Learning Dexterous Grasping from Sparse Taxonomy Guidance

TL;DR

GRIT is a two-stage framework that learns dexterous grasping from sparse taxonomy guidance, improving generalization and controllability in robotic manipulation tasks.

Contribution

It introduces a novel approach combining taxonomy-based grasp specification with continuous control, enhancing flexibility and success rates in dexterous manipulation.

Findings

Achieves an 87.9% success rate in grasping tasks.

Demonstrates improved generalization to novel objects.

Enables high-level control adjustments in real-world experiments.

Abstract

Dexterous manipulation requires planning a grasp configuration suited to the object and task, which is then executed through coordinated multi-finger control. However, specifying grasp plans with dense pose or contact targets for every object and task is impractical. Meanwhile, end-to-end reinforcement learning from task rewards alone lacks controllability, making it difficult for users to intervene when failures occur. To this end, we present GRIT, a two-stage framework that learns dexterous control from sparse taxonomy guidance. GRIT first predicts a taxonomy-based grasp specification from the scene and task context. Conditioned on this sparse command, a policy generates continuous finger motions that accomplish the task while preserving the intended grasp structure. Our result shows that certain grasp taxonomies are more effective for specific object geometries. By leveraging this relationship, GRIT improves generalization to novel objects over baselines and achieves an overall success rate of 87.9%. Moreover, real-world experiments demonstrate controllability, enabling grasp strategies to be adjusted through high-level taxonomy selection based on object geometry and task intent.