elib
DLR-Header
DLR-Logo -> http://www.dlr.de
DLR Portal Home | Imprint | Privacy Policy | Contact | Deutsch
Fontsize: [-] Text [+]

Dextrous Tactile In-Hand Manipulation Using a Modular Reinforcement Learning Architecture

Pitz, Johannes and Röstel, Lennart and Sievers, Leon and Bäuml, Berthold (2023) Dextrous Tactile In-Hand Manipulation Using a Modular Reinforcement Learning Architecture. In: 2023 IEEE International Conference on Robotics and Automation, ICRA 2023, pp. 1852-1858. IEEE. International Conference on Robotics and Automation, 2023-05-29 - 2023-06-03, London, UK. doi: 10.1109/ICRA48891.2023.10160756. ISBN 979-835032365-8. ISSN 1050-4729.

Full text not available from this repository.

Official URL: https://ieeexplore.ieee.org/document/10160756

Abstract

Dextrous in-hand manipulation with a multi-fingered robotic hand is a challenging task, esp. when performed with the hand oriented upside down, demanding permanent force-closure, and when no external sensors are used. For the task of reorienting an object to a given goal orientation (vs. infinitely spinning it around an axis), the lack of external sensors is an additional fundamental challenge as the state of the object has to be estimated all the time, e.g., to detect when the goal is reached. In this paper, we show that the task of reorienting a cube to any of the 24 possible goal orientations in a Pi/2-raster using the torque-controlled DLR-Hand II is possible. The task is learned in simulation using a modular deep reinforcement learning architecture: the actual policy has only a small observation time window of 0.5s but gets the cube state as an explicit input which is estimated via a deep differentiable particle filter trained on data generated by running the policy. In simulation, we reach a success rate of 92% while applying significant domain randomization. Via zero-shot Sim2Real-transfer on the real robotic system, all 24 goal orientations can be reached with a high success rate.

Item URL in elib:https://elib.dlr.de/195400/
Document Type:Conference or Workshop Item (Poster)
Additional Information:https://arxiv.org/abs/2303.04705
Title:Dextrous Tactile In-Hand Manipulation Using a Modular Reinforcement Learning Architecture
Authors:
AuthorsInstitution or Email of AuthorsAuthor's ORCID iDORCID Put Code
Pitz, JohannesUNSPECIFIEDhttps://orcid.org/0000-0002-2629-1892UNSPECIFIED
Röstel, LennartUNSPECIFIEDUNSPECIFIEDUNSPECIFIED
Sievers, LeonUNSPECIFIEDhttps://orcid.org/0000-0001-6430-4618UNSPECIFIED
Bäuml, BertholdUNSPECIFIEDhttps://orcid.org/0000-0002-4545-4765UNSPECIFIED
Date:June 2023
Journal or Publication Title:2023 IEEE International Conference on Robotics and Automation, ICRA 2023
Refereed publication:Yes
Open Access:No
Gold Open Access:No
In SCOPUS:Yes
In ISI Web of Science:No
DOI:10.1109/ICRA48891.2023.10160756
Page Range:pp. 1852-1858
Publisher:IEEE
ISSN:1050-4729
ISBN:979-835032365-8
Status:Published
Keywords:In-Hand Manipulation, Deep Reinforcement Learning
Event Title:International Conference on Robotics and Automation
Event Location:London, UK
Event Type:international Conference
Event Start Date:29 May 2023
Event End Date:3 June 2023
Organizer:IEEE
HGF - Research field:Aeronautics, Space and Transport
HGF - Program:Space
HGF - Program Themes:Robotics
DLR - Research area:Raumfahrt
DLR - Program:R RO - Robotics
DLR - Research theme (Project):R - Autonomous learning robots [RO]
Location: Oberpfaffenhofen
Institutes and Institutions:Institute of Robotics and Mechatronics (since 2013)
Deposited By: Pitz, Johannes
Deposited On:21 Sep 2023 10:32
Last Modified:24 Apr 2024 20:55

Repository Staff Only: item control page

Browse
Search
Help & Contact
Information
electronic library is running on EPrints 3.3.12
Website and database design: Copyright © German Aerospace Center (DLR). All rights reserved.