Learning to Use Toes in a Humanoid Robot
- In this paper we show that a model-free approach to learn behaviors in joint space can be successfully used to utilize toes of a humanoid robot. Keeping the approach model-free makes it applicable to any kind of humanoid robot, or robot in general. Here we focus on the benefit on robots with toes which is otherwise more difficult to exploit. The task has been to learn different kick behaviors onIn this paper we show that a model-free approach to learn behaviors in joint space can be successfully used to utilize toes of a humanoid robot. Keeping the approach model-free makes it applicable to any kind of humanoid robot, or robot in general. Here we focus on the benefit on robots with toes which is otherwise more difficult to exploit. The task has been to learn different kick behaviors on simulated Nao robots with toes in the RoboCup 3D soccer simulator. As a result, the robot learned to step on its toe for a kick that performs 30% better than learning the same kick without toes.…
| Author: | Klaus DorerGND |
|---|---|
| Editor: | Hidehisa Akiyama, Oliver Obst, Claude Sammut, Flavio Tonidandel |
| Publisher: | Springer |
| Place of publication: | Heidelberg |
| Year of Publication: | 2017 |
| Pagenumber: | 12 |
| Language: | English |
| Tag: | Humanoid Robots; Machine Learning; Robotic Soccer |
| Parent Title (English): | RoboCup 2017: Robot World Cup XXI |
| First Page: | 100 |
| Last Page: | 111 |
| Document Type: | Conference Proceeding |
| Institutes: | Bibliografie |
| Acces Right: | Frei zugänglich |
| Release Date: | 2018/01/13 |
| Licence (German): |  Es gilt das UrhG |
| Note: | Konferenz: RoboCup2017, Nagoya, Japan(ロボカップ2017) |
| URL: | https://www.robocup2017.org/file/symposium/RoboCup_Symposium_2017_paper_6.pdf |