FloMo: Tractable Motion Prediction with Normalizing Flows
This addresses motion prediction for autonomous driving, providing a tractable method with likelihoods, but it is incremental as it builds on existing generative models with specific improvements.
The paper tackled the problem of uncertain future motion prediction for autonomous agents by modeling it as a density estimation problem using normalizing flows, achieving state-of-the-art performance on three datasets with a significant gap over competitors.
The future motion of traffic participants is inherently uncertain. To plan safely, therefore, an autonomous agent must take into account multiple possible trajectory outcomes and prioritize them. Recently, this problem has been addressed with generative neural networks. However, most generative models either do not learn the true underlying trajectory distribution reliably, or do not allow predictions to be associated with likelihoods. In our work, we model motion prediction directly as a density estimation problem with a normalizing flow between a noise distribution and the future motion distribution. Our model, named FloMo, allows likelihoods to be computed in a single network pass and can be trained directly with maximum likelihood estimation. Furthermore, we propose a method to stabilize training flows on trajectory datasets and a new data augmentation transformation that improves the performance and generalization of our model. Our method achieves state-of-the-art performance on three popular prediction datasets, with a significant gap to most competing models.