Temporal logic control of general Markov decision processes by approximate policy refinement
For researchers in formal verification and control of stochastic systems, this provides a method to handle uncountable state spaces via abstraction, but the approach is incremental.
This work addresses correct-by-design control of general Markov decision processes (gMDPs) with temporal logic properties using approximate probabilistic relations. It introduces robust satisfaction to ensure that properties satisfied on an abstract model are also satisfied on the original model with a refined control strategy.
The formal verification and controller synthesis for Markov decision processes that evolve over uncountable state spaces are computationally hard and thus generally rely on the use of approximations. In this work, we consider the correct-by-design control of general Markov decision processes (gMDPs) with respect to temporal logic properties by leveraging approximate probabilistic relations between the original model and its abstraction. We newly work with a robust satisfaction for the construction and verification of control strategies, which allows for both deviations in the outputs of the gMDPs and in the probabilistic transitions. The computation is done over the reduced or abstracted models, such that when a property is robustly satisfied on the abstract model, it is also satisfied on the original model with respect to a refined control strategy.