Show simple item record

dc.contributor.author
Lavaei, Abolfazl
dc.contributor.author
Zhong, Bingzhuo
dc.contributor.author
Caccamo, Marco
dc.contributor.author
Zamani, Majid
dc.date.accessioned
2021-08-03T09:29:33Z
dc.date.available
2021-07-30T02:41:36Z
dc.date.available
2021-08-03T09:29:33Z
dc.date.issued
2021-05
dc.identifier.isbn
978-1-4503-8399-8
en_US
dc.identifier.other
10.1145/3457335.3461705
en_US
dc.identifier.uri
http://hdl.handle.net/20.500.11850/498481
dc.description.abstract
Artificial intelligence-based (a.k.a. AI-based) controllers have received significant attentions in the past few years due to their broad applications in cyber-physical systems (CPSs) to accomplish complex control missions. However, guaranteeing safety and reliability of CPSs equipped with this kind of (uncertified) controllers is currently very challenging, which is of vital importance in many real-life safety-critical applications. To cope with this difficulty, we propose a Safe-visor architecture for sandboxing AI-based controllers in stochastic CPSs. The proposed framework contains (i) a history-based supervisor which checks inputs from the AI-based controller and makes compromise between functionality and safety of the system, and (ii) a safety advisor that provides fallback when the AI-based controller endangers the safety of the system. By employing this architecture, we provide formal probabilistic guarantees on the satisfaction of those classes of safety specifications which can be represented by the accepting languages of deterministic finite automata (DFA), while AI-based controllers can still be employed in the control loop even though they are not reliable.
en_US
dc.language.iso
en
en_US
dc.publisher
Association for Computing Machinery
en_US
dc.subject
Trustworthy AI
en_US
dc.subject
Safe-visor architecture
en_US
dc.subject
AI-based controllers
en_US
dc.subject
Stochastic cyber-physical systems
en_US
dc.subject
Artificial intelligence
en_US
dc.title
Towards trustworthy AI: Safe-visor architecture for uncertified controllers in stochastic cyber-physical systems
en_US
dc.type
Conference Paper
dc.date.published
2021-05-19
ethz.book.title
Proceedings of the Workshop on Computation-Aware Algorithmic Design for Cyber-Physical Systems (CAADCPS '21)
en_US
ethz.pages.start
7
en_US
ethz.pages.end
8
en_US
ethz.event
2021 Workshop on Computation-Aware Algorithmic Design for Cyber-Physical Systems (CAADCPS 2021)
en_US
ethz.event.location
Online
en_US
ethz.event.date
May 18, 2021
en_US
ethz.notes
A satellite workshop of the 2021 CPS-IoT week, May 18-21, 2021.
en_US
ethz.identifier.scopus
ethz.publication.place
New York, NY
en_US
ethz.publication.status
published
en_US
ethz.date.deposited
2021-07-30T02:41:48Z
ethz.source
SCOPUS
ethz.eth
yes
en_US
ethz.availability
Metadata only
en_US
ethz.rosetta.installDate
2021-08-03T09:29:40Z
ethz.rosetta.lastUpdated
2021-08-03T09:29:40Z
ethz.rosetta.versionExported
true
ethz.COinS
ctx_ver=Z39.88-2004&rft_val_fmt=info:ofi/fmt:kev:mtx:journal&rft.atitle=Towards%20trustworthy%20AI:%20Safe-visor%20architecture%20for%20uncertified%20controllers%20in%20stochastic%20cyber-physical%20systems&rft.date=2021-05&rft.spage=7&rft.epage=8&rft.au=Lavaei,%20Abolfazl&Zhong,%20Bingzhuo&Caccamo,%20Marco&Zamani,%20Majid&rft.isbn=978-1-4503-8399-8&rft.genre=proceeding&rft_id=info:doi/10.1145/3457335.3461705&rft.btitle=Proceedings%20of%20the%20Workshop%20on%20Computation-Aware%20Algorithmic%20Design%20for%20Cyber-Physical%20Systems%20(CAADCPS%20'21)
 Search print copy at ETH Library

Files in this item

FilesSizeFormatOpen in viewer

There are no files associated with this item.

Publication type

Show simple item record