Learning Dexterous Grasping from Sparse Taxonomy Guidance
This addresses the challenge of specifying grasp plans for dexterous manipulation in robotics, offering a more controllable and generalizable approach, though it is incremental in combining existing concepts.
The paper tackles the problem of dexterous grasping by proposing GRIT, a two-stage framework that learns from sparse taxonomy guidance, improving generalization to novel objects with an overall success rate of 87.9%.
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.