Discovering Synergies for Robot Manipulation with Multi-Task Reinforcement Learning

10/04/2021
by   Zhanpeng He, et al.
0

Controlling robotic manipulators with high-dimensional action spaces for dexterous tasks is a challenging problem. Inspired by human manipulation, researchers have studied generating and using postural synergies for robot hands to accomplish manipulation tasks, leveraging the lower dimensional nature of synergistic action spaces. However, many of these works require pre-collected data from an existing controller in order to derive such a subspace by means of dimensionality reduction. In this paper, we present a framework that simultaneously discovers a synergy space and a multi-task policy that operates on this low-dimensional action space to accomplish diverse manipulation tasks. We demonstrate that our end-to-end method is able to perform multiple tasks using few synergies, and outperforms sequential methods that apply dimensionality reduction to independently collected data. We also show that deriving synergies using multiple tasks can lead to a subspace that enables robots to efficiently learn new manipulation tasks and interactions with new objects.

READ FULL TEXT

page 1

page 4

page 6

research
07/10/2017

Vision-Based Multi-Task Manipulation for Inexpensive Robots Using End-To-End Learning from Demonstration

In this paper, we propose a multi-task learning from demonstration metho...
research
11/28/2022

CLAS: Coordinating Multi-Robot Manipulation with Central Latent Action Spaces

Multi-robot manipulation tasks involve various control entities that can...
research
10/06/2020

Policy learning in SE(3) action spaces

In the spatial action representation, the action space spans the space o...
research
05/07/2020

Shared Autonomy with Learned Latent Actions

Assistive robots enable people with disabilities to conduct everyday tas...
research
03/19/2020

VisuoSpatial Foresight for Multi-Step, Multi-Task Fabric Manipulation

Robotic fabric manipulation has applications in cloth and cable manageme...
research
03/03/2022

Implicit Kinematic Policies: Unifying Joint and Cartesian Action Spaces in End-to-End Robot Learning

Action representation is an important yet often overlooked aspect in end...
research
02/12/2018

Intuitive Hand Teleoperation by Novice Operators Using a Continuous Teleoperation Subspace

Human-in-the-loop manipulation is useful in when autonomous grasping is ...

Please sign up or login with your details

Forgot password? Click here to reset