End-to-End paradigms use a unified framework to implement multi-tasks in an autonomous driving system. Despite simplicity and clarity, the performance of end-to-end autonomous driving methods on sub-tasks is still far behind the single-task methods. Meanwhile, the widely used dense BEV features in previous end-to-end methods make it costly to extend to more modalities or tasks. In this paper, we propose a Sparse query-centric paradigm for end-to-end Autonomous Driving (SparseAD), where the sparse queries completely represent the whole driving scenario across space, time and tasks without any dense BEV representation. Concretely, we design a unified sparse architecture for perception tasks including detection, tracking, and online mapping. Moreover, we revisit motion prediction and planning, and devise a more justifiable motion planner framework. On the challenging nuScenes dataset, SparseAD achieves SOTA full-task performance among end-to-end methods and significantly narrows the performance gap between end-to-end paradigms and single-task methods. Codes will be released soon.
@article{arxiv.2404.06892,
title = {SparseAD: Sparse Query-Centric Paradigm for Efficient End-to-End Autonomous Driving},
author = {Diankun Zhang and Guoan Wang and Runwen Zhu and Jianbo Zhao and Xiwu Chen and Siyu Zhang and Jiahao Gong and Qibin Zhou and Wenyuan Zhang and Ningzi Wang and Feiyang Tan and Hangning Zhou and Ziyao Xu and Haotian Yao and Chi Zhang and Xiaojun Liu and Xiaoguang Di and Bin Li},
journal= {arXiv preprint arXiv:2404.06892},
year = {2024}
}