본문

서브메뉴

Secure, Intermittent, and Model-Free Learning Framework for Autonomous Cyber Physical Systems
Secure, Intermittent, and Model-Free Learning Framework for Autonomous Cyber Physical Syst...
Secure, Intermittent, and Model-Free Learning Framework for Autonomous Cyber Physical Systems

상세정보

자료유형  
 학위논문 서양
최종처리일시  
20260202105513
ISBN  
9798263340858
DDC  
658.404
저자명  
Connelly, Prachi Pratyusha.
서명/저자  
Secure, Intermittent, and Model-Free Learning Framework for Autonomous Cyber Physical Systems
발행사항  
[Sl] : Georgia Institute of Technology, 2024
발행사항  
Ann Arbor : ProQuest Dissertations & Theses, 2024
형태사항  
153 p
주기사항  
Source: Dissertations Abstracts International, Volume: 87-05, Section: B.
주기사항  
Advisor: Vamvoudakis, Kyriakos G.;Costello, Mark.
학위논문주기  
Thesis (Ph.D.)--Georgia Institute of Technology, 2024.
초록/해제  
요약Cyber-physical systems (CPS) integrate humans and computational systems to control physical plants using sensing signals and feedback loops. The interconnection, modularity, and exposure to human environments, reveal CPS to attacks from adversaries. In an adversarial, informationconstrained, and model-agnostic environment, reinforcement learning (RL) methods design optimal control policies using reward signals that possess decentralized learning and deployment protocols. Increased continuous data-sharing between sensors, actuators, and the cloud adds complexities to the network leading to degradation of communication effectiveness and increased the possibility of reinforcement signal dropouts, leaving the learning framework with access to a sparse set of rewards or even state measurements.This work presents a Moving Target Defense (MTD) framework, to switch between allowable actuator combinations. MTD allows the input-generation framework to maintain controllability, enhance performance, and safeguard the CPS from. A data-driven controllability algorithm identifies allowable actuator combinations to guarantee controllability. During sensing breakdown, lack of complete state measurements dictates an intermittent framework to approximate value functions and tune embedded neural networks. When continuous communication between the reinforcement channel and the controller is challenged, the proposed framework will develop an self learning algorithm that will provide conditions wherein the internal reinforcement signal is rich enough to allow learning of the optimal control policy and when a trade-off between internal and external reinforcements is utilized. Finally, bionic, human brain-centric learning for CPS is formalised using B.F. Skinner's operant conditioning re-contextualised for learning in CPS, as well as sparse evolutionary training of neuronal configurations informed with physics of the system to approximate solutions to the non-linear Hamilton-Jacobi-Bellman (HJB) is presented.
일반주제명  
Schedules
일반주제명  
Communication channels
일반주제명  
Decision making
일반주제명  
Neural networks
일반주제명  
Military deployment
일반주제명  
Communications networks
일반주제명  
Design
일반주제명  
Cognition & reasoning
일반주제명  
Cognitive psychology
일반주제명  
Electrical engineering
일반주제명  
Military studies
기타저자  
Georgia Institute of Technology.
기본자료저록  
Dissertations Abstracts International. 87-05B.
전자적 위치 및 접속  
로그인 후 원문을 볼 수 있습니다.

