用智能门控减少大模型调用，让自动驾驶规划又快又准
ASSCG: Just-Right Gating over Chattering for Fast-Slow LLM Planning in Autonomous Driving
- 基于长时序决策设计动态门控，每帧决定是否调用慢速大模型
- 在nuPlan上降低60%延迟，分数提升2.28；在NAVSIM上速度提高25%
- 适合追求实时性与高精度的自动驾驶系统部署
大语言模型能提升自动驾驶规划能力，但在线调用成本高。现有快慢系统依赖人工规则，常过度调用或时机不准。本文将慢模型调用建模为资源感知的序列决策问题，提出自适应慢系统控制门控（ASSCG），实现帧级查询、缓存或丢弃决策，以刷新、复用或抑制慢模型引导。ASSCG采用RWKV架构实现高效长时序控制，并通过监督微调结合类似GRPO的算力感知强化学习进行训练。在两种架构上验证：(i) AsyncDriver在nuPlan Hard20闭环测试中，分数达67.28（+2.28），平均推理延迟降低60%；(ii) 自建RecogDrive双系统，用轻量ViT快模型替换原VLM-2B模块并加入LLM慢模型，在NAVSIM上达成91.4 PDMS（+0.6），平均速度提升25%。
原文摘要 · Abstract (English)
Large language models (LLMs) can improve autonomous driving planning but are costly to query online, and existing fast-slow planners often rely on hand-designed triggering rules that either over-call the slow system or call it at the wrong times. We formulate slow-system invocation as a resource-aware sequential decision problem and propose the Adaptive Slow-System Control Gate (ASSCG), which makes frame-level Query/Cache/Drop decisions to refresh, reuse, or suppress slow guidance. ASSCG uses an RWKV backbone for efficient long-horizon gating and is trained with supervised fine-tuning followed by GRPO-style compute-aware reinforcement fine-tuning. We apply ASSCG to two different fast-slow architectures: (i) AsyncDriver on nuPlan Hard20 closed-loop evaluation, where ASSCG improves score to 67.28 (+2.28) while reducing average end-to-end inference latency by 60%; and (ii) a RecogDrive-based dual system that we build by replacing its original VLM-2B module with a lightweight ViT-based fast planner and adding an LLM slow planner, evaluated on NAVSIM, where ASSCG achieves 91.4 PDMS (+0.6) and increases average speed by 25%. The project page, including video visualizations and additional results, is available at https://williamxuanyu.github.io/asscg/.
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。