An Improved Velocity Forecasts Method Considering the Speed and Following Distance of Two Vehicles

Accurate vehicle speed prediction is essential to the optimal control of electric vehicles. Compared with the traditional vehicle speed prediction, with the rapid development of the Internet of Vehicles technology, it’s more practical to make predictions using the speed and traffic information of the vehicles ahead. In this paper, we propose an improved velocity prediction algorithm for intelligent connected vehicles based on the historical speed of the target vehicle, the historical speed of the preceding vehicle, and the following distance. First, build a single-lane traffic model in VISSIM and simulate it to get the speed and distance of the two vehicles. Then, an improved BP neural network prediction algorithm is proposed to train the neural network based on the obtained multivariate traffic information. Finally, the proposed velocity prediction algorithm is used to predict the speed of the target vehicle. In order to verify the method proposed in this paper, the prediction effect of the improved BP neural network prediction algorithm is compared with that of gray prediction and traditional BP neural network. The results show that the accuracy of velocity prediction is significantly improved in a simple traffic environment.