用线性分组RNN统一处理自动驾驶多模态数据，高效且性能强。
UniLION: Towards Unified Autonomous Driving Model with Linear Group RNNs
- 采用线性分组RNN处理点云、图像和时序数据
- 在多个任务上达到顶尖性能，无需额外融合模块
- 适合构建统一的多模态自动驾驶系统
尽管变换器在多个领域表现出色，但其二次注意力机制在处理长序列数据时计算开销大。本文提出统一自动驾驶模型UniLION，基于线性分组RNN（linear group RNN）操作，高效处理大规模激光雷达点云、高分辨率多视角图像及时间序列数据。令人瞩目的是，UniLION作为单一通用架构，可无缝支持多种专用变体（如仅激光雷达、时序激光雷达、多模态、多模态时序融合），无需显式的时间或跨模态融合模块。此外，该模型在3D感知（如3D目标检测、3D目标跟踪、3D占用预测、鸟瞰图地图分割）、预测（如运动预测）和规划（如端到端规划）等核心任务中均表现出竞争力甚至领先性能。这一统一范式自然简化了多模态与多任务自动驾驶系统的设计，同时保持卓越性能。我们希望UniLION为自动驾驶3D基础模型的发展提供新思路。代码已开源：https://github.com/happinesslz/UniLION。
原文摘要 · Abstract (English)
Although transformers have demonstrated remarkable capabilities across various domains, their quadratic attention mechanisms introduce significant computational overhead when processing long-sequence data. In this paper, we present a unified autonomous driving model, UniLION, which efficiently handles large-scale LiDAR point clouds, high-resolution multi-view images, and even temporal sequences based on the linear group RNN operator (i.e., performs linear RNN for grouped features). Remarkably, UniLION serves as a single versatile architecture that can seamlessly support multiple specialized variants (i.e., LiDAR-only, temporal LiDAR, multi-modal, and multi-modal temporal fusion configurations) without requiring explicit temporal or multi-modal fusion modules. Moreover, UniLION consistently delivers competitive and even state-of-the-art performance across a wide range of core tasks, including 3D perception (e.g., 3D object detection, 3D object tracking, 3D occupancy prediction, BEV map segmentation), prediction (e.g., motion prediction), and planning (e.g., end-to-end planning). This unified paradigm naturally simplifies the design of multi-modal and multi-task autonomous driving systems while maintaining superior performance. Ultimately, we hope UniLION offers a fresh perspective on the development of 3D foundation models in autonomous driving. Code is available at https://github.com/happinesslz/UniLION
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。