Neural-network-based modeling and prediction of the future state of a Stop&Go behavior in urban areas

The goal of Stop&Go systems is to assist drivers in traffic jams by reducing the need for them to repeatedly accelerate and/or stop their vehicles. There have been attempts to model Stop&Go waves via microscopic and macroscopic traffic models. But predicting the future state of the behavior of a Driver-Vehicle-Unit (DVU) in this maneuver has not been studied much. The purpose of this study is to design neural-network-based models to simulate and predict the future state of the Stop&Go maneuver in real traffic flow for different steps ahead. These models are designed based on the real traffic data and model the acceleration of the vehicle which performs a Stop&Go maneuver. The models were validated at the microscopic level, and the results showed very close agreement between field data and models output. The proposed models can be employed in ITS applications, Drier Assistant devices, Collision Prevention systems and etc.