Title :
On the performance of unscented Kalman filter with intermittent observations
Author :
De Carli Foletto, Tanisia ; Franco Moreno, Ubirajara
Author_Institution :
IFF-JC, Farroupilha Fed. Inst., Julio de Castilhos, Brazil
Abstract :
The insertion of the communication network in feedback control loop makes the analysis and design of wireless networked control systems (WNCS) more complex, induces challenges in the form of: delays and packet losses, that degrade the control system´s performance and even cause instability. From the viewpoint of application, the main challenge found in wireless networked control system (WNCS) is the packet loss, thus the use of observers as estimators states has been presented as an alternative solution to treat intermittent measurements. The purpose of this paper is study the statistical convergence properties of the estimation error covariance of unscented Kalman filter with intermittent observation (UKFI), showing the existence of a critical value for the arrival rate of the observations, beyond which a transition to an unbounded state error covariance occurs.
Keywords :
Kalman filters; feedback; radio networks; statistical analysis; telecommunication control; communication network; estimation error covariance; feedback control loop; intermittent measurements; intermittent observations; packet loss; statistical convergence properties; unbounded state error covariance; unscented Kalman filter; wireless networked control systems; Covariance matrices; Estimation; Kalman filters; Noise; Numerical models; Packet loss;
Conference_Titel :
Industrial Informatics (INDIN), 2014 12th IEEE International Conference on
Conference_Location :
Porto Alegre
DOI :
10.1109/INDIN.2014.6945592