Estimator for Multirate Sampling Systems With Multiple Random Measurement Time Delays

• 成果类型: Article
• 署名作者: Lin, Honglei; Sun, Shuli
• 署名单位: Heilongjiang University
• 刊物名称: IEEE TRANSACTIONS ON AUTOMATIC CONTROL
• ISSN/ISSBN: 0018-9286
• DOI: 10.1109/TAC.2021.3070299
• 发表日期: 2022
• 页码: 1589-1596
• 关键词: Particle measurements; Atmospheric measurements; Noise measurement; Stochastic processes; Time measurement; DELAYS; Delay effects; Multiple random measurement time delay; multirate sampling system; optimal nonaugmented estimator; projection theory

摘要

The optimal linear estimation problem is studied for multirate sampling systems with multiple random measurement time delays (TDs). The considered multirate sampling scheme is that the sensor uniformly samples at a slow rate and the state uniformly updates at a fast rate. Known stochastic variable sequences obeying Bernoulli distributions are adopted to depict random measurement TDs, including missing measurements as a special case. First, using a state iterating method, the original system with multirate sampling and delayed measurements is transformed into a state-space model with single-rate sampling and delay-free measurements at measurement sampling (MS) instants. Then, by utilizing projection theory, a nonaugmented recursive optimal linear state filter is presented based on the established model in the linear minimum variance sense, where the estimators for the process noise are involved. Furthermore, the state estimator at state update instants is achieved through filtering or prediction based on the filter at MS instants. Finally, the centralized fusion estimator and the distributed covariance intersection fusion estimator are proposed for the multisensor case. Simulation research on a vehicle suspension system verifies the effectiveness of the algorithms.