SparseAD: Sparse Query-Centric Paradigm for Efficient End-to-End Autonomous Driving

End-to-end paradigms use a unified framework to implement multitasks in an autonomous driving system. Despite simplicity and clarity, the performance of end-to-end autonomous driving methods on subtasks is still far behind the single-task methods. Meanwhile, the widely used dense bird’s-eye-view (BEV) features in previous end-to-end methods make it costly to extend to more modalities or tasks. In this article, 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 state-of-the-art full-task performance among end-to-end methods and significantly narrows the performance gap between end-to-end paradigms and single-task methods.