arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2604.02408cs.RO
收藏

F2F-AP: 流向未来异步策略用于实时动态操控

F2F-AP: Flow-to-Future Asynchronous Policy for Real-time Dynamic Manipulation

Department of Automation, Tsinghua University(清华大学自动化系)

Haoyu Wei, Xiuwei Xu, Ziyang Cheng, Hang Yin, Angyuan Ma, Jie Zhou, Jiwen Lu, Bingyao Yu

AI总结:

本文提出F2F-AP策略,通过预测物体流合成未来观测,利用基于流的对比学习目标对齐视觉特征,提升实时动态操控任务的响应性和成功率。

摘要:

异步推断已成为机器人操控中的主流范式,实现了轨迹平滑和效率的显著进步。然而,固有的延迟导致生成的动作不可避免地滞后于实时环境。在动态场景中,这种时间错位严重削弱了策略对快速变化环境的解读和反应能力。本文提出了一种新的框架,利用预测的物体流来合成未来观测,并结合基于流的对比学习目标,对齐预测观测的视觉特征表示与真实未来状态。借助这种预期的视觉上下文,我们的异步策略具备主动规划和运动的能力,使其能够显式补偿延迟并稳健地执行涉及主动移动物体的操控任务。实验结果表明,我们的方法在复杂动态操控任务中显著提高了响应性和成功率。

英文摘要:

Asynchronous inference has emerged as a prevalent paradigm in robotic manipulation, achieving significant progress in ensuring trajectory smoothness and efficiency. However, a systemic challenge remains unresolved, as inherent latency causes generated actions to inevitably lag behind the real-time environment. This issue is particularly exacerbated in dynamic scenarios, where such temporal misalignment severely compromises the policy's ability to interpret and react to rapidly evolving surroundings. In this paper, we propose a novel framework that leverages predicted object flow to synthesize future observations, incorporating a flow-based contrastive learning objective to align the visual feature representations of predicted observations with ground-truth future states. Empowered by this anticipated visual context, our asynchronous policy gains the capacity for proactive planning and motion, enabling it to explicitly compensate for latency and robustly execute manipulation tasks involving actively moving objects. Experimental results demonstrate that our approach significantly enhances responsiveness and success rates in complex dynamic manipulation tasks.

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。

↑