用生成模型控制快速动态任务，无需专家示范。
Generative Predictive Control: Flow Matching Policies for Dynamic and Difficult-to-Demonstrate Tasks
- 结合采样控制与生成建模，直接训练高速动态任务策略
- 训练后可在推理时快速热启动，保持时间一致性
- 适合仿真易但示范难的复杂机器人任务
生成式控制策略近期在机器人领域取得显著进展，通过扩散或流匹配生成动作序列，依赖专家示范进行训练。但现有方法存在两大局限：需获取专家示范（难以获得），且仅适用于较慢、准静态的任务。本文利用基于采样的预测控制与生成建模之间的紧密联系，提出生成式预测控制（Generative Predictive Control），一种针对动态快、易仿真但难示范的任务的监督学习框架。我们进一步展示，训练好的流匹配策略可在推理时实现热启动，维持时间一致性并支持高频反馈。该方法为行为克隆提供了互补路径，有望推动超越准静态示范任务的通用型策略发展。
原文摘要 · Abstract (English)
Generative control policies have recently unlocked major progress in robotics. These methods produce action sequences via diffusion or flow matching, with training data provided by demonstrations. But existing methods come with two key limitations: they require expert demonstrations, which can be difficult to obtain, and they are limited to relatively slow, quasi-static tasks. In this paper, we leverage a tight connection between sampling-based predictive control and generative modeling to address each of these issues. In particular, we introduce generative predictive control, a supervised learning framework for tasks with fast dynamics that are easy to simulate but difficult to demonstrate. We then show how trained flow-matching policies can be warm-started at inference time, maintaining temporal consistency and enabling high-frequency feedback. We believe that generative predictive control offers a complementary approach to existing behavior cloning methods, and hope that it paves the way toward generalist policies that extend beyond quasi-static demonstration-oriented tasks.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。