Data-driven ballistic coefficient learning for future state prediction of high-speed vehicles

This paper describes a methodology to predict a future state of unknown high-speed vehicles by applying machine learning techniques. Traditionally, the state estimation of high-speed vehicles is carried out by the variations of Kalman filters, but such state estimation is limited to the temporal moment of the observation. Therefore, the future state of high-speed vehicles has been obtained through a number of predictive iterations with a dynamics equation. This dynamic equation requires a key parameter, i.e. ballistic coefficient, and this coefficient were merely fixed or modeled as another dynamics model in the past. The novelty of this paper lies on the utilization of machine learning models, i.e. Gaussian process regression and support vector regression, to predict the future ballistic coefficient. Our simulation experiments show that there is a reduction in the position error and the ballistic coefficient error when the machine learning models were used.