traffic congestion detection and avoidance using vehicular communication

Traffic congestion is a serious problem in big cities. With the number of vehicles increasing rapidly, especially in cities whose economy is booming, the situation is getting even worse. Drivers, unaware of congestion ahead eventually join it and increase the severity of it. The ability of a driver to know the traffic conditions on the roads ahead enables him/her to seek alternate routes through which time and fuel can be saved. Due to recent advancements in vehicular technologies, vehicular communication has emerged. The objective of this work is to check feasibility of using infrastructure based vehicular communication for detecting and avoiding traffic congestion. In this paper we propose a Signal Agent (SA) and Car Agent(CA)based approach for detecting and avoiding traffic congestion. We analyze performance of the proposed approach for two different road network scenarios using simulations: structured grid network (like Gandhinagar City of Gujarat, India) and apart of typical city road network ( Tiwan city). With the proposed approach we get reduction of 10.05% in trip duration of vehicles, reduction of 10.08% in number of vehicles in entire traffic road network and 9.82% in heavy traffic area. In an accident scenario, about 72.63% vehicles changed their route due to awareness of congestion. Error in trip time estimation and vehicle count estimation is observed to be less than 1%.