SFMGNet: A Physics-based Neural Network To Predict Pedestrian Trajectories
This addresses the problem of safe and interpretable interaction for autonomous robots and vehicles in mixed-traffic environments, representing an incremental improvement over existing methods.
The paper tackles pedestrian trajectory prediction for autonomous systems by introducing SFMGNet, a physics-based neural network combining a social force model with an MLP, which achieves better than state-of-the-art accuracy and realistic, interpretable predictions even when trained only on synthetic data.
Autonomous robots and vehicles are expected to soon become an integral part of our environment. Unsatisfactory issues regarding interaction with existing road users, performance in mixed-traffic areas and lack of interpretable behavior remain key obstacles. To address these, we present a physics-based neural network, based on a hybrid approach combining a social force model extended by group force (SFMG) with Multi-Layer Perceptron (MLP) to predict pedestrian trajectories considering its interaction with static obstacles, other pedestrians and pedestrian groups. We quantitatively and qualitatively evaluate the model with respect to realistic prediction, prediction performance and prediction "interpretability". Initial results suggest, the model even when solely trained on a synthetic dataset, can predict realistic and interpretable trajectories with better than state-of-the-art accuracy.