Salvato in:
Dettagli Bibliografici
Autori principali: Wu, Zhengfei, Pan, Shuaixi, Chen, Shuohan, Yang, Shuo, Huang, Yanjun
Natura: Preprint
Pubblicazione: 2026
Soggetti:
Accesso online:https://arxiv.org/abs/2602.03112
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911420488089600
author Wu, Zhengfei
Pan, Shuaixi
Chen, Shuohan
Yang, Shuo
Huang, Yanjun
author_facet Wu, Zhengfei
Pan, Shuaixi
Chen, Shuohan
Yang, Shuo
Huang, Yanjun
contents End-to-end autonomous driving is increasingly adopting a multimodal planning paradigm that generates multiple trajectory candidates and selects the final plan, making candidate-set design critical. A fixed trajectory vocabulary provides stable coverage in routine driving but often misses optimal solutions in complex interactions, while scene-adaptive refinement can cause over-correction in simple scenarios by unnecessarily perturbing already strong vocabulary trajectories.We propose CdDrive, which preserves the original vocabulary candidates and augments them with scene-adaptive candidates generated by vocabulary-conditioned diffusion denoising. Both candidate types are jointly scored by a shared selection module, enabling reliable performance across routine and highly interactive scenarios. We further introduce HATNA (Horizon-Aware Trajectory Noise Adapter) to improve the smoothness and geometric continuity of diffusion candidates via temporal smoothing and horizon-aware noise modulation. Experiments on NAVSIM v1 and NAVSIM v2 demonstrate leading performance, and ablations verify the contribution of each component. Code: https://github.com/WWW-TJ/CdDrive.
format Preprint
id arxiv_https___arxiv_org_abs_2602_03112
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving
Wu, Zhengfei
Pan, Shuaixi
Chen, Shuohan
Yang, Shuo
Huang, Yanjun
Robotics
End-to-end autonomous driving is increasingly adopting a multimodal planning paradigm that generates multiple trajectory candidates and selects the final plan, making candidate-set design critical. A fixed trajectory vocabulary provides stable coverage in routine driving but often misses optimal solutions in complex interactions, while scene-adaptive refinement can cause over-correction in simple scenarios by unnecessarily perturbing already strong vocabulary trajectories.We propose CdDrive, which preserves the original vocabulary candidates and augments them with scene-adaptive candidates generated by vocabulary-conditioned diffusion denoising. Both candidate types are jointly scored by a shared selection module, enabling reliable performance across routine and highly interactive scenarios. We further introduce HATNA (Horizon-Aware Trajectory Noise Adapter) to improve the smoothness and geometric continuity of diffusion candidates via temporal smoothing and horizon-aware noise modulation. Experiments on NAVSIM v1 and NAVSIM v2 demonstrate leading performance, and ablations verify the contribution of each component. Code: https://github.com/WWW-TJ/CdDrive.
title A Unified Candidate Set with Scene-Adaptive Refinement via Diffusion for End-to-End Autonomous Driving
topic Robotics
url https://arxiv.org/abs/2602.03112