In the transport system of the future, and before the full integration of autonomous vehicles (AVs), mixed traffic conditions involving both AVs and human-driven vehicles will coexist. In these transitional phases, traffic management strategies should be able to adapt to all users’ requirements and their effectiveness will be essential to maintain traffic flow and safety. Such TMS may refer to vehicle platooning—where AVs and human-driven vehicles travel in close, coordinated formations— or dynamic lane reversal strategies – where a specific part of the road changes its use based on the dynamic demand. This need is particularly crucial during non-recurrent events on the network, such as roadworks, accidents, and temporary lane closures. In such scenarios, adaptive traffic control mechanisms will help manage diverse vehicle behaviors, ensuring that both AVs and human-driven vehicles can navigate safely and seamlessly through unpredictable road conditions. In FRODDO AI-assisted traffic management strategies will be developed and tested in simulation environments to estimate large-scale impacts on the network.

More specifically, in the Athens pilot several scenarios will be tested using the up to date microscopic, simulated network of the Athens’ city center owned and maintained by the NTUA Laboratory of Traffic Engineering. Within the simulation traffic management strategies will be evaluated under various mixed traffic compositions, enabling the estimation of large-scale impacts on traffic, emissions and network efficiency. By modeling scenarios with diverse mixes of autonomous vehicles and traditional vehicles, simulation allows for safe, cost-effective experimentation across a wide range of traffic densities and behavioral patterns.