MCMAS-SLK: A Model Checker for the Verification of Strategy Logic Specifications

Abstract

Model checking has come of age. A number of techniques are increasingly used in industrial setting to verify hardware and software systems, both against models and concrete implementations. While it is generally accepted that obstacles still remain, notably handling infinite state systems efficiently, much of this work involves refining and improving existing techniques such as predicate abstraction. At scientific level a major avenue of work remains the development of verification techniques against rich and expressive specification languages. Over the years there has been a natural progression from checking reachability only, to a large number of techniques (BDDs, BMC, abstraction, etc) catering for LTL [24], CTL [10], and CTL [11]. More recently, ATL and ATL [3] were introduced to analyse systems in which some components, or agents, can enforce temporal properties on the system. The paths so identified correspond to infinite games between a coalition and its complement. ATL is well explored theoretically and at least two toolkits now support it [4, 16, 17]. It has however been observed that ATL suffers from a number of limitations when one tries to apply it to multi-agent system reasoning and games [1, 2, 5, 19, 27]. One of these is the lack of support for binding strategies explicitly to various agents or to the same agent in different contexts [22, 23]. To overcome this and other difficulties, Strategy Logic (SL) has been put forward. By using SL a number of multi-agent based specifications involving cooperation become naturally expressible. Also, key game-theoretic properties such as Nash equilibria, which were previously not logically-representable, can easily be captured in SL. In this paper, we describe MCMAS-SLK, a first model checker for SL. This tool supports the verification of SL specifications (hence ATL and CTL), the synthesis of agents’ strategies to satisfy a given parametric specification, as well as basic counterexample generation. MCMAS-SLK, released as open-source, implements novel labelling algorithms for SL, encoded on BDDs, and reuses existing algorithms for the verification of epistemic specifications [25].

DOI: 10.1007/978-3-319-08867-9_34

Extracted Key Phrases

1 Figure or Table

01020201520162017
Citations per Year

Citation Velocity: 13

Averaging 13 citations per year over the last 3 years.

Learn more about how we calculate this metric in our FAQ.

Cite this paper

@inproceedings{Cermk2014MCMASSLKAM, title={MCMAS-SLK: A Model Checker for the Verification of Strategy Logic Specifications}, author={Petr Cerm{\'a}k and Alessio Lomuscio and Fabio Mogavero and Aniello Murano}, booktitle={CAV}, year={2014} }