让机器人在噪声中更稳，通过预测动作和不确定性实现精准控制
SUREFlow: State-space Uncertainty-aware REsidual Flow Matching for Robust Robot Manipulation
- 用Mamba架构联合预测动作速度与不确定性
- 在无反馈情况下提升动作可靠性，成功率达92.5%
- 小模型实现大模型效果，适合资源受限场景
生成式视觉-语言-动作策略虽推动了机器人操作发展，但在噪声、部分可观测性和随机初始条件下易不稳定。长时间推演中微小速度误差累积导致执行可靠性下降。现有基于扩散和流的方法通常假设残差同方差，且未显式建模动作生成中的不确定性，限制了迭代推演的鲁棒性。我们提出SUREFlow，一种基于Mamba骨干的时序状态空间不确定性感知残差流匹配框架。该方法联合预测动作速度与输入相关残差不确定性，可在无环境反馈下选择性优化不可靠动作维度，同时保持计算效率。在LIBERO上，SUREFlow平均成功率（SR）达92.5%，优于基于Mamba的MaIL方法34.2%；在LIBERO-PRO上仅用179M参数即达到约49%的SR，性能媲美拥有3-7B参数的大规模视觉-语言动作模型。代码已开源：https://github.com/tanvirnwu/SUREFlow
原文摘要 · Abstract (English)
Generative vision-language-action policies have advanced robot manipulation, but they often exhibit instability under noise, partial observability, and stochastic initial conditions. During extended rollouts, small velocity errors accumulate, degrading execution reliability. Existing diffusion and flow-based policies typically assume homoscedastic residuals and lack explicit uncertainty modeling within action generation, limiting robustness during iterative rollout. We propose SUREFlow, a state-space uncertainty-aware residual flow matching framework built on a Mamba backbone. The method jointly predicts action velocities and input-dependent residual uncertainty, enabling selective refinement of unreliable action dimensions without environment feedback while preserving computational efficiency. On LIBERO, SUREFlow achieves 92.5% average success rate (SR), outperforming the Mamba-based MaIL by 34.2%. On LIBERO-PRO, it attains around 49% SR using only 179M parameters, achieving performance comparable to large VLAs with 3-7B parameters. SUREFlow source code is available on: https://github.com/tanvirnwu/SUREFlow
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。