oru.sePublications
Change search
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Safe-To-Explore State Spaces: Ensuring Safe Exploration in Policy Search with Hierarchical Task Optimization
Intelligent Robotics Group, Aalto University, Helsinki, Finland.
Royal Institute of Technology, Stockholm, Sweden.
Örebro University, School of Science and Technology. (Centre for Applied Autonomous Sensor Systems (AASS))ORCID iD: 0000-0003-4026-7490
Örebro University, School of Science and Technology. (Centre for Applied Autonomous Sensor Systems (AASS))ORCID iD: 0000-0002-6013-4874
Show others and affiliations
2018 (English)In: IEEE-RAS Conference on Humanoid Robots / [ed] Asfour, T, IEEE, 2018, p. 132-138Conference paper, Published paper (Refereed)
Abstract [en]

Policy search reinforcement learning allows robots to acquire skills by themselves. However, the learning procedure is inherently unsafe as the robot has no a-priori way to predict the consequences of the exploratory actions it takes. Therefore, exploration can lead to collisions with the potential to harm the robot and/or the environment. In this work we address the safety aspect by constraining the exploration to happen in safe-to-explore state spaces. These are formed by decomposing target skills (e.g., grasping) into higher ranked sub-tasks (e.g., collision avoidance, joint limit avoidance) and lower ranked movement tasks (e.g., reaching). Sub-tasks are defined as concurrent controllers (policies) in different operational spaces together with associated Jacobians representing their joint-space mapping. Safety is ensured by only learning policies corresponding to lower ranked sub-tasks in the redundant null space of higher ranked ones. As a side benefit, learning in sub-manifolds of the state-space also facilitates sample efficiency. Reaching skills performed in simulation and grasping skills performed on a real robot validate the usefulness of the proposed approach.

Place, publisher, year, edition, pages
IEEE, 2018. p. 132-138
Series
IEEE-RAS International Conference on Humanoid Robots, ISSN 2164-0572
Keywords [en]
Sensorimotor learning, Grasping and Manipulation, Concept and strategy learning
National Category
Computer Sciences Computer Vision and Robotics (Autonomous Systems)
Identifiers
URN: urn:nbn:se:oru:diva-71311ISI: 000458689700019OAI: oai:DiVA.org:oru-71311DiVA, id: diva2:1277232
Conference
IEEE-RAS 18th Conference on Humanoid Robots (Humanoids 2018), Beijing, China, November 6-9, 2018
Funder
Swedish Foundation for Strategic Research
Note

Funding Agency:

Academy of Finland  314180

Available from: 2019-01-09 Created: 2019-01-09 Last updated: 2019-03-01Bibliographically approved

Open Access in DiVA

No full text in DiVA

Authority records BETA

Schaffernicht, ErikStoyanov, Todor

Search in DiVA

By author/editor
Schaffernicht, ErikStoyanov, Todor
By organisation
School of Science and Technology
Computer SciencesComputer Vision and Robotics (Autonomous Systems)

Search outside of DiVA

GoogleGoogle Scholar

urn-nbn

Altmetric score

urn-nbn
Total: 86 hits
CiteExportLink to record
Permanent link

Direct link
Cite
Citation style
  • apa
  • harvard1
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf