cs.ROOct 30, 2026

Direct Action-Manifold Modeling for Efficient and Robust Bimanual Manipulation

Authors: Hongren Wang, Mingyu You, Zihan Li, Zhipeng Wang, Shuo Jiang, Bin He

Organizations: Shanghai Research Institute for Intelligent Autonomous Systems, Tongji University, Shanghai 201210, China · College of Electronics and Information Engineering, Tongji University, Shanghai 201804, China

Abstract

Diffusion and flow matching visuomotor policies require iterative denoising or numerical integration for multimodal action generation, introducing latency and execution mismatch. We instead consider a different formulation: conditioned on vision and proprioception, demonstrated bimanual action chunks concentrate in a structured observation-dependent region of joint-action space, making the action target more direct to model than the generative process. Based on this view, we propose the Direct Manifold Action Transformer (DMAT), which reformulates action generation as the direct modeling of clean, observation-conditioned joint-space action chunks from stochastic interpolants, thereby enabling single-step inference without relying on iterative generative inference or auxiliary trajectory-straightening procedures. Within DMAT, local 3D geometry anchors predictions to spatial structure, while State-Consistent Spectral-Time Refinement (STR) reconciles predicted chunks with proprioception and suppresses high-frequency residuals. Across 567 standard and out-of-distribution (OOD) trials per method, DMAT outperformed Improved 3D Diffusion Policy (IDP3) and DiTFlow3D in aggregate success rate, achieving relative improvements of 129.1% and 95.7%, respectively, while generating action chunks 2.06x and 1.34x faster and remaining robust under repeated 50 N interventions. These results pave the way for a novel generative paradigm for efficient and robust bimanual manipulation.

Figures & tables

Explore similar work

CardsList
  1. JAMB: Joint Action-Motion Diffusion for Bimanual Manipulation

    Sep 21, 2026Chuyang Xiao, Peilin Meng, David Held3D Point TrackingBimanual Robotic Manipulation

  2. High-Fidelity One-Step Generative Visuomotor Policy via Recursive Correction, Frequency Consistency, and Contrastive Flow Matching

    Jul 4, 2026Yuran Chen, Xinye Cai, Zhonglin Gong +1Flow MatchingContrastive Flow Matching