arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2202.09344cs.LOcs.MA
收藏

面向多智能体系统的模型检测与运行时验证结合研究

Towards the Combination of Model Checking and Runtime Verification on Multi-Agent Systems

Angelo Ferrando, Vadim Malvone

AI总结:

针对多智能体系统形式化验证难、模型检测问题通常不可判定的问题,提出结合模型检测与运行时验证的方法,用模型检测器验证可判定片段子模型、运行时监控器验证其余部分,并实现技术验证了效果。

摘要:

多智能体系统(Multi-Agent Systems, MAS)复杂度极高且难以验证。事实上,为MAS建模并非易事,即便构建出模型,也未必总能通过形式化方法验证其实际行为符合预期。通常而言,明确智能体是否有能力完成自身目标十分关键。检测这一点的一种可行方法是模型检测,具体可通过验证交替时间时序逻辑(Alternating-time Temporal Logic, ATL)属性来实现,该逻辑可描述实现目标的策略概念。但遗憾的是,由此产生的模型检测问题通常是不可判定的。本文提出一种结合模型检测与运行时验证的验证流程:属于可判定片段的MAS子模型由模型检测器验证,其余部分则通过运行时监控器验证。此外,我们实现了该技术并展示了实验结果。

英文摘要:

Multi-Agent Systems (MAS) are notoriously complex and hard to verify. In fact, it is not trivial to model a MAS, and even when a model is built, it is not always possible to verify, in a formal way, that it is actually behaving as we expect. Usually, it is relevant to know whether an agent is capable of fulfilling its own goals. One possible way to check this is through Model Checking. Specifically, by verifying Alternating-time Temporal Logic (ATL) properties, where the notion of strategies for achieving goals can be described. Unfortunately, the resulting model checking problem is not decidable in general. In this paper, we present a verification procedure based on combining Model Checking and Runtime Verification, where sub-models of the MAS model belonging to decidable fragments are verified by a model checker, and runtime monitors are used to verify the rest. Furthermore, we implement our technique and show experimental results.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