A novel model for linear dynamic system with random delays

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

[1]  E. Yaz,et al.  Linear unbiased state estimation for random models with sensor delay , 1996, Proceedings of 35th IEEE Conference on Decision and Control.

[2]  Yingting Luo.  Networked Multisensor Decision and Estimation Fusion: Based on Advanced Mathematical Methods , 2012 .

[3]  Donghua Wang,et al.  Globally Optimal Multisensor Distributed Random Parameter Matrices Kalman Filtering Fusion with Applications , 2008, Sensors.

[4]  Ilya V. Kolmanovsky,et al.  Optimal control of continuous-time linear systems with a time-varying, random delay , 2001, Syst. Control. Lett..

[5]  Edwin Engin Yaz,et al.  Minimum variance generalized state estimators for multiple sensors with different delay rates , 2007, Signal Process..

[6]  Raquel Caballero-Águila,et al.  A New Estimation Algorithm from Measurements with Multiple-Step Random Delays and Packet Dropouts , 2010 .

[7]  Yunmin Zhu,et al.  Globally optimal Kalman filtering with finite-time correlated noises , 2010, 49th IEEE Conference on Decision and Control (CDC).

[9]  L. Schenato.  Optimal estimation in networked control systems subject to random delay and packet loss , 2006, Proceedings of the 45th IEEE Conference on Decision and Control.

[10]  Edwin Engin Yaz,et al.  State estimation of uncertain nonlinear stochastic systems with general criteria , 2001, Appl. Math. Lett..

[11]  Ling Shi,et al.  Kalman Filtering Over a Packet-Dropping Network: A Probabilistic Perspective , 2010, IEEE Transactions on Automatic Control.

[12]  E. Yaz.  Full and reduced-order observer design for discrete stochastic bilinear systems , 1992 .

[13]  Raquel Caballero-Águila,et al.  Least-squares linear filtering using observations coming from multiple sensors with one- or two-step random delay , 2009, Signal Process..

[14]  Seiichi Nakamori,et al.  Recursive estimators of signals from measurements with stochastic delays using covariance information , 2005, Appl. Math. Comput..

[15]  Zidong Wang,et al.  Robust filtering with stochastic nonlinearities and multiple missing measurements , 2009, Autom..

[16]  Johan Nilsson,et al.  Stochastic Analysis and Control of Real-Time Systems with Random Time Delays , 1996 .