Synthesizing nonanomalous event-based controllers for liveness goals

We present SGR(1), a novel synthesis technique and methodological guidelines for automatically constructing event-based behavior models. Our approach works for an expressive subset of liveness properties, distinguishes between controlled and monitored actions, and differentiates system goals from environment assumptions. We show that assumptions must be modeled carefully in order to avoid synthesizing anomalous behavior models. We characterize nonanomalous models and propose assumption compatibility, a sufficient condition, as a methodological guideline.