Estimating pedestrian intentions from trajectory data

In autonomous driving systems, one of the most crucial aspects is to anticipate the movements of other traffic participants. In this paper, several machine learning methods are used to train classifiers capable of estimating the intention of a pedestrian to cross a zebra crossing. Their results are compared to a Bayesian network-an approach commonly used in autonomous driving. The data used for the estimation contain only position and heading of the pedestrians. The best performing method achieved the F ₂ score of 92.37%.