Adaptive and robust experimental design for linear dynamical models using Kalman filter

被引:2
作者
Strouwen, Arno [1 ]
Nicolai, Bart M. M. [1 ]
Goos, Peter [1 ,2 ]
机构
[1] Katholieke Univ Leuven, Dept Biosyst, Leuven, Belgium
[2] Univ Antwerp, Dept Engn Management, Antwerp, Belgium
关键词
Optimal experimental design; Bayesian experimental design; Adaptive experimental design; Dynamical system; Kalman filter;
D O I
10.1007/s00362-023-01438-9
中图分类号
O21 [概率论与数理统计]; C8 [统计学];
学科分类号
020208 ; 070103 ; 0714 ;
摘要
Current experimental design techniques for dynamical systems often only incorporate measurement noise, while dynamical systems also involve process noise. To construct experimental designs we need to quantify their information content. The Fisher information matrix is a popular tool to do so. Calculating the Fisher information matrix for linear dynamical systems with both process and measurement noise involves estimating the uncertain dynamical states using a Kalman filter. The Fisher information matrix, however, depends on the true but unknown model parameters. In this paper we combine two methods to solve this issue and develop a robust experimental design methodology. First, Bayesian experimental design averages the Fisher information matrix over a prior distribution of possible model parameter values. Second, adaptive experimental design allows for this information to be updated as measurements are being gathered. This updated information is then used to adapt the remainder of the design.
引用
收藏
页码:1209 / 1231
页数:23
相关论文
共 38 条
[1]   Optimal design of pharmacokinetic studies described by stochastic differential equations [J].
Anisimov, Vladimir V. ;
Fedorov, Valerii V. ;
Leonov, Sergei L. .
MODA 8 - ADVANCES IN MODEL-ORIENTED DESIGN AND ANALYSIS, 2007, :9-+
[2]  
ATKINSON A. C., 2007, Oxford Statistical Science Series, V34
[3]   Julia: A Fresh Approach to Numerical Computing [J].
Bezanson, Jeff ;
Edelman, Alan ;
Karpinski, Stefan ;
Shah, Viral B. .
SIAM REVIEW, 2017, 59 (01) :65-98
[4]   On computing the expected Fisher information matrix for state-space model parameters [J].
Cavanaugh, JE ;
Shumway, RH .
STATISTICS & PROBABILITY LETTERS, 1996, 26 (04) :347-355
[5]   Bayesian experimental design: A review [J].
Chaloner, K ;
Verdinelli, I .
STATISTICAL SCIENCE, 1995, 10 (03) :273-304
[6]   Mixture Kalman filters [J].
Chen, R ;
Liu, JS .
JOURNAL OF THE ROYAL STATISTICAL SOCIETY SERIES B-STATISTICAL METHODOLOGY, 2000, 62 :493-508
[7]  
EFRON B, 1978, BIOMETRIKA, V65, P457, DOI 10.1093/biomet/65.3.457
[8]   OPTIMUM ALLOCATION IN LINEAR REGRESSION THEORY [J].
ELFVING, G .
ANNALS OF MATHEMATICAL STATISTICS, 1952, 23 (02) :255-262
[9]   Pharmacokinetic Studies Described by Stochastic Differential Equations: Optimal Design for Systems with Positive Trajectories [J].
Fedorov, Valerii V. ;
Leonov, Sergei L. ;
Vasiliev, Vyacheslav A. .
MODA 9 - ADVANCES IN MODEL-ORIENTED DESIGN AND ANALYSIS, 2010, :73-80
[10]  
Fedorov Valerii Vadimovich, 2013, Theory of Optimal Experiments