• DocumentCode
    434605
  • Title

    Assume-guarantee reasoning for hybrid I/O-automata by over-approximation of continuous interaction

  • Author

    Frehse, Goran ; Han, Zhi ; Krogh, Bruce

  • Author_Institution
    Dept. of Electr. & Comput. Eng., Carnegie Mellon Univ., Pittsburgh, PA, USA
  • Volume
    1
  • fYear
    2004
  • fDate
    17-17 Dec. 2004
  • Firstpage
    479
  • Abstract
    Assume-guarantee reasoning (AGR) is recognized as a means to counter the state explosion problem in the verification of safety properties. We propose a novel assume-guarantee rule for hybrid systems based on simulation relations. This makes it possible to perform compositional reasoning that is conservative in the sense of over-approximating the composed behaviors. The framework is formally based on hybrid input/output automata and their labeled transition system semantics. In contrast to previous approaches that require global receptivity conditions, the circularity is broken in our approach by a state-based nonblocking condition that can be checked in the course of computing the AGR simulation relations. The proposed procedures for AGR are implemented in a computational tool, called PHAVer, for the class of linear hybrid I/O automata, and the approach is illustrated with a simple example.
  • Keywords
    automata theory; continuous time systems; discrete time systems; formal verification; inference mechanisms; AGR simulation relations; PHAVer computational tool; assume-guarantee reasoning; compositional reasoning; continuous interaction; global receptivity conditions; hybrid I/O-automata; hybrid systems; over-approximation; state explosion problem; state-based nonblocking condition; transition system semantics; Automata; Automatic control; Computational complexity; Computational modeling; Control system synthesis; Counting circuits; Digital control; Electrical safety; Explosions; Sufficient conditions;
  • fLanguage
    English
  • Publisher
    ieee
  • Conference_Titel
    Decision and Control, 2004. CDC. 43rd IEEE Conference on
  • Conference_Location
    Nassau
  • ISSN
    0191-2216
  • Print_ISBN
    0-7803-8682-5
  • Type

    conf

  • DOI
    10.1109/CDC.2004.1428676
  • Filename
    1428676