A Distributed Strategy for Gait Adaptation in Modular Robots

David Johan Christensen; Ulrik Pagh Schultz; Kasper Stoy

doi:10.1109/ROBOT.2010.5509942

A Distributed Strategy for Gait Adaptation in Modular Robots

David Johan Christensen, Ulrik Pagh Schultz, Kasper Stoy

Research output: Contribution to journal › Conference article › Research › peer-review

Abstract

In this paper we study online gait opti- mization for modular robots. The learning strategy we apply is distributed, independent on robot mor- phology, and easy to implement. First we demonstrate how the strategy allows an ATRON robot to adapt to faults and changes in its morphology and we study the strategy’s scalability. Second we extend the strategy to learn the parameters of gait-tables for ATRON and M-TRAN robots.We conclude that the presented strategy is effective for online learning of gaits for most types of modular robots and that learning can effectively be distributed by having independent pro- cesses learning in parallel.

Original language	English
Journal	I E E E International Conference on Robotics and Automation. Proceedings
Pages (from-to)	2765-2770
ISSN	1050-4729
DOIs	https://doi.org/10.1109/ROBOT.2010.5509942
Publication status	Published - 2010
Externally published	Yes
Event	2010 IEEE International Conference on Robotics and Automation - Anchorage, AK, United States Duration: 3 May 2010 → 8 May 2010 http://icra2010.grasp.upenn.edu/

Conference

Conference	2010 IEEE International Conference on Robotics and Automation
Country	United States
City	Anchorage, AK
Period	03/05/2010 → 08/05/2010
Internet address	http://icra2010.grasp.upenn.edu/

Access to Document

10.1109/ROBOT.2010.5509942

Cite this

@inproceedings{99382f8061bf4d5fb8c0f27401636f95,

title = "A Distributed Strategy for Gait Adaptation in Modular Robots",

abstract = "In this paper we study online gait opti- mization for modular robots. The learning strategy we apply is distributed, independent on robot mor- phology, and easy to implement. First we demonstrate how the strategy allows an ATRON robot to adapt to faults and changes in its morphology and we study the strategy{\textquoteright}s scalability. Second we extend the strategy to learn the parameters of gait-tables for ATRON and M-TRAN robots.We conclude that the presented strategy is effective for online learning of gaits for most types of modular robots and that learning can effectively be distributed by having independent pro- cesses learning in parallel.",

author = "Christensen, {David Johan} and Schultz, {Ulrik Pagh} and Kasper Stoy",

year = "2010",

doi = "10.1109/ROBOT.2010.5509942",

language = "English",

pages = "2765--2770",

journal = "I E E E International Conference on Robotics and Automation",

issn = "2152-4092",

publisher = "IEEE",

note = "2010 IEEE International Conference on Robotics and Automation, ICRA 2010 ; Conference date: 03-05-2010 Through 08-05-2010",

url = "http://icra2010.grasp.upenn.edu/",

}

TY - GEN

T1 - A Distributed Strategy for Gait Adaptation in Modular Robots

AU - Christensen, David Johan

AU - Schultz, Ulrik Pagh

AU - Stoy, Kasper

PY - 2010

Y1 - 2010

N2 - In this paper we study online gait opti- mization for modular robots. The learning strategy we apply is distributed, independent on robot mor- phology, and easy to implement. First we demonstrate how the strategy allows an ATRON robot to adapt to faults and changes in its morphology and we study the strategy’s scalability. Second we extend the strategy to learn the parameters of gait-tables for ATRON and M-TRAN robots.We conclude that the presented strategy is effective for online learning of gaits for most types of modular robots and that learning can effectively be distributed by having independent pro- cesses learning in parallel.

AB - In this paper we study online gait opti- mization for modular robots. The learning strategy we apply is distributed, independent on robot mor- phology, and easy to implement. First we demonstrate how the strategy allows an ATRON robot to adapt to faults and changes in its morphology and we study the strategy’s scalability. Second we extend the strategy to learn the parameters of gait-tables for ATRON and M-TRAN robots.We conclude that the presented strategy is effective for online learning of gaits for most types of modular robots and that learning can effectively be distributed by having independent pro- cesses learning in parallel.

U2 - 10.1109/ROBOT.2010.5509942

DO - 10.1109/ROBOT.2010.5509942

M3 - Conference article

SP - 2765

EP - 2770

JO - I E E E International Conference on Robotics and Automation

JF - I E E E International Conference on Robotics and Automation

SN - 2152-4092

T2 - 2010 IEEE International Conference on Robotics and Automation

Y2 - 3 May 2010 through 8 May 2010

ER -

A Distributed Strategy for Gait Adaptation in Modular Robots

Abstract

Conference

Access to Document

Fingerprint

Cite this