On Policy Learning Robust to Irreversible Events: An Application to Robotic In-Hand Manipulation
- 31 January 2018
- journal article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 3 (3) , 1482-1489
Abstract
In this letter, we present an approach for learning in-hand manipulation skills with a low-cost, underactuated prosthetic hand in the presence of irreversible events. Our approach combines reinforcement learning based on visual perception with low-level reactive control based on tactile perception, which aims to avoid slipping. The objective of the reinforcement learning level consists not only in fulfilling the in-hand manipulation goal, but also in minimizing the intervention of the tactile reactive control. This way, the occurrence of object slipping during the learning procedure, which we consider an irreversible event, is significantly reduced. When an irreversible event occurs, the learning process is considered failed. We show the performance in two tasks, which consist in reorienting a cup and a bottle only using the fingers. The experimental results show that the proposed architecture allows reaching the goal in the Cartesian space and reduces significantly the occurrence of object slipping during the learning procedure. Moreover, without the proposed synergy between reactive control and reinforcement learning it was not possible to avoid irreversible events and, therefore, to learn the task.Keywords
All Related Versions
Funding Information
- Marie Sklodowska-Curie Individual Fellowship LEACON (659265)
- Technical University of Munich, International Graduate School of Science and Engineering and Helmholtz Association
This publication has 22 references indexed in Scilit:
- Enhance In-Hand Dexterous Micromanipulation by Exploiting Adhesion ForcesIEEE Transactions on Robotics, 2017
- Data-efficient control policy search using residual dynamics learning2017
- Learning dexterous manipulation for a soft robotic hand from human demonstrations2016
- Learning task-specific models for dexterous, in-hand manipulation with simple, adaptive robot hands2016
- Optimal control with learned local models: Application to dexterous manipulation2016
- Learning robot in-hand manipulation with tactile features2015
- Dexterous manipulation using both palm and fingers2014
- Planning in-hand object manipulation with multifingered hands considering task constraints2013
- On rolling contact motion by robotic fingers via prescribed performance control2013
- Associating and reshaping of whole body motions for object manipulation2009