A method for connected vehicle trajectory prediction and collision warning algorithm based on V2V communication

Connected vehicle communication technology is rapidly developing in recent years, and host vehicle (HV) can send or receive the basic safety message (BSM) from the remote vehicles (RVs). However, there are few applications using this information to improve the driving safety. In this paper, we propose a collision warning predicted framework that provides connected automated vehicle and alert driver when time to collision (TTC) is within specified thresholds. After preprocessor RV BSM data, this paper transforms the RV position and calculates the relative position, distance and speed. Then, the RV trajectory is estimated by using Kalman filter algorithm, and the error statistics of the prediction of latitude and longitude are analysed. The prediction results show that it works on straight, corner and the curve road, but the latitude error is higher than the longitude error. At last this paper constructs the relative position radar map for the HV, in which it can show the relative position, speed and TTC information. Vehicle collision can be detected in real time and the vehicle can prevent the potential conflict accordingly by using these information.