Optimal distributed signal control for platooning in connected vehicle networks.

Pressure-based algorithms such as Back Pressure and Max Weight have been widely adopted in traffic signal control because of their simplicity and proven throughput maximization. This paper proposes a new optimal Max Weight control mechanism to enhance the network throughput and reduce vehicle delays using a platooning concept enabled by Connected Vehicles (CVs). In particular, the weight (or pressure) of a traffic movement at an intersection in our proposed algorithm consists of several features: platoon delay, speed and position. To the best of our knowledge, there is no existing scheme that considers multiple configurable attributes when formulating the pressure. Furthermore, we also develop a machine learning procedure in order to optimize the weight of each attribute contributing to the total pressure. A number of simulations demonstrate the feasibility of the learning procedure and show that our algorithm can perform better than the state-of-the-art and well-known pressure-based algorithms.