Learning Synergies between Pushing and Grasping with Self-supervised Deep Reinforcement Learning

03/27/2018
by   Andy Zeng, et al.
0

Skilled robotic manipulation benefits from complex synergies between non-prehensile (e.g. pushing) and prehensile (e.g. grasping) actions: pushing can help rearrange cluttered objects to make space for arms and fingers; likewise, grasping can help displace objects to make pushing movements more precise and collision-free. In this work, we demonstrate that it is possible to discover and learn these synergies from scratch through model-free deep reinforcement learning. Our method involves training two fully convolutional networks that map from visual observations to actions: one infers the utility of pushes for a dense pixel-wise sampling of end effector orientations and locations, while the other does the same for grasping. Both networks are trained jointly in a Q-learning framework and are entirely self-supervised by trial and error, where rewards are provided from successful grasps. In this way, our policy learns pushing motions that enable future grasps, while learning grasps that can leverage past pushes. During picking experiments in both simulation and real-world scenarios, we find that our system quickly learns complex behaviors amid challenging cases of clutter, and achieves better grasping success rates and picking efficiencies than baseline alternatives after only a few hours of training. We further demonstrate that our method is capable of generalizing to novel objects. Qualitative results (videos), code, pre-trained models, and simulation environments are available at http://vpg.cs.princeton.edu

READ FULL TEXT

page 1

page 3

page 5

research
07/25/2019

Robot Learning of Shifting Objects for Grasping in Cluttered Environments

Robotic grasping in cluttered environments is often infeasible due to ob...
research
07/06/2021

Fast-Learning Grasping and Pre-Grasping via Clutter Quantization and Q-map Masking

Grasping objects in cluttered scenarios is a challenging task in robotic...
research
10/03/2017

Robotic Pick-and-Place of Novel Objects in Clutter with Multi-Affordance Grasping and Cross-Domain Image Matching

This paper presents a robotic pick-and-place system that is capable of g...
research
07/19/2022

Self-Supervised Interactive Object Segmentation Through a Singulation-and-Grasping Approach

Instance segmentation with unseen objects is a challenging problem in un...
research
07/31/2023

Deep Reinforcement Learning of Dexterous Pre-grasp Manipulation for Human-like Functional Categorical Grasping

Many objects such as tools and household items can be used only if grasp...
research
03/08/2020

Online Self-Supervised Learning for Object Picking: Detecting Optimum Grasping Position using a Metric Learning Approach

Self-supervised learning methods are attractive candidates for automatic...
research
07/04/2021

Hierarchical Policies for Cluttered-Scene Grasping with Latent Plans

6D grasping in cluttered scenes is a longstanding robotic manipulation p...

Please sign up or login with your details

Forgot password? Click here to reset