Investment and financing of roadway digital infrastructure for automated driving
研究了在不同交通流量下，如何最优投资车载通信与路侧传感器，并分析了最优收费能否覆盖数字基础设施的建设与装备成本。
Connected automated vehicles (CAVs) are equipped with sensors, enabling them to scan and analyze their surrounding environment. This capability empowers CAVs to make informed and efficient decisions regarding their motion; however, the limited spatial range and resolution of these sensors present challenges for achieving full autonomy. Cooperative sensing through vehicle-to-vehicle (V2V) and vehicle-to-infrastructure (V2I) communications offers an alternative approach to enrich CAVs’ environmental understanding. This study explores the optimal investment policy for vehicular connectivity and road-side sensor deployment under varying traffic flow conditions. It also extends the self-financing theorem to the sensor equipped roads and investigates whether an optimal toll can cover both the construction costs and the costs of equipping roads with sensing components. The stylized model of CAV mobility considers the interplay between stationary sensors installed road-side as a part of the infrastructure and mobile sensors of CAVs. Results indicate that under constrained budgets and low traffic flow, investing in infrastructure improvement is preferred. However, as traffic flow increases, prioritizing connectivity and data sharing among CAVs becomes more lucrative. Notably, in high traffic flow, a shift back to investing in stationary sensors may occur, depending on system settings. The findings provide insights into budget allocation to enhance CAV performance, advancing the development of efficient and safe automated driving systems. The analyses on the self-financing theorem also show that the optimal user tolls do not cover the cost of constructing digital infrastructure. However, if social planners consider the safety benefits of sensor equipped roads, the construction of digital infrastructure can be covered by the optimal user tolls. In addition, the revenue from optimal user tolls can cover the cost of equipping existing roads with sensors if their flow-capacity ratio is greater than a certain threshold. • We investigate optimal investment strategies in vehicular connectivity and road-side sensors. • For a given budget and low traffic flow, investing in roadside sensors is preferred. • For high traffic flows, investment should be allocated more to vehicular connectivity. • In high traffic flows, a shift back to investing in stationary sensors may occur depending on certain conditions.