A Centralized V2I LoRaWAN Architecture for Real-Time Adaptive Route Planning and Traceability of Electric Vehicles

Smart mobility applications in urban areas involve the use of V2I infrastructure, based on different communication technologies. Low Power Wide Area Networks (LPWAN) are known by their higher power efficiency reducing considerably the implementation and operation costs. In particular, LoRaWAN is an effective LPWAN technology which supports a great number of connected devices over large areas with low bit rates. This paper presents an architecture, aimed at smart mobility applications, which allows implementing complex algorithms, typically based on artificial intelligence techniques, using real-time V2I data. Both roadside units (RSUs) and on board units (OBUs) are connected to the decision centres by means of LoRaWAN-enabled devices. The architecture was validated experimentally in a use case scenario involving the so-called Dynamic Electric Vehicle Routing Problem (DEVRP). Namely, it was used for planning dynamically the route of electric vehicle. The algorithms executed in the central layer compute optimal routes, based on continuously updated environmental and operational variables. These algorithms intend to enhance fleet efficiency, reduce energy consumption, and extend the battery life of the vehicles. In this case study, live data, from vehicles OBUs and RSUs, was continuously forwarded to the centralized computation layer where the algorithms reacted to diverse traffic events such as slow traffic points, traffic jams or closed roads.