Combining Explicit and Symbolic Approaches for Better On-the-Fly LTL Model Checking

Abstract: We present two new hybrid techniques that replace the synchronized product used in the automata-theoretic approach for LTL model checking. The proposed products are explicit graphs of aggregates (symbolic sets of states) that can be interpreted as B\"uchi automata. These hybrid approaches allow on the one hand to use classical emptiness-check algorithms and build the graph on-the-fly, and on the other hand, to have a compact encoding of the state space thanks to the symbolic representation of the aggregates. The Symbolic Observation Product assumes a globally stuttering property (e.g., LTL \ X) to aggregate states. The Self-Loop Aggregation Product} does not require the property to be globally stuttering (i.e., it can tackle full LTL), but dynamically detects and exploits a form of stuttering where possible. Our experiments show that these two variants, while incomparable with each other, can outperform other existing approaches.