An Agent-Based Model to Predict Pedestrians Trajectories with an Autonomous Vehicle in Shared Spaces
Un modèle à base d'agents pour prédire les trajectoires des piétons dans des espaces partagés avec un véhicule autonome
Résumé
This paper addresses modeling and simulating pedestrian trajectories when interacting with an autonomous vehicle in a shared space. Pedestrian motion models integrating pedestrians interactions with an autonomous vehicle already exist. However, they fail to accurately predict the individual trajectory of each pedestrian, and they do not deal with the diversity of possible pedestrian interactions with the vehicle in a shared space (front, back or lateral). Moreover, previous works do not sufficiently provide a quantitative evaluation of the model's predictions. In this paper, we propose an hybrid pedestrian model that combines the social force model and a new decision model for conflicting pedestrian-vehicle interactions. The proposed model integrates different observed pedestrians behaviors, as well as the behaviors of the social groups of pedestrians. We validate the model and evaluate its predictive potential through qualitative and quantitative comparisons with ground truth trajectories. The proposed model reproduces observed behaviors that have not been replicated by the social force model and outperforms the social force model at predicting pedestrians trajectories on the used dataset. This model will be used by an autonomous vehicle in a shared space to predict the trajectories of surrounding pedestrians.
Origine : Fichiers produits par l'(les) auteur(s)