Towards Combining Model Checking and Proof Checking

Abstract

Model checking and automated theorem proving are two pillars of formal methods. This paper investigates model checking from an automated theorem proving perspective, aiming at combining the expressiveness of automated theorem proving and the complete automaticity of model checking. The focus of this paper is on the verification of temporal logic properties of Kripke models. The main contributions of this paper are: first the definition of an extended computation tree logic that allows polyadic predicate symbols, then a proof system for this logic, taking Kripke models as parameters, then, the design of a proofsearch algorithm for this calculus and a new automated theorem prover to implement it. The verification process is completely automatic, and produces either a counterexample when the property does not hold, or a certificate when it does. The experimental result compares well to existing state-of-the-art tools on some benchmarks, including an application to air traffic control and the design choices that lead to this efficiency are discussed. arXiv:1606.08668v2 [cs.LO] 30 Sep 2017 5

Scratch-Based User-Friendly Requirements Definition for Formal Verification of Control Systems

“…However, one should remain aware of the limitations of formal methods (Tempelmeier, 2011), including environmental influences, unaccountable world knowledge or the misbehavior of neighbouring systems. One of the most approved methods is the model checking technique (Pakonen et al, 2016;Clarke et al, 1999l Emerson, 2008Jiang et al, 2018), which has recently been entering slowly even into quantum systems (Ying and Feng, 2018). Generally speaking, in model checking the specification is validated against some user-defined requirements.…”