Saved in:
| Main Authors: | , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2503.12170 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866908455204290560 |
|---|---|
| author | Wang, Tao Zhang, Cong Qu, Xingguang Li, Kun Liu, Weiwei Huang, Chang |
| author_facet | Wang, Tao Zhang, Cong Qu, Xingguang Li, Kun Liu, Weiwei Huang, Chang |
| contents | End-to-end autonomous driving (E2E-AD) has rapidly emerged as a promising approach toward achieving full autonomy. However, existing E2E-AD systems typically adopt a traditional multi-task framework, addressing perception, prediction, and planning tasks through separate task-specific heads. Despite being trained in a fully differentiable manner, they still encounter issues with task coordination, and the system complexity remains high. In this work, we introduce DiffAD, a novel diffusion probabilistic model that redefines autonomous driving as a conditional image generation task. By rasterizing heterogeneous targets onto a unified bird's-eye view (BEV) and modeling their latent distribution, DiffAD unifies various driving objectives and jointly optimizes all driving tasks in a single framework, significantly reducing system complexity and harmonizing task coordination. The reverse process iteratively refines the generated BEV image, resulting in more robust and realistic driving behaviors. Closed-loop evaluations in Carla demonstrate the superiority of the proposed method, achieving a new state-of-the-art Success Rate and Driving Score. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2503_12170 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | DiffAD: A Unified Diffusion Modeling Approach for Autonomous Driving Wang, Tao Zhang, Cong Qu, Xingguang Li, Kun Liu, Weiwei Huang, Chang Robotics Computer Vision and Pattern Recognition End-to-end autonomous driving (E2E-AD) has rapidly emerged as a promising approach toward achieving full autonomy. However, existing E2E-AD systems typically adopt a traditional multi-task framework, addressing perception, prediction, and planning tasks through separate task-specific heads. Despite being trained in a fully differentiable manner, they still encounter issues with task coordination, and the system complexity remains high. In this work, we introduce DiffAD, a novel diffusion probabilistic model that redefines autonomous driving as a conditional image generation task. By rasterizing heterogeneous targets onto a unified bird's-eye view (BEV) and modeling their latent distribution, DiffAD unifies various driving objectives and jointly optimizes all driving tasks in a single framework, significantly reducing system complexity and harmonizing task coordination. The reverse process iteratively refines the generated BEV image, resulting in more robust and realistic driving behaviors. Closed-loop evaluations in Carla demonstrate the superiority of the proposed method, achieving a new state-of-the-art Success Rate and Driving Score. |
| title | DiffAD: A Unified Diffusion Modeling Approach for Autonomous Driving |
| topic | Robotics Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2503.12170 |