Conditional Unscented Autoencoders for Trajectory Prediction

Faris Janjoš,Marcel Hallgarten,Anthony Knittel,Maxim Dolgov,Andreas Zell,J. Marius Zöllner

2024-02-27

Abstract:The CVAE is one of the most widely-used models in trajectory prediction for AD. It captures the interplay between a driving context and its ground-truth future into a probabilistic latent space and uses it to produce predictions. In this paper, we challenge key components of the CVAE. We leverage recent advances in the space of the VAE, the foundation of the CVAE, which show that a simple change in the sampling procedure can greatly benefit performance. We find that unscented sampling, which draws samples from any learned distribution in a deterministic manner, can naturally be better suited to trajectory prediction than potentially dangerous random sampling. We go further and offer additional improvements including a more structured Gaussian mixture latent space, as well as a novel, potentially more expressive way to do inference with CVAEs. We show wide applicability of our models by evaluating them on the INTERACTION prediction dataset, outperforming the state of the art, as well as at the task of image modeling on the CelebA dataset, outperforming the baseline vanilla CVAE. Code is available at <a class="link-external link-https" href="https://github.com/boschresearch/cuae-prediction" rel="external noopener nofollow">this https URL</a>.

Robotics,Artificial Intelligence,Computer Vision and Pattern Recognition,Machine Learning

What problem does this paper attempt to address?

### Problems Addressed by the Paper

The paper primarily focuses on the issue of human-driven vehicle trajectory prediction in autonomous driving systems and attempts to improve the performance of Conditional Variational Autoencoders (CVAE) in this task. Specifically, the paper challenges the key components of CVAE and proposes a series of improvements:

1. **Alternative Methods to Random Sampling**:
- A deterministic sampling method (such as Unscented Transform) is proposed to replace traditional random sampling, thereby enhancing the reliability and safety of trajectory prediction.

2. **Improvements in Latent Space Structure**:
- A Gaussian Mixture Model (GMM) is introduced as a modeling approach for the latent space to better capture the multimodal output space.
- A new inference method, Conditional Posterior Estimation (CXP), is proposed. This method constructs a more complex distribution for sampling during inference instead of directly using the trained latent space.

Through these improvements, the paper aims to address several key issues of CVAE in trajectory prediction tasks, including:
- Unreasonable or dangerous trajectory predictions caused by random sampling.
- An overly simplistic latent space that struggles to capture multimodal outputs.
- Ensuring that the generated trajectories have higher diversity and representativeness.

The experimental section demonstrates the superior performance of these improved methods on the INTERACTION dataset and also achieves good results in image modeling tasks on the CelebA dataset.