Collision Avoidance Models and Algorithms in the Era of Internet of Vehicles

Abstract

Rear-end crashes and intersection collisions are two common types of accidents in roadways, and a lot of technology and approaches have been used to analyze the vehicle distance and provide effective precautions for drivers. Vehicular Ad-Hoc Network (VANET) can be adopted to share traffic information, select optimal driving route and realize safe driving without any collisions. In this paper, we reviewed a series of models and algorithms in establishing avoidance system for rear-end and intersection collisions. As for rear-end crashes, VK model and neural network had been exploited to release warning messages and realize emergency braking to avoid collisions. Concerning intersection collisions, a spectrum of effective algorithms were put forward to realize collision avoidance in intersection such as Markvo chain. Yet, a handful of challenges should not be neglected in building up efficient collision avoidance system in VANET such as MAC design, information security and effective communications among vehicles. To establish more efficient and accurate collision avoidance system in the environment of VANET, cloud computing, artificial intelligence, 5G technology and automatic technology can be developed and applied in this field in the future.