GeCo-SRT：面向跨任务仿真到现实迁移的几何感知持续适应
GeCo-SRT: Geometry-aware Continual Adaptation for Cross-Task Sim-to-Real Transfer
把局部几何特征当作域不变、任务不变的迁移基底，并用几何感知的混合专家加优先经验回放来累积跨任务仿真到现实的经验，这套「几何先验+专家路由+经验回放」的组合可直接迁移到视觉策略的持续域适应与机器人操作策略学习中。
原文依据 It utilizes domain-invariant and task-invariant knowledge from local geometric features as a transferable foundation
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论弥合仿真到现实的差距，对于将低成本仿真数据应用于真实机器人系统至关重要。然而，此前的方法严重受限于将每次迁移视为孤立的任务，需要反复且昂贵的调参，并浪费了先前的迁移经验。为超越孤立的仿真到现实迁移，我们构建了一种以跨迭代迁移的知识累积为核心的持续跨任务仿真到现实迁移范式，从而实现对新颖任务的有效且高效的适应。为此，我们提出 GeCo-SRT，一种几何感知的持续适应方法。它利用来自局部几何特征的域不变与任务不变知识作为可迁移的基础，以加速后续仿真到现实迁移过程中的适应。该方法首先采用几何感知的混合专家模块，该模块动态激活专家以专门处理不同的几何知识，从而弥合观测层面的仿真到现实差距。进一步，几何专家引导的优先经验回放模块优先从未被充分利用的专家中采样，刷新专门知识以对抗遗忘并保持稳健的跨任务性能。借助迭代迁移过程中累积的知识，GeCo-SRT 方法不仅相较基线取得 52% 的平均性能提升，还展现出显著的数据效率，仅用 1/6 的数据即可适应新任务。我们希望这项工作能为高效、低成本的跨任务仿真到现实迁移方法带来启发。
英文原文摘要 1513 字符
Bridging the sim-to-real gap is important for applying low-cost simulation data to real-world robotic systems. However, previous methods are severely limited by treating each transfer as an isolated endeavor, demanding repeated, costly tuning and wasting prior transfer experience.To move beyond isolated sim-to-real, we build a continual cross-task sim-to-real transfer paradigm centered on knowledge accumulation across iterative transfers, thereby enabling effective and efficient adaptation to novel tasks. Thus, we propose GeCo-SRT, a geometry-aware continual adaptation method. It utilizes domain-invariant and task-invariant knowledge from local geometric features as a transferable foundation to accelerate adaptation during subsequent sim-to-real transfers. This method starts with a geometry-aware mixture-of-experts module, which dynamically activates experts to specialize in distinct geometric knowledge to bridge observation sim-to-real gap. Further, the geometry-expert-guided prioritized experience replay module preferentially samples from underutilized experts, refreshing specialized knowledge to combat forgetting and maintain robust cross-task performance.Leveraging knowledge accumulated during iterative transfer, GeCo-SRT method not only achieves 52% average performance improvement over the baseline, but also demonstrates significant data efficiency for new task adaptation with only 1/6 data.We hope this work inspires approaches for efficient, low-cost cross-task sim-to-real transfer.