Autonomous vehicles (AV) rely on environmental perception to take manoeuvre decisions. Safety assurance for AV thus hinges on achieving confidence in all percepts that are safe-guarding critical manoeuvres. As the safety targets for such critical manoeuvres are considerably higher than the statistical figures for the reliability of at least current learning-enabled classification algorithms within the environmental perception, we need means for assuring that the overall system is “safer than perception” in that the frequency of erratically adopting a critical manoeuvre is considerably lower than the frequency of individual misclassifications. We present a methodology for constructively generating reformulations of guard conditions that are more resilient to misperception than the original condition. The synthesized, rephrased guard conditions reconcile a given safety target, i.e. a given a societally accepted upper bound on erratically activating a critical manoeuvre due to a false positive in guard evaluation, with maximal availability, i.e. maximal true positive rate. By synthesizing a resilient rephrasing of the guard condition, the constructive setting presented herein complements the analytical setting from a previous companion paper [6], which merely analysed a given condition for its safety under uncertain perception.