行动、感知、行动：从大规模自我中心人类数据中学习主动感知
Act, Sense, Act: Learning Active Perception from Large-Scale Egocentric Human Data
AI总结：
研究如何让机器人在无约束环境中实现可泛化操作，提出CoMe - VLA框架，利用大规模人类自我中心数据学习主动感知，集成认知辅助头和双轨记忆系统，经三个阶段训练模型，实验证明该方法在多场景长期任务中有强鲁棒性和适应性。
摘要：
在无约束环境中实现可泛化操作需要机器人主动解决信息不确定性，即主动感知能力。现有方法常局限于有限类型的传感行为。本文将主动感知形式化为由信息寻求动作和决策分支驱动的依赖历史的感知 - 行动循环，给出视觉主动感知范式的结构化分类。在此基础上引入CoMe - VLA框架，利用大规模人类自我中心数据学习通用探索和操作先验。该框架集成认知辅助头和双轨记忆系统，通过融合本体感觉和视觉时间上下文维持自我和环境感知一致性。在统一自我中心动作空间中对齐人类和机器人手眼协调行为，分三个阶段逐步训练模型。基于轮式人形机器人的大量实验表明该方法在多种主动感知场景的长期任务中具有强大的鲁棒性和适应性。
英文摘要：
Achieving generalizable manipulation in unconstrained environments requires the robot to proactively resolve information uncertainty, i.e., the capability of active perception. However, existing methods are often confined in limited types of sensing behaviors, restricting their applicability to complex environments. In this work, we formalize active perception as a history-dependent perception-action loop driven by information-seeking action and decision branching, providing a structured categorization of visual active perception paradigms. Building on this perspective, we introduce CoMe-VLA, a cognitive and memory-aware vision-language-action (VLA) framework that leverages large-scale human egocentric data to learn versatile exploration and manipulation priors. Our framework integrates a cognitive auxiliary head for autonomous sub-task transitions and a dual-track memory system to maintain consistent self and environmental awareness by fusing proprioceptive and visual temporal contexts. By aligning human and robot hand-eye coordination behaviors in a unified egocentric action space, we train the model progressively in three stages. Extensive experiments on a wheel-based humanoid have demonstrated strong robustness and adaptability of our proposed method across diverse long-horizon tasks spanning multiple active perception scenarios.
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。
