Coordination of Autonomous Vehicles and Dynamic Traffic Rules in Mixed Automated/Manual Traffic

We consider the coordination of multiple connected automated vehicles (CAVs) by a central coordinator (CC) in mixed traffic where also non-controlled vehicles (NCVs) are present. The CC directly provides motion target commands to the CAVs and affects the behavior of NCVs by controlling dynamic traffic rules, such as changing traffic lights. We model the traffic rules for these scenarios by mixed-logical constraints, and the response of the NCVs to the changing traffic rules by a switched system with switches triggered by logic conditions. The proposed approach is implemented as a mixed-integer programming (MIP) problem to be solved in the central coordinator, which also handles timing constraints.