Probabilistic Model Checking and Autonomy

The design and control of autonomous systems that operate in uncertain or adversarial environments can be facilitated by formal modeling analysis. Probabilistic model checking is a technique to automatically verify, for given temporal logic specification, system satisfies the as well synthesize an optimal strategy its control. This method has recently been extended multiagent exhibit competitive cooperative behavior modeled via stochastic games synthesis equilibria strategies. In this article, we provide overview probabilistic checking, focusing on models supported PRISM PRISM-games checkers. includes fully observable partially Markov decision processes, turn-based concurrent games, together with associated logics. We demonstrate applicability framework through illustrative examples from systems. Finally, highlight research challenges suggest directions future work area.