Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion
Movement generation that is consistent with observed or demonstrated behaviour is an efficient way to seed movement planning in complex, high-dimensional movement systems like humanoid robots. We present a method for learning potential-based policies from constrained motion data. In contrast to prev...
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Wiley
2008-01-01
|
| Series: | Applied Bionics and Biomechanics |
| Online Access: | http://dx.doi.org/10.1080/11762320902789830 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1832555175089274880 |
|---|---|
| author | Matthew Howard Stefan Klanke Michael Gienger Christian Goerick Sethu Vijayakumar |
| author_facet | Matthew Howard Stefan Klanke Michael Gienger Christian Goerick Sethu Vijayakumar |
| author_sort | Matthew Howard |
| collection | DOAJ |
| description | Movement generation that is consistent with observed or demonstrated behaviour is an efficient way to seed movement planning in complex, high-dimensional movement systems like humanoid robots. We present a method for learning potential-based policies from constrained motion data. In contrast to previous approaches to direct policy learning, our method can combine observations from a variety of contexts where different constraints are in force, to learn the underlying unconstrained policy in form of its potential function. This allows us to generalise and predict behaviour where novel constraints apply. We demonstrate our approach on systems of varying complexity, including kinematic data from the ASIMO humanoid robot with 22 degrees of freedom. |
| format | Article |
| id | doaj-art-3b9d6d8285fa43caa1ea94e463ea6ae1 |
| institution | Kabale University |
| issn | 1176-2322 1754-2103 |
| language | English |
| publishDate | 2008-01-01 |
| publisher | Wiley |
| record_format | Article |
| series | Applied Bionics and Biomechanics |
| spelling | doaj-art-3b9d6d8285fa43caa1ea94e463ea6ae12025-02-03T05:49:27ZengWileyApplied Bionics and Biomechanics1176-23221754-21032008-01-015419521110.1080/11762320902789830Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained MotionMatthew Howard0Stefan Klanke1Michael Gienger2Christian Goerick3Sethu Vijayakumar4School of Informatics, University of Edinburgh, Edinburgh EH9 3JZ, UKSchool of Informatics, University of Edinburgh, Edinburgh EH9 3JZ, UKHonda Research Institute Europe GmbH, Offenbach/Main D-63073, GermanyHonda Research Institute Europe GmbH, Offenbach/Main D-63073, GermanySchool of Informatics, University of Edinburgh, Edinburgh EH9 3JZ, UKMovement generation that is consistent with observed or demonstrated behaviour is an efficient way to seed movement planning in complex, high-dimensional movement systems like humanoid robots. We present a method for learning potential-based policies from constrained motion data. In contrast to previous approaches to direct policy learning, our method can combine observations from a variety of contexts where different constraints are in force, to learn the underlying unconstrained policy in form of its potential function. This allows us to generalise and predict behaviour where novel constraints apply. We demonstrate our approach on systems of varying complexity, including kinematic data from the ASIMO humanoid robot with 22 degrees of freedom.http://dx.doi.org/10.1080/11762320902789830 |
| spellingShingle | Matthew Howard Stefan Klanke Michael Gienger Christian Goerick Sethu Vijayakumar Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion Applied Bionics and Biomechanics |
| title | Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion |
| title_full | Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion |
| title_fullStr | Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion |
| title_full_unstemmed | Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion |
| title_short | Behaviour Generation in Humanoids by Learning Potential-Based Policies from Constrained Motion |
| title_sort | behaviour generation in humanoids by learning potential based policies from constrained motion |
| url | http://dx.doi.org/10.1080/11762320902789830 |
| work_keys_str_mv | AT matthewhoward behaviourgenerationinhumanoidsbylearningpotentialbasedpoliciesfromconstrainedmotion AT stefanklanke behaviourgenerationinhumanoidsbylearningpotentialbasedpoliciesfromconstrainedmotion AT michaelgienger behaviourgenerationinhumanoidsbylearningpotentialbasedpoliciesfromconstrainedmotion AT christiangoerick behaviourgenerationinhumanoidsbylearningpotentialbasedpoliciesfromconstrainedmotion AT sethuvijayakumar behaviourgenerationinhumanoidsbylearningpotentialbasedpoliciesfromconstrainedmotion |