Skip to Main content Skip to Navigation

Design of resilient networked control systems

Abstract : This thesis presents several methodological contributions for the analysis and synthesis of networked control systems subjected to multiple network-induced factors, such as packet dropouts, time delays, finite word length effects. Specifically, the following researches are considered in the thesis: 1) the stochastic LQ control under asymptotic tracking is considered for multiple-input-multiple-output discrete-time system over lossy channels, where the controller-coding co-design approach is adopted to take full advantage of the network resources; 2) the robust filtering problem is considered also over lossy channels, where an adjoint operator-based approach is proposed to solve the considered H∞ filtering problem and a modified discrete-time algebraic Riccati equation-based solution is further proposed to design the corresponding H∞ filter; 3) taking into consideration the networked estimators, implemented in a multi-subsystem way, a methodology making possible both their design and the choice of the most resilient realization is introduced. This methodology is based on a descriptor model used to describe in a unifying framework all the equivalent realizations, with the internal time delays and finite word length effects. Based on this descriptor model, the design and realization optimization for the estimator is thus achieved. To verify the effectiveness of the proposed design and realization methods mentioned above, a system with two cooperative mobile robots is introduced as an illustrative example. Several simulations permit us to test and validate our different methodological contributions on a concrete and realistic problem.
Document type :
Complete list of metadata
Contributor : Abes Star :  Contact
Submitted on : Monday, July 26, 2021 - 12:51:10 PM
Last modification on : Wednesday, January 19, 2022 - 3:48:24 PM
Long-term archiving on: : Wednesday, October 27, 2021 - 6:17:16 PM


Version validated by the jury (STAR)


  • HAL Id : tel-03299347, version 1


Rongyao Ling. Design of resilient networked control systems. Automatic. Ecole nationale supérieure Mines-Télécom Atlantique; Zhejiang University (Hangzhou, Chine), 2020. English. ⟨NNT : 2020IMTA0193⟩. ⟨tel-03299347⟩



Les métriques sont temporairement indisponibles