CNDM: Customized Noise Diffusion Model for Trajectory Prediction
IEEE Transactions on Intelligent Transportation Systems · 2026-07 · 27 (7) · 8100-8112
DOI: 10.1109/tits.2026.3681222 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Intelligent Transportation Systems, 2026-07. Verfasst von Entao Chang und 5 weiteren Autor:innen. Die Arbeit nennt 44 Literaturangaben. Gefördert von: National Natural Science Foundation of China.
Abstract (Originalsprache)
Precise and reliable multi-agent trajectory prediction is fundamental to enabling safe autonomous navigation. While denoising diffusion models have demonstrated remarkable capability in capturing the multimodal uncertainty inherent in this task, their reliance on a generic, isotropic Gaussian noise prior poses a critical limitation. This uniform prior is fundamentally misaligned with the highly structured, heterogeneous uncertainty of agent motion—shaped by dynamics, type, and environmental constraints—forcing the denoising network to implicitly relearn complex motion priors from scratch. To bridge this gap, we propose the Customized Noise Diffusion Model (CNDM), a novel framework that introduces a learned, agent-specific noise prior. At the core of CNDM is a Prior-Guidance Network (PGN) that distills an agent’s history, type, and scene context into a parametric, anisotropic Gaussian distribution. This customized prior provides a physically-grounded starting point for the diffusion process. To enable efficient training on such anisotropic noise, we leverage a Mahalanobis whitening transformation to standardize the denoising task. Extensive experiments on the Waymo Open Motion and Argoverse 2 datasets show that CNDM achieves competitive performance against state-of-the-art methods, excelling particularly in capturing diverse motion patterns and improving probabilistic calibration. Ablation studies confirm that the performance gains are directly attributable to our customized noise design, underscoring the importance of integrating structured domain knowledge into the generative foundation of diffusion models.
Förderung
- National Natural Science Foundation of China
Zitieren
APA
Chang, E., Fu, J., Gao, W., Chen, S., Xin, J., & Zheng, N. (2026). CNDM: Customized Noise Diffusion Model for Trajectory Prediction. IEEE Transactions on Intelligent Transportation Systems, 27(7), 8100-8112. https://doi.org/10.1109/tits.2026.3681222
Vancouver
Chang E, Fu J, Gao W, Chen S, Xin J, Zheng N. CNDM: Customized Noise Diffusion Model for Trajectory Prediction. IEEE Transactions on Intelligent Transportation Systems. 2026;27(7):8100-8112. doi:10.1109/tits.2026.3681222
BibTeX
@article{Chang2026,
author = {Chang, Entao and Fu, Jiawei and Gao, Wenjie and Chen, Shitao and Xin, Jingmin and Zheng, Nanning},
title = {CNDM: Customized Noise Diffusion Model for Trajectory Prediction},
journal = {IEEE Transactions on Intelligent Transportation Systems},
year = {2026},
volume = {27},
number = {7},
pages = {8100--8112},
doi = {10.1109/tits.2026.3681222}
}
EUVIMED