arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.04161cs.RO
收藏

推理时的自适应动作分块

Adaptive Action Chunking at Inference-time for Vision-Language-Action Models

National University of Singapore(新加坡国立大学);Shenzhen University of Advanced Technology(深圳理工大学);Sangfor Technologies Inc.(深信服科技股份有限公司);Mininglamp Technology(明略科技);Shenzhen Technology University(深圳技术大学);City University of Hong Kong(香港城市大学)

Yuanchang Liang, Xiaobo Wang, Kai Wang, Shuo Wang, Xiaojiang Peng, Haoyu Chen, David Kim Huat Chua, Prahlad Vadakkepat

AI总结:

本文提出自适应动作分块策略,通过动作熵动态调整分块大小,提升机器人操作的反应与一致性。

摘要:

在视觉-语言-动作(VLA)模型中,动作分块(即执行一系列动作而不进行中间重新规划)是提高机器人操作能力的关键技术。然而,较大的分块大小会降低模型对新信息的响应能力,而较小的分块大小则容易导致模式跳跃和不连贯的行为。因此,选择最佳的分块大小是平衡模型反应性和一致性的重要需求。不幸的是,当前VLA模型的一个主要趋势是在推理时采用经验固定的分块长度,限制了其在多样化操作任务中的优越性和扩展性。为了解决这个问题,我们提出了一种新颖的自适应动作分块(AAC)策略,利用动作熵作为线索,根据当前预测动态确定分块大小。在广泛模拟和现实世界机器人操作任务上的大量实验表明,我们的方法在性能上显著优于现有最先进的方法。视频和源代码可在https://lance-lot.github.io/adaptive-chunking.github.io/上公开获取。

英文摘要:

In Vision-Language-Action (VLA) models, action chunking (i.e., executing a sequence of actions without intermediate replanning) is a key technique to improve robotic manipulation abilities. However, a large chunk size reduces the model's responsiveness to new information, while a small one increases the likelihood of mode-jumping, jerky behavior resulting from discontinuities between chunks. Therefore, selecting the optimal chunk size is an urgent demand to balance the model's reactivity and consistency. Unfortunately, a dominant trend in current VLA models is an empirical fixed chunk length at inference-time, hindering their superiority and scalability across diverse manipulation tasks. To address this issue, we propose a novel Adaptive Action Chunking (AAC) strategy, which exploits action entropy as the cue to adaptively determine the chunk size based on current predictions. Extensive experiments on a wide range of simulated and real-world robotic manipulation tasks have demonstrated that our approach substantially improves performance over the state-of-the-art alternatives. The videos and source code are publicly available at https://lance-lot.github.io/adaptive-chunking.github.io/.

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。

↑