EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

Frequency-Domain Trajectory Planning for Autonomous Driving in Highly Dynamic Scenarios

Jie Xia, Zhuo Kong, Xiaodong Wu, Boran Shi, Yuanbo Han, Min Xu

Applied Sciences · 03.03.2026 · 16 (5) · 2447

DOI: 10.3390/app16052447 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 03.03.2026. Verfasst von Jie Xia und 5 weiteren Autor:innen. Die Arbeit nennt 47 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: National Natural Science Foundation of China.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Trajectory planning is a central problem in autonomous driving, requiring long-horizon reasoning, strict safety guarantees, and robustness to rare but critical events. Recent learning-based planners increasingly formulate planning as an autoregressive sequence generation problem, analogous to large language models, where future motions are discretized into action tokens and predicted by Transformer-based neural sequence models. Despite promising empirical results, most existing approaches adopt time-domain action representations, in which consecutive actions are highly correlated. When combined with autoregressive decoding, this design induces degenerate generation behavior in learning-based planners, encouraging local action continuation and leading to rapid error accumulation during closed-loop execution, particularly in safety-critical corner cases such as sudden pedestrian emergence. To address this limitation of time-domain autoregressive planning, we propose a unified trajectory planning framework built upon three core ideas: (1) explicit action tokenization for long-horizon planning, (2) transformation of the action space from the time domain to the frequency domain, and (3) a hybrid learning paradigm that combines imitation learning with reinforcement learning. By representing future motion using compact frequency-domain action coefficients rather than per-timestep actions, the proposed planner is encouraged to reason about global motion intent before refining local details. This change in action representation fundamentally alters the inductive bias of learning-based autoregressive planning, mitigates exposure bias, and enables earlier and more decisive responses in complex and safety-critical environments. We present the model formulation, learning objectives, and training strategy, and outline a comprehensive experimental protocol.

Förderung

  • National Natural Science Foundation of China

Zitieren

APA

Xia, J., Kong, Z., Wu, X., Shi, B., Han, Y., & Xu, M. (2026). Frequency-Domain Trajectory Planning for Autonomous Driving in Highly Dynamic Scenarios. Applied Sciences, 16(5), 2447. https://doi.org/10.3390/app16052447

Vancouver

Xia J, Kong Z, Wu X, Shi B, Han Y, Xu M. Frequency-Domain Trajectory Planning for Autonomous Driving in Highly Dynamic Scenarios. Applied Sciences. 2026;16(5):2447. doi:10.3390/app16052447

BibTeX
@article{Xia2026,
  author = {Xia, Jie and Kong, Zhuo and Wu, Xiaodong and Shi, Boran and Han, Yuanbo and Xu, Min},
  title = {Frequency-Domain Trajectory Planning for Autonomous Driving in Highly Dynamic Scenarios},
  journal = {Applied Sciences},
  year = {2026},
  volume = {16},
  number = {5},
  pages = {2447},
  doi = {10.3390/app16052447}
}
Kostenlos starten →