发表机构
Shenzhen Research Institute of Big Data; Shenzhen Institutes of Advanced Technology (SIAT), Chinese Academy of Sciences; The University of New South Wales; University of Macau; The University of Hong Kong（深圳大数据研究院; 中国科学院深圳先进技术研究院; 新南威尔士大学; 澳门大学; 香港大学）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本文提出SIGMA仿真在环框架，通过评估不确定性对规划成本的影响来决定云端VLM调用，以减少不必要的云端交互并提升自动驾驶导航成功率。
AI 中文摘要
大型视觉语言模型（VLM）为自动驾驶提供了强大的开放世界感知和推理能力，但其高昂的计算成本和推理延迟使得持续在云端使用不切实际。这促使了快慢协作（fast-slow collaboration）的产生，其中高效的车载模块负责实时感知和控制，而云端模型仅在需要时提供高层推理。关键挑战在于决定何时云端推理应影响时间关键的驾驶决策。现有方法通常依赖感知不确定性、启发式触发或资源驱动策略，而未评估解决不确定性是否会改善规划。我们提出了SIGMA，一个用于任务导向快慢协作的仿真在环框架。SIGMA将规划器嵌入不确定性评估中，评估语义和几何不确定性下合理的场景实现如何影响可行轨迹和规划成本。基于这些结果，它估计解决不确定性所带来的规划成本预期降低。我们进一步引入了预期规划增益（EPG），一种用于云端调用、云端引导集成以及在截止时间和资源约束下请求优先级的决策级指标。在CARLA中的实验表明，SIGMA减少了不必要的云端交互，同时提高了静态和动态障碍场景中的规划、效率和导航成功率。与固定周期协作相比，SIGMA将不必要的云端交互减少了50%，导航成功率提高了6%以上，并在动态场景中将完成时间缩短了最多26.2%。
英文摘要
Large vision-language models (VLMs) provide powerful open-world perception and reasoning for autonomous driving, but their high computational cost and inference latency make continuous cloud-side use impractical. This motivates fast--slow collaboration, where efficient onboard modules handle real-time perception and control while cloud models provide high-level reasoning only when needed. The key challenge is deciding when cloud reasoning should influence time-critical driving decisions. Existing methods often rely on perception uncertainty, heuristic triggers, or resource-driven policies, without assessing whether resolving an uncertainty will improve planning. We propose \textbf{SIGMA}, a simulation-in-the-loop framework for task-oriented fast--slow collaboration. SIGMA embeds the planner into uncertainty assessment and evaluates how plausible scene realizations under semantic and geometric uncertainty affect feasible trajectories and planning cost. Based on these outcomes, it estimates the expected reduction in planning cost from resolving uncertainty. We further introduce expected planning gain (EPG), a decision-level metric for cloud invocation, cloud-guidance integration, and request prioritization under deadline and resource constraints. Experiments in CARLA show that SIGMA reduces unnecessary cloud interactions while improving planning, efficiency, and navigation success in static and dynamic obstacle scenarios. Compared with fixed-period collaboration, SIGMA reduces unnecessary cloud interactions by 50\%, improves navigation success by more than 6\%, and cuts finish time by up to 26.2\% in dynamic scenarios.
