Fast Generative Grasping via Lie Group-Constrained MeanFlow
2026-08-26 • Robotics
Robotics
AI summaryⓘ
The authors focus on teaching robots how to quickly find good ways to grab objects, which is tricky because many possible grasps exist. They use a new mathematics-based method called MeanFlow on a special space to speed up the process of suggesting grasps without losing quality. Their method runs much faster than previous ones and works well even when tested on real robots without extra training. This helps robots grasp objects reliably and quickly, which is important for practical use.
grasp synthesisrobotic manipulationgenerative modelsdiffusion modelsflow-based modelsLie groupsMeanFlowRiemannian geometryACRONYM datasetnetwork evaluation
Authors
S. Talha Bukhari, Yi Wei, Ruiqi Ni, Zachary Kingston, Aniket Bera
Abstract
Grasp synthesis is a core task in robotic manipulation, for which the solution typically forms a multimodal distribution rather than a point estimate. Generative robotic grasping aims to learn this distribution with deep generative models such as diffusion and flow-based approaches. The iterative nature of such generative models makes them flexible and generalizable; however, multi-step sampling impedes the time-critical operation required in robotics. We devise an approach to fast generative grasping based on MeanFlow on the product Lie group $\mathcal{G} = \mathrm{SO}(3) \times \mathbb{R}^3$. The training objective couples a purely algebraic semigroup consistency condition with Riemannian Conditional Flow Matching on $\mathcal{G}$ that anchors the average velocity to the data distribution. The resulting Lie Group-constrained MeanFlow formulation samples reliable grasps in $\leq 5$ network evaluations, matching the grasp generation performance of state-of-the-art diffusion and flow-based models on the ACRONYM dataset at millisecond-scale inference latency (up to $39\times$ speed-up). We further demonstrate that the approach directly translates to real-world robotic grasping without additional training or domain adaptation, exhibiting robust grasp synthesis under observation noise.