Verhaltensentscheidung für automatisierte Fahrzeuge mittels Arbitrationsgraphen

Abstract

This work proposes a system architecture for secure and robust behavioral decision-making for automated vehicles. It assembles basic behavior blocks in a hierarchical arbitration graph and ensures safety through verification and diverse levels of fallback. The presented method contributes to a transparent and comprehensible decision-making process and produces safe and stable driving behavior even at high failure rates.