BiDexGrasp: Coordinated Bimanual Dexterous Grasps across Object Geometries and Sizes

2026-04-07

AI Summary

This work addresses the challenge of generalizing bimanual robotic dexterous grasping to objects with diverse geometries and sizes, hindered by the lack of large-scale datasets and efficient grasp generation models. The authors introduce a large-scale bimanual grasping dataset comprising 6,351 objects and 9.7 million annotated grasps, along with a novel grasp generation framework that integrates coordination awareness and geometric size adaptability. The framework employs a two-stage synthesis strategy featuring region-based grasp initialization, decoupled force-closure optimization, bimanual coordination modeling, and a size-adaptive generation network to efficiently produce physically feasible grasps. Experimental results demonstrate that the proposed method significantly outperforms existing approaches in both simulation and real-world settings, enabling high-quality grasping of previously unseen objects.

Abstract

Bimanual dexterous grasping is a fundamental and promising area in robotics, yet its progress is constrained by the lack of comprehensive datasets and powerful generation models. In this work, we propose BiDexGrasp, consists of a large-scale bimanual dexterous grasp dataset and a novel generation model. For dataset, we propose a novel bimanual grasp synthesis pipeline to efficiently annotate physically feasible data for dataset construction. This pipeline addresses the challenges of high-dimensional bimanual grasping through a two-stage synthesis strategy of efficient region-based grasp initialization and decoupled force-closure grasp optimization. Powered by this pipeline, we construct a large-scale bimanual dexterous grasp dataset, comprising 6351 diverse objects with sizes ranging from 30 to 80 cm, along with 9.7 million annotated grasp data. Based on this dataset, we further introduce a bimanual-coordinated and geometry-size-adaptive dexterous grasping generation framework. The framework lies in two key designs: a bimanual coordination module and a geometry-size-adaptive grasp generation strategy to generate coordinated and high-quality grasps on unseen objects. Extensive experiments conducted in both simulation and real world demonstrate the superior performance of our proposed data synthesis pipeline and learned generative framework.

Problem

• bimanual grasping
• dexterous manipulation
• grasp dataset
• object geometry
• size variation

Innovation

• bimanual dexterous grasping
• grasp synthesis pipeline
• force-closure optimization
• geometry-size-adaptive generation
• coordinated bimanual manipulation