Minimizing Traffic Congestion through V2V Communication

Abstract

With the rising level of vehicular traffic congestion, it is important to find a way to manage these congestions at an early stage. Connected vehicle technology is a potential solution that can significantly improve traffic system efficiency. Cars use short range radio signals to communicate with each other, so that they are aware of other vehicles’ position in roads. Drivers can, subsequently, receive notifications and alert others of dangerous situations in the road ahead. A heuristic approach was developed in this research to design a dissemination strategy based on a dynamic set covering model that ensures full coverage with the minimum number of relay vehicles. This model selects the optimal relay vehicles at each dissemination step, which in turn will communicate the information to other vehicles. The selection of the optimal relay vehicles was solved by formulating a linear programming model that minimizes the number of the selected relay vehicles while maintain full coverage for all vehicles. The model was tested and coded using the Matlab software simulating a highway and an urban environment configuration. Model validation and evaluation were carried for both configurations. A comparison was performed with other proposed models included in the literature review. The proposed model showed a significant decrease in the dissemination time of a warning message. The longest time recorded was 1.4 seconds for a road length of 1 km with vehicle density of 0.125 vehicle per lane per meter in a highway scenario with a number of transmitted messages of 353741. In the urban scenario of two 1 km roads and vehicle density of 0.125 the model recorded a time of 0.8 seconds with 481562 messages disseminated. A better performance in terms of delay time was achieved for the proposed model compared to any of the other models studied. This short delay time gives a better opportunity for the drivers to take an alternative route in case of congestions. The increase in road lengths does not need to increase the number of relay vehicles significantly, which shows the efficiency of the proposed model.