Design and Implementation of an RFID-Based Intelligent Transportation System (ITS) using IoT and Data Analytics

Abstract

The creation of fine-grained vehicle traffic data for an entire city or a wider area is mostly dependent on an Intelligent Transportation System (ITS). A road network's traffic flow is optimised using real-time traffic data. However, developing nations like India have limited ITS infrastructure because to the costly expense of sending and support, making it trying to create continuous traffic data on a wide scale. Therefore, an affordable ITS solution is required. In India, a significant portion of the road network is covered by the extensive cellular network deployment. However, cellular network-based positioning data is not appropriate for estimating edge level transit time or speed due to its substantial location inaccuracy (250–500 meters). The growing number of smart phone users and GPS-enabled cars makes GPS probe data a desirable source for estimating travel speeds in real time. However, it is unsuitable for developing nations like India due to its limited penetration and the fact that only certain types of vehicles—cars and buses—have GPS. About three-quarters of all vehicles on Indian arterial roadways are two-wheelers. In this work, we investigate the issue of producing cost-effective real-time traffic information. Numerous proposals that are tested under different traffic situations and make use of one or more alternative sources of traffic information have been published in the literature. The majority of these solutions either employ information from different sources, either regardless of combination, or they utilize cell network information alone, which brings about a significant mistake in the produced rush hour gridlock data. In order to produce precise traffic statistics, this dissertation proposes a novel method of combining data from restricted ITS infrastructure, GPS probe data, and broadly accessible cellular network data.