Title :
Parameter identification for nonlinear state-space models of a biological network via linearization and robust state estimation
Author :
Jie Xiong ; Tong Zhou
Author_Institution :
Dept. of Autom., Tsinghua Univ., Beijing, China
Abstract :
Developing mathematical models of biological systems and estimating their parameters hold a key to understanding and predicting the dynamic behaviors of biological systems which contain gene regulatory networks, signal transduction pathways, etc. A widely adopted way to model dynamic biological systems is to employ nonlinear state-space models, in which the extended Kalman filter (EKF) is sometimes used for estimating both their states and parameters. However, first-order linearization usually results in modeling errors, but the EKF based method does not take either unmodeled or parametric uncertainty into account. As a result, the estimation performance of the EKF based method may not be satisfactory, such as slow convergence speed and low estimation accuracy. To overcome these problems, a sensitivity penalization based robust state estimator is suggested for estimating parameters of nonlinear biological systems. Simulation results show that the estimation accuracy of the proposed method can be significantly higher than that of the EKF based method.
Keywords :
Kalman filters; linearisation techniques; nonlinear dynamical systems; nonlinear estimation; state estimation; state-space methods; uncertainty handling; EKF based method; dynamic behavior prediction; dynamic biological system model; extended Kalman filter; gene regulatory network; linearization; mathematical model; modeling error; nonlinear biological system; nonlinear state-space model; parameter estimation accuracy; parameter identification; parametric uncertainty; sensitivity penalization based robust state estimation; signal transduction pathway; Accuracy; Biological system modeling; Estimation; Mathematical model; Parameter estimation; Robustness; Uncertainty; Biological Networks; Extended Kalman Filter; Nonlinear State-Space Model; Sensitivity Penalization Based Robust State Estimation;
Conference_Titel :
Control Conference (CCC), 2013 32nd Chinese
Conference_Location :
Xi´an