Download PDFOpen PDF in browser

Verification of Stochastic Systems by Stochastic Satisfiability Modulo Theories with Continuous Domain

9 pagesPublished: December 18, 2015

Abstract

Stochastic Satisfiability Modulo Theories (SSMT) is a quantitative exten-
sion of classical Satisfiability Modulo Theories (SMT) inspired by stochastic
logics. It extends SMT by the usual as well as randomized quantifiers, fa-
cilitating capture of stochastic game properties in the logic, like reachability
analysis of hybrid-state Markov decision processes. Solving for SSMT for-
mulae with quantification over finite and thus discrete domain has been ad-
dressed by Tino Teige et al. In our work, we extend their work to SSMT
over continuous quantifier domains (CSSMT) in order to enable capture of,
e.g., continuous disturbances and uncertainty in hybrid systems. We extend
the semantics of SSMT and introduce a corresponding solving procedure. A
discussion regarding to reachability analysis is given to demonstrate applica-
bility of our framework to reachability problems in hybrid systems.

Keyphrases: constraint solving, hybrid system, reachability analysis, Stochastic Satisfiability Modulo Theory

In: Sergiy Bogomolov and Ashish Tiwari (editors). Symbolic and Numerical Methods for Reachability Analysis, 1st International Workshop, SNR 2015, vol 37, pages 2--10

Links:
BibTeX entry
@inproceedings{SNR2015:Verification_of_Stochastic_Systems,
  author    = {Yang Gao and Martin Fr\textbackslash{}"anzle},
  title     = {Verification of Stochastic Systems by Stochastic Satisfiability Modulo Theories with Continuous Domain},
  booktitle = {Symbolic and Numerical Methods for Reachability Analysis, 1st International Workshop, SNR 2015},
  editor    = {Sergiy Bogomolov and Ashish Tiwari},
  series    = {EPiC Series in Computing},
  volume    = {37},
  pages     = {2--10},
  year      = {2015},
  publisher = {EasyChair},
  bibsource = {EasyChair, https://easychair.org},
  issn      = {2398-7340},
  url       = {https://easychair.org/publications/paper/TWf},
  doi       = {10.29007/wm3j}}
Download PDFOpen PDF in browser