We extend Pratt's worst-case optimal decision procedure for PDL to a richer logic with nominals, difference modalities, and inverse actions. We prove correctness and worst-case optimality. Our correctness proof is based on syntactic models called demos. The main theorem states that a formula is satisfiable if and only if it is contained in a demo. From this theorem the correctness of the decision procedure is easily obtained. Our development is modular and we extend it stepwise from modal logic with eventualities to the full logic.
DOI (LNCS version): 10.1007/978-3-642-22119-4_16
Download PDF Show BibTeX
Login to edit