Three years of EU-wide cooperation, eight use cases in four cities, hundreds of virtual autonomous vehicles driving on various digital roads in traffic models. After the successful completion of CoEXist, the partners decided to summarise the main findings in a short and concise brochure, focusing on each individual use case, the modelling tools used, highlighting the main findings, providing an executive summary and providing recommendations for municipalities.
The brochure provides an overview of the analysis from the four partner cities of Stuttgart (DE), Helmond (NL), Milton Keynes (UK) and Gothenburg (SE), who first assessed their mobility needs in terms of automated vehicles, defined use cases, created models and drew conclusions from them. These models were divided in micro- and macro-simulations, looking at various environments, which are possibly challenging for AVs, including shared spaces, highways, roundabouts, signalised intersections and others. During most of these assessments, a closer look was taken at the behaviour of AVs with different levels of automation, ranging from driver assistance mechanisms (low levels) up to fully autonomous cars (high levels).
Overall, major conclusions could be drawn from all use cases, as they suggest that traffic flows will be significantly hampered during the introduction phase of low-level AVs. As safety mechanisms are programmed to behave with extreme caution in difficult environments, low-level AVs often stop at the slightest obstacle, as seen in the shared space use case in Gothenburg. Once AVs are more advanced and widespread, traffic flows ease, and the level of efficiency rises. Some assessments suggest that traffic flows are more efficient, once a certain saturation of AVs is reached.
If you want to learn more, we invite you to read our brochure.