Network user equilibrium problems with infrastructure-enabled autonomy

Although autonomous vehicle (AV) technology is expected to bring dramatic societal, environmental, and economic benefits, the high cost might slow adoption of AVs. This paper explores an infrastructure-enabled driving system, which a promising remedy Specifically, system combines vehicles infrastructure in realization driving. Equipped with roadside sensing, computing, communicating devices, ordinary road can be upgraded into “automated road” enabling service for minimum required on-board devices. The costs thus significantly reduced. We envision that automated roads will deployed transportation networks serve new type called (IEAVs), driven autonomously only on but manually roads. Therefore, IEAV users may experience inconvenience due transitions between manual driving, frequent switch inevitably yield costs. minimize their individual travel cost, they have decide whether mode when heading road. Considering such unique feature IEAVs, we proposed group driving-mode-choice equilibrium conditions describe drivers’ choice behaviors, considered time costs, charges change. Combining traditional route-choice conditions, developed user (UE) model flow distributions network mixed-autonomy traffic. UE formulated as novel non-linear complementarity problem, its solution existence discussed. To solve model, expansion method reformulate standard path-based model. A route-swapping-based algorithm then used reformulated Numerical studies are presented demonstrate models algorithms.