Reducing congestion and emissions via roadside unit deployment under mixed traffic flow

Abstract

Abstract It is expected that for a long time the future road traffic will be composed of both regular vehicles (RVs) and connected autonomous vehicles (CAVs). As a vehicle-to-infrastructure technology dedicated to facilitating CAV under the mixed traffic flow, roadside units (RSUs) can also improve the quality of information received by CAVs, thereby influencing the routing behavior of CAV users. This paper explores the possibility of leveraging the RSU deployment to affect the route choices of both CAVs and RVs and the adoption rate of CAVs so as to reduce the network congestion and emissions. To this end, we first establish a logit-based stochastic user equilibrium model to capture drivers’ route choice and vehicle type choice behaviors provided the RSU deployment plan is given. Particularly, CAV users’ perception error can be reduced by higher CAV penetration and denser RSUs deployed on the road due to the improved information quality. With the established equilibrium model, the RSU deployment problem is then formulated as a mathematical program with equilibrium constraints. An active-set algorithm is presented to solve the deployment problem efficiently. Numerical results suggest that an optimal RSU deployment plan can effectively drive the system towards one with lower network delay and emissions.

Keywords

• Roadside unit deployment
• Connected autonomous vehicle
• Information quality
• Congestion and emission