0598043 20250317090005.720240910235959.9 85202902631 001296047100114 10.1016/j.ifacol.2024.07.297 6 s. E cav_un_epca*05984482405-8963676-681AmsterdamElsevier2024 active fault detection tensor decomposition value iteration cav_un_auth*0472250 Punčochář I. CZ 40 cav_un_auth*0434606 Straka O. CZ 30 cav_un_auth*0101212 Tichavský Petr UTIA-B Stochastická informatika Department of Stochastic Informatics SI SI Department of Stochastic Informatics 30 Ústav teorie informace a automatizace AV ČR, v. v. i. https://library.utia.cas.cz/separaty/2024/SI/tichavsky-0598043.pdf https://www.sciencedirect.com/science/article/pii/S2405896324003811?via%3Dihub GA22-11101S GA ČR CZ cav_un_auth*0435406 The paper deals with the use of tensor train decomposition for active fault detection of stochastic systems. The faulty and faulty-free behavior of the system is represented using multiple models. A non-rectangular grid in the space of the information state is proposed. An approximate Bellman function is represented using the tensor train decomposition and is calculated using the value iteration algorithm. cav_un_auth*0472251 IFAC Symposium on Fault Detection, Supervision, and Safety for Technical Processes, Safe Process 2024 /12./ 20240604 20240607 Ferrara IT IN 20000 20200 20204 2025 3 PR RVO:67985556 https://hdl.handle.net/11104/0356113 S 0.354 Q4 2024 85202902631 SCOPUS PUBMED 001296047100114 WOS cav_un_epca*0598448 IFAC-PapersOnLine. Volume 58, Issue 4 -12th IFAC Symposium on Fault Detection, Supervision and Safety for Technical Processes SAFEPROCESS 2024 Elsevier 2024 Amsterdam 676 681 2405-8963