An Edge Computing Based Public Vehicle System for Smart Transportation
Jie Lin,Wei Yu,Xinyu Yang,Peng Zhao,Hanlin Zhang,Wei Zhao
DOI: https://doi.org/10.1109/tvt.2020.3028497
IF: 6.8
2020-11-01
IEEE Transactions on Vehicular Technology
Abstract:As a key smart transportation service, public vehicle systems are intended to improve traffic efficiency and vehicle occupancy ratios, and to reduce the number of vehicles on roads, by inducing travelers to share rides with others. Despite the clear logic behind this service, achieving a viable model for matching multiple riders to vehicles with low latency and high satisfaction remains an open issue. In this paper, we propose an Edge Computing based Public Vehicle (ECPV) system to improve traffic efficiency and vehicle occupancy ratios by scheduling ridesharing among travelers and reduce the delay of decision making by leveraging edge computing. Particularly, by introducing a metric of traveler satisfaction jointly considering travel time, distance, and costs (i.e., charge), we formalize public vehicle scheduling problem as an optimization problem with maximizing traveler satisfaction as objective to reduce travel time and improve traffic efficiency. Further, as achieving globally optimal matching of rides and travelers is time consuming, to reduce the delay of decision making (i.e., response to ride-traveler pairs matching) and improve real time ridesharing service and traveler satisfaction, an edge computing based ride request transmission mechanism and a tree based heuristic matching mechanism are proposed to effectively exchange ride requests and select appropriate vehicles for mostof ride requests on edge devices. Also, our ECPV system considers driverless cars as public vehicles to remove the subjective influence of drivers and improve the efficiency of vehicle scheduling, and introduces a graph partition based depot placement mechanism to determine vehicleparking locations and balance vehicle distribution across system. Through extensive performance evaluations, our experimental results show that our proposed ECPV system can effectively match ride requests to public vehicles, reduce travel times and distances for all trips, reducecharges to travelers, and i-prove vehicle occupancy.
telecommunications,engineering, electrical & electronic,transportation science & technology