Time-Dependent Lane-Level Navigation With Spatiotemporal Mobility Modeling Based on the Internet of Vehicles

Abstract

In this article, we propose a time-dependent lane-level navigation (TDLN) framework with spatiotemporal mobility modeling based on the Internet of Vehicles (IoV). The proposed TDLN framework can provide drivers with the fastest navigation path that can avoid passing congestion areas and predict vehicle spatiotemporal mobility of future traffic flows by estimating the travel time of road segments and the waiting time of intersections. According to our review of relevant research, TDLN is the first lane-level navigation solution that can provide the following features: 1) it can navigate vehicles in a lane-level manner and classify the queuing state of each vehicle as passing through an intersection; 2) it can estimate the driving time of lanes and the stopping time of intersections in different lanes to calculate the total delay time of passing through each lane and intersection; and 3) it can predict future traffic flows to determine the congestion level of each lane and explore predicted flow conditions on the road network to achieve the fastest navigation path planning. Simulation results show that TDLN outperforms existing methods and can plan the lane-level navigation path with the shortest travel time.