This Paper:Browse 98 Download 0 |
码上扫一扫! |
Distributed order estimation for continuous-time stochastic systems |
XinghuaZhu1,2,ZhixinLiu1,2,XiaomingHu3 |
|
(1 Key Laboratory of Systems and Control, Academy of Mathematics and Systems Science, Beijing 100190, China
2 School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100190, China;3 Department of Mathematics, KTH Royal Institute of Technology, Stockholm 10044, Sweden) |
|
摘要: |
In this paper, we investigate the distributed estimation problem of continuous-time stochastic dynamic systems over sensor
networks when both the system order and parameters are unknown. We propose a local information criterion (LIC) based on
the L0 penalty term. By minimizing LIC at the diffusion time instant and utilizing the continuous-time diffusion least squares
algorithm, we obtain a distributed estimation algorithm to simultaneously estimate the unknown order and the parameters
of the system. By dealing with the effect of the system noises and the coupling relationship between estimation of system
orders and parameters, we establish the almost sure convergence results of the proposed distributed estimation algorithm.
Furthermore, we give a simulation example to verify the effectiveness of the distributed algorithm in estimating the system
order and parameters. |
关键词: Stochastic differential equations · Sensor networks · Distributed order estimation · Cooperative excitation condition · Convergence |
DOI:https://doi.org/10.1007/s11768-023-00190-7 |
|
基金项目:This work was supported by the National Key R&D Program of China (No. 2018YFA0703800), the Natural Science Foundation of China (No. T2293770), the Strategic Priority Research Program of Chinese Academy of Sciences (No. XDA27000000) and the National Science Foundation of Shandong Province (No. ZR2020ZD26). |
|
Distributed order estimation for continuous-time stochastic systems |
Xinghua Zhu1,2,Zhixin Liu1,2,Xiaoming Hu3 |
(1 Key Laboratory of Systems and Control, Academy of Mathematics and Systems Science, Beijing 100190, China
2 School of Mathematical Sciences, University of Chinese Academy of Sciences, Beijing 100190, China;3 Department of Mathematics, KTH Royal Institute of Technology, Stockholm 10044, Sweden) |
Abstract: |
In this paper, we investigate the distributed estimation problem of continuous-time stochastic dynamic systems over sensor
networks when both the system order and parameters are unknown. We propose a local information criterion (LIC) based on
the L0 penalty term. By minimizing LIC at the diffusion time instant and utilizing the continuous-time diffusion least squares
algorithm, we obtain a distributed estimation algorithm to simultaneously estimate the unknown order and the parameters
of the system. By dealing with the effect of the system noises and the coupling relationship between estimation of system
orders and parameters, we establish the almost sure convergence results of the proposed distributed estimation algorithm.
Furthermore, we give a simulation example to verify the effectiveness of the distributed algorithm in estimating the system
order and parameters. |
Key words: Stochastic differential equations · Sensor networks · Distributed order estimation · Cooperative excitation condition · Convergence |