Intelligent Traffic Based on Hybrid Control Policy of Connected Autonomous Vehicles in Multiple Unsignalized Intersections

Over the last decades, several advanced intersection control systems are built to alleviate traffic congestion. Connected Autonomous Vehicles (CAVs) can be more easily developed for cooperative navigation than regular traffic. Due to whole uncertainty in a transportation network, the conventional motion planning for local areas may lead to undesirable consequences in long term. In this context, this paper presents the Micro-Macro Flow Control (MiMaFC) strategy to explore CAVs’ global navigation performance in a traffic network. In prior work, a separate management layer named local supervisor is constructed to control adjacent unsignalized intersections by considering traffic aggregated velocity and vehicle crossing priority. This paper extends the problem to multiple interacting road intersections. Correspondingly, a hybrid control policy is implemented in local areas to solve CAVs conflicts while improving the traffic flow. Further, enhanced intersection navigation protocols are exploited to deal with continued traffic streams. Simulations including a congested traffic network are presented to evaluate the proposed MiMaFC strategy. It is shown in the paper that the mobility in the urban network can be improved by the proposed motion planning framework compared to the non-supervised CAVs system in the same reference conditions.