MARC

 008260126s2024        us                              c    eng  d
■001000017360365
■00520260202105513
■006m          o    d                
■007cr#unu||||||||
■020    ▼a9798263340858
■035    ▼a(MiAaPQ)AAI32309260
■035    ▼a(MiAaPQ)GeorgiaTech75178
■040    ▼aMiAaPQ▼cMiAaPQ
■0820  ▼a658.404
■1001  ▼aConnelly,  Prachi  Pratyusha.
■24510▼aSecure,  Intermittent,  and  Model-Free  Learning  Framework  for  Autonomous  Cyber  Physical  Systems
■260    ▼a[Sl]▼bGeorgia  Institute  of  Technology▼c2024
■260  1▼aAnn  Arbor▼bProQuest  Dissertations  &  Theses▼c2024
■300    ▼a153  p
■500    ▼aSource:  Dissertations  Abstracts  International,  Volume:  87-05,  Section:  B.
■500    ▼aAdvisor:  Vamvoudakis,  Kyriakos  G.;Costello,  Mark.
■5021  ▼aThesis  (Ph.D.)--Georgia  Institute  of  Technology,  2024.
■520    ▼aCyber-physical  systems  (CPS)  integrate  humans  and  computational  systems  to  control  physical  plants  using  sensing  signals  and  feedback  loops.  The  interconnection,  modularity,  and  exposure  to  human  environments,  reveal  CPS  to  attacks  from  adversaries.  In  an  adversarial,  informationconstrained,  and  model-agnostic  environment,  reinforcement  learning  (RL)  methods  design  optimal  control  policies  using  reward  signals  that  possess  decentralized  learning  and  deployment  protocols.  Increased  continuous  data-sharing  between  sensors,  actuators,  and  the  cloud  adds  complexities  to  the  network  leading  to  degradation  of  communication  effectiveness  and  increased  the  possibility  of  reinforcement  signal  dropouts,  leaving  the  learning  framework  with  access  to  a  sparse  set  of  rewards  or  even  state  measurements.This  work  presents  a  Moving  Target  Defense  (MTD)  framework,  to  switch  between  allowable  actuator  combinations.  MTD  allows  the  input-generation  framework  to  maintain  controllability,  enhance  performance,  and  safeguard  the  CPS  from.  A  data-driven  controllability  algorithm  identifies  allowable  actuator  combinations  to  guarantee  controllability.  During  sensing  breakdown,  lack  of  complete  state  measurements  dictates  an  intermittent  framework  to  approximate  value  functions  and  tune  embedded  neural  networks.  When  continuous  communication  between  the  reinforcement  channel  and  the  controller  is  challenged,  the  proposed  framework  will  develop  an  self  learning  algorithm  that  will  provide  conditions  wherein  the  internal  reinforcement  signal  is  rich  enough  to  allow  learning  of  the  optimal  control  policy  and  when  a  trade-off  between  internal  and  external  reinforcements  is  utilized.  Finally,  bionic,  human  brain-centric  learning  for  CPS  is  formalised  using  B.F.  Skinner's  operant  conditioning  re-contextualised  for  learning  in  CPS,  as  well  as  sparse  evolutionary  training  of  neuronal  configurations  informed  with  physics  of  the  system  to  approximate  solutions  to  the  non-linear  Hamilton-Jacobi-Bellman  (HJB)  is  presented.
■590    ▼aSchool  code:  0078.
■650  4▼aSchedules
■650  4▼aCommunication  channels
■650  4▼aDecision  making
■650  4▼aNeural  networks
■650  4▼aMilitary  deployment
■650  4▼aCommunications  networks
■650  4▼aDesign
■650  4▼aCognition  &  reasoning
■650  4▼aCognitive  psychology
■650  4▼aElectrical  engineering
■650  4▼aMilitary  studies
■690    ▼a0389
■690    ▼a0800
■690    ▼a0633
■690    ▼a0544
■690    ▼a0750
■71020▼aGeorgia  Institute  of  Technology.
■7730  ▼tDissertations  Abstracts  International▼g87-05B.
■790    ▼a0078
■791    ▼aPh.D.
■792    ▼a2024
■793    ▼aEnglish
■85640▼uhttp://www.riss.kr/pdu/ddodLink.do?id=T17360365▼nKERIS▼z이  자료의  원문은  한국교육학술정보원에서  제공합니다.

미리보기

내보내기

chatGPT토론

Ai 추천 관련 도서


    신착도서 더보기
    최근 3년간 통계입니다.

    소장정보

    • 예약
    • 소재불명신고
    • 나의폴더
    • 우선정리요청
    • 비도서대출신청
    • 야간 도서대출신청
    소장자료
    등록번호 청구기호 소장처 대출가능여부 대출정보
    TF14570 전자도서 대출가능 마이폴더 부재도서신고 비도서대출신청 야간 도서대출신청

    * 대출중인 자료에 한하여 예약이 가능합니다. 예약을 원하시면 예약버튼을 클릭하십시오.

    해당 도서를 다른 이용자가 함께 대출한 도서

    관련 인기도서

    로그인 후 이용 가능합니다.