A novel model for linear dynamic system with random delays. (January 2019)

• Record Type: Journal Article
• Authors: Pang, Naijiao; Luo, Yingting; Zhu, Yunmin
• Is Part Of: Automatica. Volume 99(2019)
• Page Start: 346
• Page End: 351
• Publication Date: 2019-01
• Subjects: Random parameters matrices -- Kalman filtering -- Random delay -- Conditional probabilities -- Markov chain; Automatic control -- Periodicals; Automation -- Periodicals; 629.805
• DOI: 10.1016/j.automatica.2018.10.057
• Languages: English
• ISSNs: 0005-1098

Abstract

Abstract: This paper explores a novel model to describe a linear dynamic system with random delays. Compared with existing studies, the probabilities of random delays in the novel model are calculated by conditional probabilities. Therefore, the process and measurement noises in the new model for random delay problems are infinitely correlated. By treating the model as a random parameter matrix Kalman filter and a random parameter matrix Kalman filter with one-step correlated noises approximately, the new state estimators are presented. The numerical examples show that these new estimators work better than the existing algorithm in many cases.