arXiv 2606.147722026-06-16cs.CVcs.AI

ScoutVLA：面向开放世界具身问答的无人机中心主动感知双专家VLA模型

ScoutVLA: UAV-Centric Active Perception via a Dual-Expert VLA Model for Open-World Embodied Question Answering

National Key Laboratory of Digital Intelligent Modeling and Simulation, National University of Defense Technology（国防科技大学数字智能建模与仿真国家重点实验室）；GigaAI（极佳科技）

Wenhao Lu, Zhengqiu Zhu, Xiaofeng Wang, Xiaoran Zhang, Yatai Ji, Yong Zhao, Yue Hu, Yingzhen Nie, Jinlong Zhu, Zheng Zhu

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

针对无人机在室外具身问答中细粒度视角调整不足的问题，提出ScoutVLA模型，采用解耦双专家架构（视觉语言专家推断语义意图，动作专家生成连续视角调整轨迹），并通过知识隔离机制平衡连续控制与语义推理，在仿真和真实实验中显著优于基线方法。

摘要：

空中具身问答（EQA）要求无人机（UAV）主动感知环境并回答自然语言问题。现有的室外EQA系统通常在目标进入无人机视野后停止，导致寻找证据所需的问题的细粒度视角调整问题仍未解决。为解决此问题，我们引入FG-EQA，一个细粒度主动感知EQA基准，包含超过4万条模拟轨迹和1千条真实轨迹。受侦察蜂“摇摆舞”的启发（它们迭代调整飞行路径以验证目标信息），我们提出ScoutVLA，一种用于室外EQA的证据驱动视觉-语言-动作模型。为模拟这种主动探索行为，ScoutVLA采用解耦双专家架构：视觉语言专家推断语义意图以识别缺失证据，而独立动作专家使用高自由度流匹配生成连续视角调整轨迹。为平衡连续控制和语义推理的竞争需求，我们设计了一种解耦训练策略，其中包含知识隔离机制，防止动作梯度抹除模型的多模态推理能力。大量仿真实验和定性真实世界实地研究均验证了ScoutVLA相对于最先进基线的优越性，平均严格成功率高10.48倍，平均QA正确率高7.72倍。

英文摘要：

Aerial Embodied Question Answering (EQA) requires Unmanned Aerial Vehicles (UAVs) to actively perceive the environment and answer natural language questions. Existing outdoor EQA systems usually stop once the target enters the UAV's field of view, leaving the fine-grained viewpoint adjustment needed for evidence-seeking questions largely unresolved. To address this issue, we introduce FG-EQA, a fine-grained active perception EQA benchmark with more than 40K simulated trajectories and 1K real-world trajectories. Drawing inspiration from the ``waggle dance'' of scout bees, which iteratively adjust their flight paths to verify target information, we propose ScoutVLA, an evidence-driven Vision-Language-Action model for outdoor EQA. To emulate this active exploration behavior, ScoutVLA features a decoupled dual-expert architecture: a vision-language expert infers the semantic intent to identify missing evidence, while an independent action expert employs high-DoF flow matching to generate continuous viewpoint-refinement trajectories. To balance the competing demands of continuous control and semantic reasoning, we devise a decoupled training strategy with a knowledge insulation mechanism that prevents the action gradients from erasing the model's multimodal reasoning ability. Extensive simulated experiments and a qualitative real-world field study both verify the superiority of ScoutVLA over the state-of-the-art baselines, demonstrating a 10.48$\boldsymbol{\times}$ higher average strict success rate and a 7.72$\boldsymbol{\times}$ higher average QA correctness.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。