用生成模型提升复杂抓取任务的实时规划效率
Generative Models From and For Sampling-Based MPC: A Bootstrapped Approach For Adaptive Contact-Rich Manipulation
- 从仿真采样数据中训练生成模型，替代传统迭代优化
- 实测提升采样效率，减少规划所需时间步数
- 适用于真实四足机器人接触丰富的操作任务
我们提出一种生成式预测控制（GPC）框架，通过在仿真中收集的采样基模型预测控制（SPC）序列上训练条件流匹配模型，将SPC的计算开销分摊。与依赖迭代优化或梯度求解的方法不同，本工作首次证明可直接从噪声化的SPC数据中学习有意义的提案分布，从而在在线规划中实现更高效、更智能的采样。我们进一步首次将该方法应用于真实世界的接触丰富型四足机器人运动-操作任务。仿真与硬件实验均表明，该方法显著提升采样效率，降低规划时域需求，并在多种任务变化下保持良好泛化能力。
原文摘要 · Abstract (English)
We present a generative predictive control (GPC) framework that amortizes sampling-based Model Predictive Control (SPC) by bootstrapping it with conditional flow-matching models trained on SPC control sequences collected in simulation. Unlike prior work relying on iterative refinement or gradient-based solvers, we show that meaningful proposal distributions can be learned directly from noisy SPC data, enabling more efficient and informed sampling during online planning. We further demonstrate, for the first time, the application of this approach to real-world contact-rich loco-manipulation with a quadruped robot. Extensive experiments in simulation and on hardware show that our method improves sample efficiency, reduces planning horizon requirements, and generalizes robustly across task variations.
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。