A Novel Traffic Characteristics Aware and Context Prediction Protocol for Intelligent Connected Vehicles

Abstract

Digital maps have been installed and attached to vehicles recently. They help with the GPS receivers to determine the relative locations of vehicles to other existing traffic and objects over the road network such as entrance/exit points, obstacles, road intersections, etc. This helps drivers or autonomous vehicles to decide the most appropriate reaction, in terms of speed, take-over, or stop operations ahead of time. Several daily traveling vehicles do not have digital maps. Besides, digital maps are vulnerable to being destroyed or inaccurate. They require regular updates due to the continuous construction and re-design of the road networks. These constructions aimed to enhance the road design and the traffic efficiency there. Moreover, accidents, broken vehicles, traffic congestion, or other ad-hoc obstacles appear unpredictably over the road network. In this paper, we aim to introduce a prediction protocol that gathers and analyzes the traffic characteristics of vehicles over the investigated road scenario using wireless transceivers in vehicles. Then, it predicts the physical and traffic context based on the analyzed traffic data. This protocol can replace the absent or broken digital maps in vehicles. It also can be used to verify the correctness of the digital map in vehicles. From the experimental results, we can infer that the proposed protocol has succeeded in predicting the road context over highways and downtown scenarios. More accurate and better predictions are acquired by increasing the percentage of wireless transceiver-equipped vehicles.