Formal Modeling of Road Network-Based Autonomous Vehicle Validation Scenarios With Intersections and Pedestrians

Autonomous vehicles (AVs) are an exciting emerging technology that reduces the active human participation in driving and extends awareness beyond human limitations of perception and reaction, increasing driving quality and safety. Validation testing of AVs provides the assurance of quality and safety, and is also an integral part of new AV research and development. Following in suite of their predecessor, i.e. non-autonomous vehicles, scenario-based validation testing is a safe, time and cost effective means of testing these complex, costly, and safety critical systems. In this paper we propose an open-source implementation of a scenario-based AV validation testing approach for AVs in traffic environments with pedestrians and intersections. Furthermore, we utilize the approach to implement scenarios of previously unimplementable AV scenarios from the Atomic Blocks dataset.