DexTransfer: Real World Multi-fingered Dexterous Grasping with Minimal Human Demonstrations

09/28/2022
by   Zoey Qiuyu Chen, et al.
4

Teaching a multi-fingered dexterous robot to grasp objects in the real world has been a challenging problem due to its high dimensional state and action space. We propose a robot-learning system that can take a small number of human demonstrations and learn to grasp unseen object poses given partially occluded observations. Our system leverages a small motion capture dataset and generates a large dataset with diverse and successful trajectories for a multi-fingered robot gripper. By adding domain randomization, we show that our dataset provides robust grasping trajectories that can be transferred to a policy learner. We train a dexterous grasping policy that takes the point clouds of the object as input and predicts continuous actions to grasp objects from different initial robot states. We evaluate the effectiveness of our system on a 22-DoF floating Allegro Hand in simulation and a 23-DoF Allegro robot hand with a KUKA arm in real world. The policy learned from our dataset can generalize well on unseen object poses in both simulation and the real world

READ FULL TEXT

page 1

page 5

page 11

page 12

research
10/24/2022

Learning Robust Real-World Dexterous Grasping Policies via Implicit Shape Augmentation

Dexterous robotic hands have the capability to interact with a wide vari...
research
02/15/2020

Learning Pregrasp Manipulation of Objects from Ungraspable Poses

In robotic grasping, objects are often occluded in ungraspable configura...
research
11/28/2020

AdaGrasp: Learning an Adaptive Gripper-Aware Grasping Policy

This paper aims to improve robots' versatility and adaptability by allow...
research
02/13/2023

Learning Tri-mode Grasping for Ambidextrous Robot Picking

Object picking in cluttered scenes is a widely investigated field of rob...
research
05/30/2022

Multi-Object Grasping – Types and Taxonomy

This paper proposes 12 multi-object grasps (MOGs) types from a human and...
research
08/28/2023

Uncertainty-driven Affordance Discovery for Efficient Robotics Manipulation

Robotics affordances, providing information about what actions can be ta...
research
03/04/2022

GraspARL: Dynamic Grasping via Adversarial Reinforcement Learning

Grasping moving objects, such as goods on a belt or living animals, is a...

Please sign up or login with your details

Forgot password? Click here to reset