机器人与具身智能 颠覆级 有讲解视频
发表时间
2026-08-07
arXiv
2608.06729

收录解读

现有视觉-语言-动作(VLA)模型在部分可观测和长时程任务中受限于固有反应式范式,当仅依赖腕部相机时,存在物体离开视野导致的感知遗忘以及多步执行中的任务进度遗忘,严重制约了其性能表现。

AtlasVLA提出从直接反应式操作转向主动推理,通过双记忆架构(4D持久世界状态记忆和自我工作状态记忆)构建全局更新的体素哈希空间状态,并用扩散Transformer联合世界-自我状态进行空间推理,解决了视觉盲区和进度遗忘问题。

实验覆盖LIBERO、RLBench和真实世界基准,仅使用腕部相机即达到当前最优性能,在LIBERO-Long上相对多视角基线绝对成功率提升9.4%,在真实世界长时程任务上提升17.5%,展示了跨基准的显著效果。

这一工作为VLA模型注入了持久状态和推理能力,但当前证据仅限于摘要,未报告消融、统计细节或代码开源,且局限性在于仅验证了操作任务,仍未超出操作领域,未来需评估其在更多具身任务中的泛化能力。

原始摘要

While Vision-Language-Action (VLA) models have advanced embodied AI, their fundamentally reactive paradigm severely limits performance in partially observable and long-horizon tasks. When restricted to a single wrist-mounted camera, they inevitably suffer from perception forgetting as objects exit the field of view, and temporal task-progress forgetting} during multi-step execution. To overcome these bottlenecks, we propose AtlasVLA, a novel framework that transitions from direct reactive manipulation to proactive reasoning through a persistent world-ego state. AtlasVLA features a dual-memory architecture: a 4D Persistent World State Memory that lifts transient 2D observations into a globally updated, voxel-hashed spatial state to resolve visual blind spots, and an Ego-Working State Memory that tracks historical ego state and task progress. By conditioning a diffusion transformer (DiT) on this joint World-Ego state, AtlasVLA enables robust spatial reasoning. Extensive evaluations across LIBERO, RLBench, and real-world benchmarks demonstrate that AtlasVLA achieves state-of-the-art performance using solely a wrist camera. Remarkably, it decisively outperforms multi-view baselines, yielding absolute success rate improvements of 9.4% on LIBERO-Long and 17.5% in real-world long-horizon tasks.

解读视频

相关论文

链接