Towards Generalizable and Interpretable Motion Prediction: A Deep Variational Bayes Approach

Juanwu Lu 1, Wei Zhan 2, Masayoshi Tomizuka 2, Yeping Hu 3
1 Purdue University 2 University of California, Berkeley 3 Lawrence Livermore National Laboratory
The 27th International Conference on Artificial Intelligence and Statistics
Paper Code arXiv

Abstract

framework
Overview of the GNeVA Framework

Estimating the potential behavior of the surrounding human-driven vehicles is crucial for the safety of autonomous vehicles in a mixed traffic flow. Recent state-of-the-art achieved accurate prediction using deep neural networks. However, these end-to-end models are usually black boxes with weak interpretability and generalizability. This paper proposes the Goal-based Neural Variational Agent (GNeVA), an interpretable generative model for motion prediction with robust generalizability to out-of-distribution cases. For interpretability, the model achieves target-driven motion prediction by estimating the spatial distribution of long-term destinations with a variational mixture of Gaussians. We identify a causal structure among maps and agents' histories and derive a variational posterior to enhance generalizability. Experiments on motion prediction datasets validate that the fitted model can be interpretable and generalizable and can achieve comparable performance to state-of-the-art results.

Visualizing Posterior Predictive Distribution of Goals

Poster

BibTeX


@InProceedings{pmlr-v238-lu24a,
  title = 	 { Towards Generalizable and Interpretable Motion Prediction: A Deep Variational Bayes Approach },
  author =       {Lu, Juanwu and Zhan, Wei and Tomizuka, Masayoshi and Hu, Yeping},
  booktitle = 	 {Proceedings of The 27th International Conference on Artificial Intelligence and Statistics},
  pages = 	 {4717--4725},
  year = 	 {2024},
  editor = 	 {Dasgupta, Sanjoy and Mandt, Stephan and Li, Yingzhen},
  volume = 	 {238},
  series = 	 {Proceedings of Machine Learning Research},
  month = 	 {02--04 May},
  publisher =    {PMLR},
  pdf = 	 {https://proceedings.mlr.press/v238/lu24a/lu24a.pdf},
  url = 	 {https://proceedings.mlr.press/v238/lu24a.html},
  abstract = 	 { Estimating the potential behavior of the surrounding human-driven vehicles is crucial for the safety of autonomous vehicles in a mixed traffic flow. Recent state-of-the-art achieved accurate prediction using deep neural networks. However, these end-to-end models are usually black boxes with weak interpretability and generalizability. This paper proposes the Goal-based Neural Variational Agent (GNeVA), an interpretable generative model for motion prediction with robust generalizability to out-of-distribution cases. For interpretability, the model achieves target-driven motion prediction by estimating the spatial distribution of long-term destinations with a variational mixture of Gaussians. We identify a causal structure among maps and agents’ histories and derive a variational posterior to enhance generalizability. Experiments on motion prediction datasets validate that the fitted model can be interpretable and generalizable and can achieve comparable performance to state-of-the-art results. }
}