SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation

Akihiko Yamaguchi; Tsukasa Ogasawara

doi:10.1109/ICHR.2010.5686285

SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation

Akihiko Yamaguchi, Tsukasa Ogasawara

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

2 Citations (Scopus)

Abstract

This paper introduces a software library of reinforcement learning (RL) methods, named SkyAI. SkyAI is a highly modularized RL library for real/simulated robots to learn behaviors. Our ultimate goal is to develop an artificial intelligence (AI) program with which the robots can learn to behave as their users' wish. In this paper, we describe the concepts, the requirements, and the current implementation of SkyAI. SkyAI provides two conflicting features: high execution-speed enough for real robot systems and high flexibility to design learning systems. We also demonstrate the applications to crawling tasks of both a humanoid robot in simulation and a real spider robot.

Original language	English
Title of host publication	2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010
Pages	118-123
Number of pages	6
DOIs	https://doi.org/10.1109/ICHR.2010.5686285
Publication status	Published - 2010 Dec 1
Externally published	Yes
Event	2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010 - Nashville, TN, United States Duration: 2010 Dec 6 → 2010 Dec 8

Publication series

Name	2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010

Other

Other	2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010
Country	United States
City	Nashville, TN
Period	10/12/6 → 10/12/8

ASJC Scopus subject areas

Artificial Intelligence
Hardware and Architecture
Human-Computer Interaction

Access to Document

10.1109/ICHR.2010.5686285

Fingerprint Dive into the research topics of 'SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation'. Together they form a unique fingerprint.

Cite this

Yamaguchi, A., & Ogasawara, T. (2010). SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation. In 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010 (pp. 118-123). [5686285] (2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010). https://doi.org/10.1109/ICHR.2010.5686285

SkyAI : Highly modularized reinforcement learning library - Concepts, requirements, and implementation. / Yamaguchi, Akihiko; Ogasawara, Tsukasa.

2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010. 2010. p. 118-123 5686285 (2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Yamaguchi, A & Ogasawara, T 2010, SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation. in 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010., 5686285, 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010, pp. 118-123, 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010, Nashville, TN, United States, 10/12/6. https://doi.org/10.1109/ICHR.2010.5686285

@inproceedings{5313a368b75048c4a89fb203b6a7aa04,

title = "SkyAI: Highly modularized reinforcement learning library - Concepts, requirements, and implementation",

abstract = "This paper introduces a software library of reinforcement learning (RL) methods, named SkyAI. SkyAI is a highly modularized RL library for real/simulated robots to learn behaviors. Our ultimate goal is to develop an artificial intelligence (AI) program with which the robots can learn to behave as their users' wish. In this paper, we describe the concepts, the requirements, and the current implementation of SkyAI. SkyAI provides two conflicting features: high execution-speed enough for real robot systems and high flexibility to design learning systems. We also demonstrate the applications to crawling tasks of both a humanoid robot in simulation and a real spider robot.",

author = "Akihiko Yamaguchi and Tsukasa Ogasawara",

year = "2010",

month = dec,

day = "1",

doi = "10.1109/ICHR.2010.5686285",

language = "English",

isbn = "9781424486885",

series = "2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010",

pages = "118--123",

booktitle = "2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010",

note = "2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010 ; Conference date: 06-12-2010 Through 08-12-2010",

}

TY - GEN

T1 - SkyAI

T2 - 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010

AU - Yamaguchi, Akihiko

AU - Ogasawara, Tsukasa

PY - 2010/12/1

Y1 - 2010/12/1

N2 - This paper introduces a software library of reinforcement learning (RL) methods, named SkyAI. SkyAI is a highly modularized RL library for real/simulated robots to learn behaviors. Our ultimate goal is to develop an artificial intelligence (AI) program with which the robots can learn to behave as their users' wish. In this paper, we describe the concepts, the requirements, and the current implementation of SkyAI. SkyAI provides two conflicting features: high execution-speed enough for real robot systems and high flexibility to design learning systems. We also demonstrate the applications to crawling tasks of both a humanoid robot in simulation and a real spider robot.

AB - This paper introduces a software library of reinforcement learning (RL) methods, named SkyAI. SkyAI is a highly modularized RL library for real/simulated robots to learn behaviors. Our ultimate goal is to develop an artificial intelligence (AI) program with which the robots can learn to behave as their users' wish. In this paper, we describe the concepts, the requirements, and the current implementation of SkyAI. SkyAI provides two conflicting features: high execution-speed enough for real robot systems and high flexibility to design learning systems. We also demonstrate the applications to crawling tasks of both a humanoid robot in simulation and a real spider robot.

UR - http://www.scopus.com/inward/record.url?scp=79851475664&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79851475664&partnerID=8YFLogxK

U2 - 10.1109/ICHR.2010.5686285

DO - 10.1109/ICHR.2010.5686285

M3 - Conference contribution

AN - SCOPUS:79851475664

SN - 9781424486885

T3 - 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010

SP - 118

EP - 123

BT - 2010 10th IEEE-RAS International Conference on Humanoid Robots, Humanoids 2010

Y2 - 6 December 2010 through 8 December 2010

ER -