Saved in:
Bibliographic Details
Main Authors: Zong, Zhenyu, Wang, Yuchen, Lin, Haohong, Gan, Lu, Shao, Huajie
Format: Preprint
Published: 2026
Subjects:
Online Access:https://arxiv.org/abs/2602.13936
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866911449587122176
author Zong, Zhenyu
Wang, Yuchen
Lin, Haohong
Gan, Lu
Shao, Huajie
author_facet Zong, Zhenyu
Wang, Yuchen
Lin, Haohong
Gan, Lu
Shao, Huajie
contents Trajectory prediction for traffic agents is critical for safe autonomous driving. However, achieving effective zero-shot generalization in previously unseen domains remains a significant challenge. Motivated by the consistent nature of kinematics across diverse domains, we aim to incorporate domain-invariant knowledge to enhance zero-shot trajectory prediction capabilities. The key challenges include: 1) effectively extracting domain-invariant scene representations, and 2) integrating invariant features with kinematic models to enable generalized predictions. To address these challenges, we propose a novel generalizable Physics-guided Causal Model (PCM), which comprises two core components: a Disentangled Scene Encoder, which adopts intervention-based disentanglement to extract domain-invariant features from scenes, and a CausalODE Decoder, which employs a causal attention mechanism to effectively integrate kinematic models with meaningful contextual information. Extensive experiments on real-world autonomous driving datasets demonstrate our method's superior zero-shot generalization performance in unseen cities, significantly outperforming competitive baselines. The source code is released at https://github.com/ZY-Zong/Physics-guided-Causal-Model.
format Preprint
id arxiv_https___arxiv_org_abs_2602_13936
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Generalizable Physics-guided Causal Model for Trajectory Prediction in Autonomous Driving
Zong, Zhenyu
Wang, Yuchen
Lin, Haohong
Gan, Lu
Shao, Huajie
Artificial Intelligence
Trajectory prediction for traffic agents is critical for safe autonomous driving. However, achieving effective zero-shot generalization in previously unseen domains remains a significant challenge. Motivated by the consistent nature of kinematics across diverse domains, we aim to incorporate domain-invariant knowledge to enhance zero-shot trajectory prediction capabilities. The key challenges include: 1) effectively extracting domain-invariant scene representations, and 2) integrating invariant features with kinematic models to enable generalized predictions. To address these challenges, we propose a novel generalizable Physics-guided Causal Model (PCM), which comprises two core components: a Disentangled Scene Encoder, which adopts intervention-based disentanglement to extract domain-invariant features from scenes, and a CausalODE Decoder, which employs a causal attention mechanism to effectively integrate kinematic models with meaningful contextual information. Extensive experiments on real-world autonomous driving datasets demonstrate our method's superior zero-shot generalization performance in unseen cities, significantly outperforming competitive baselines. The source code is released at https://github.com/ZY-Zong/Physics-guided-Causal-Model.
title A Generalizable Physics-guided Causal Model for Trajectory Prediction in Autonomous Driving
topic Artificial Intelligence
url https://arxiv.org/abs/2602.13936