Sample Efficient Grasp Learning Using Equivariant Models

2022/02/18 by Zhu, Xupeng, Wang, Dian, Biza, Ondrej +3 · 7 citations
#FOS: Computer and information sciences #Robotics (cs.RO)

paper · doi:10.48550/arxiv.2202.09468

Abstract

In planar grasp detection, the goal is to learn a function from an image of a scene onto a set of feasible grasp poses in SE(2). In this paper, we recognize that the optimal grasp function is SE(2)-equivariant and can be modeled using an equivariant convolutional neural network. As a result, we are able to significantly improve the sample efficiency of grasp learning, obtaining a good approximation of the grasp function after only 600 grasp attempts. This is few enough that we can learn to grasp completely on a physical robot in about 1.5 hours.

Cited by

• EfficientFlow: Efficient Equivariant Flow Policy Learning for Embodied AI
• Recurrent Equivariant Constraint Modulation: Learning Per-Layer Symmetry Relaxation from Data
• Residual Rotation Correction using Tactile Equivariance
• Generalizable Hierarchical Skill Learning via Object-Centric Representation
• VBM-NET: Visual Base Pose Learning for Mobile Manipulation using Equivariant TransporterNet and GNNs
• Multi-Group Equivariant Augmentation for Reinforcement Learning in Robot Manipulation
• Constraint-Preserving Data Generation for Visuomotor Policy Learning

Related

• On Robot Grasp Learning Using Equivariant Models
• Real-Time Grasp Detection Using Convolutional Neural Networks
• OrbitGrasp: SE(3)-Equivariant Grasp Learning
• Learning to Grasp from a Single Demonstration
• Grasp Learning by Sampling from Demonstration