Controllable Generative Trajectory Prediction via Weak Preference Alignment

Fuente: arXiv
Gespeichert in:
Bibliographische Detailangaben
Hauptverfasser: Cao, Yongxi, Schumann, Julian F., Kober, Jens, Pajarinen, Joni, Zgonnikov, Arkady
Format: Preprint
Veröffentlicht: 2025
Schlagworte:
Online-Zugang:
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
_version_ 1866915548920545280
author Cao, Yongxi
Schumann, Julian F.
Kober, Jens
Pajarinen, Joni
Zgonnikov, Arkady
author_facet Cao, Yongxi
Schumann, Julian F.
Kober, Jens
Pajarinen, Joni
Zgonnikov, Arkady
contents Deep generative models such as conditional variational autoencoders (CVAEs) have shown great promise for predicting trajectories of surrounding agents in autonomous vehicle planning. State-of-the-art models have achieved remarkable accuracy in such prediction tasks. Besides accuracy, diversity is also crucial for safe planning because human behaviors are inherently uncertain and multimodal. However, existing methods generally lack a scheme to generate controllably diverse trajectories, which is arguably more useful than randomly diversified trajectories, to the end of safe planning. To address this, we propose PrefCVAE, an augmented CVAE framework that uses weakly labeled preference pairs to imbue latent variables with semantic attributes. Using average velocity as an example attribute, we demonstrate that PrefCVAE enables controllable, semantically meaningful predictions without degrading baseline accuracy. Our results show the effectiveness of preference supervision as a cost-effective way to enhance sampling-based generative models.
format Preprint
id arxiv_https___arxiv_org_abs_2510_10731
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Controllable Generative Trajectory Prediction via Weak Preference Alignment
Cao, Yongxi
Schumann, Julian F.
Kober, Jens
Pajarinen, Joni
Zgonnikov, Arkady
Robotics
Machine Learning
Deep generative models such as conditional variational autoencoders (CVAEs) have shown great promise for predicting trajectories of surrounding agents in autonomous vehicle planning. State-of-the-art models have achieved remarkable accuracy in such prediction tasks. Besides accuracy, diversity is also crucial for safe planning because human behaviors are inherently uncertain and multimodal. However, existing methods generally lack a scheme to generate controllably diverse trajectories, which is arguably more useful than randomly diversified trajectories, to the end of safe planning. To address this, we propose PrefCVAE, an augmented CVAE framework that uses weakly labeled preference pairs to imbue latent variables with semantic attributes. Using average velocity as an example attribute, we demonstrate that PrefCVAE enables controllable, semantically meaningful predictions without degrading baseline accuracy. Our results show the effectiveness of preference supervision as a cost-effective way to enhance sampling-based generative models.
title Controllable Generative Trajectory Prediction via Weak Preference Alignment
topic Robotics
Machine Learning
url https://arxiv.org/abs/2510.10731