analiza aiqbench
This research introduces EAGG, a novel framework for cross-end-effector grasp generation that dynamically aligns robot morphology with task geometry using graph-based conditioning. By maintaining synchronization between evolving end-effector states and the generative model during sampling, it achieves high success rates across diverse robotic hardware without sacrificing transfer learning capabilities. The work represents significant progress in embodied AI by proving that explicit structural alignment yields better generalization than static descriptors for dexterous manipulation tasks.
EAGG: Embodiment-Aligned Grasp Generation via Geometry-Aware Graph Conditioning
arXiv:2606.18092v1 Announce Type: new Abstract: Cross-end-effector grasp generation seeks a unified model that generalizes across objects and across embodiments ranging from parallel grippers to dexterous end effectors. Existing grasp generators are typically designed for a fixed embodiment or encode embodiment identity with a static descriptor, which weakens transfer when topology, actuation coupling, and contact geometry differ substantially. We present EAGG, an embodiment-aligned grasp generator that represents each embodiment with a topology-aware end-effector graph and an embodiment-specific low-dimensional end-effector control space. A frozen end-effector-cognition backbone converts the current articulated state into geometry-aware tokens that act as a reusable morphology prior, and iterative geometry injection refreshes these tokens throughout sampling so that conditioning remains synchronized with the evolving end-effector geometry. On the MultiGripperGrasp benchmark, EAGG reaches 56.17% average success across six training end effectors, remaining within 1.10 percentage points of specialized training while preserving transfer to finetuning and zero-shot end effectors. Iterative geometry injection further reduces the pooled median contact distance from 0.239 cm to 0.189 cm. These results show that cross-end-effector grasp generation is strengthened by aligning embodiment structure inside a shared generator rather than suppressing embodiment differences. Code is available at https://github.com/wanhaoniu/EAGG.
Abstract:Cross-end-effector grasp generation seeks a unified model that generalizes across objects and across embodiments ranging from parallel grippers to dexterous end effectors. Existing grasp generators are typically designed for a fixed embodiment or encode embodiment identity with a static descriptor, which weakens transfer when topology, actuation coupling, and contact geometry differ substantially. We present EAGG, an embodiment-aligned grasp generator that represents each embodiment with a topology-aware end-effector graph and an embodiment-specific low-dimensional end-effector control space. A frozen end-effector-cognition backbone converts the current articulated state into geometry-aware tokens that act as a reusable morphology prior, and iterative geometry injection refreshes these tokens throughout sampling so that conditioning remains synchronized with the evolving end-effector geometry. On the MultiGripperGrasp benchmark, EAGG reaches 56.17% average success across six training end effectors, remaining within 1.10 percentage points of specialized training while preserving transfer to finetuning and zero-shot end effectors. Iterative geometry injection further reduces the pooled median contact distance from 0.239 cm to 0.189 cm. These results show that cross-end-effector grasp generation is strengthened by aligning embodiment structure inside a shared generator rather than suppressing embodiment differences. Code is available at this https URL.
16 pages, 8 figures. Code is available at this https URL
Robotics (cs.RO); Artificial Intelligence (cs.AI)
arXiv:2606.18092 [cs.RO]
(or arXiv:2606.18092v1 [cs.RO] for this version)
https://doi.org/10.48550/arXiv.2606.18092 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 296 cuvinte
Subiecte conexe
DEAL-Grasp: Decoupled Alignment Representation for Geometry-Aware Dexterous Grasp Generation
arXiv · 24 sept. 2026
Function beyond Form: Functional Correspondence for Cross-Embodiment Dexterous Grasp Generation
arXiv · 1 oct. 2026
PIRAMID: Progress and Plans
LessWrong · 27 iul. 2026
GEAR-VLA: Learning Geometry-Aware Action Representations for Generalizable Robotic Manipulation
arXiv · 11 iun. 2026