constraint Robust Estimator for Multi-Sensor Networked Hybrid Uncertain Systems

• Published in: IEEE transactions on network science and engineering Vol. 8; no. 4; p. 3335
• Main Authors: Xia, Juan, Gao, Shesheng, Guo, Li, Qi, Xiaomin, Gao, Bingbing, Zhang, Jiahao
• Format: Journal Article
• Language: English
• Published: Piscataway The Institute of Electrical and Electronics Engineers, Inc. (IEEE) 2021
• Subjects: Data integration; H infinity; Hybrid systems; Mathematical models; Multisensor fusion; Noise measurement; Nonlinear systems; Random variables; Robustness (mathematics); State estimation; Tracking; Uncertainty; Upper bounds
• DOI: 10.1109/TNSE.2021.3112669

Due to the complex communication channel environment, uncertainties, including random measurement delays, missing measurements and uncertain measurement noise variances, often exist in multi-sensor networked hybrid uncertain systems. These uncertainties will lead to poor accuracy and robustness for traditional information fusion methods. Hence, this paper presents a distributed robust estimator based on the H ∞ constraint to solve the mentioned issues in sensor networked systems. The nonlinear pseudomeasurement model is first constructed with Bernoulli distributed random variables to describe the mentioned uncertainties in the communication channels. Then, a sufficient condition with a local optimal upper bound on estimation error covariance is derived rigorously with respect to the robustness definition, which is based on the H[Formula Omitted] constraint inequality to improve the robustness of state estimation. Furthermore, the proposed estimator is proven to be stable through boundedness analysis for the upper bound matrix. The numerical simulation and the target tracking experiment are finally applied to evaluate the effectiveness of the novel fusion estimator.