LSTM Learning of Inverse Dynamics with Contact in Various Environments

Daichi Furuta; Kyo Kutsuzawa; Sho Sakaino; Toshiaki Tsuji

doi:10.1109/MECATRONICS.2018.8495698

LSTM Learning of Inverse Dynamics with Contact in Various Environments

Daichi Furuta, Kyo Kutsuzawa, Sho Sakaino, Toshiaki Tsuji

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

1 Citation (Scopus)

Abstract

A machine learning method has been introduced to solve the problem of inverse dynamics with contact in various environments. Conventional methods need multiple contact models to switch according to situations, while such methods have a difficulty in dealing with different environments. We propose a machine learning method that can handle various environments with a single learning model. We use long short-term memory as a learning model with high expression ability. From the verification, the proposed method showed higher performance than Gaussian processes. In addition, the performance of the model was improved by using training data collected under various environmental conditions.

Original language	English
Title of host publication	Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	149-154
Number of pages	6
ISBN (Electronic)	9781538629826
DOIs	https://doi.org/10.1109/MECATRONICS.2018.8495698
Publication status	Published - 2018 Oct 17
Externally published	Yes
Event	12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018 - Tsu, Japan Duration: 2018 Sep 10 → 2018 Sep 12

Publication series

Name	Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018

Conference

Conference	12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018
Country	Japan
City	Tsu
Period	18/9/10 → 18/9/12

ASJC Scopus subject areas

Human-Computer Interaction
Signal Processing
Mechanical Engineering
Artificial Intelligence
Modelling and Simulation
Orthopedics and Sports Medicine
Control and Optimization
Rehabilitation

UN SDGs

This output contributes to the following Sustainable Development Goal(s)

Access to Document

10.1109/MECATRONICS.2018.8495698

Fingerprint Dive into the research topics of 'LSTM Learning of Inverse Dynamics with Contact in Various Environments'. Together they form a unique fingerprint.

Cite this

Furuta, D., Kutsuzawa, K., Sakaino, S., & Tsuji, T. (2018). LSTM Learning of Inverse Dynamics with Contact in Various Environments. In Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018 (pp. 149-154). [8495698] (Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/MECATRONICS.2018.8495698

LSTM Learning of Inverse Dynamics with Contact in Various Environments. / Furuta, Daichi; Kutsuzawa, Kyo; Sakaino, Sho; Tsuji, Toshiaki.

Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018. Institute of Electrical and Electronics Engineers Inc., 2018. p. 149-154 8495698 (Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018).

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

Furuta, D, Kutsuzawa, K, Sakaino, S & Tsuji, T 2018, LSTM Learning of Inverse Dynamics with Contact in Various Environments. in Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018., 8495698, Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018, Institute of Electrical and Electronics Engineers Inc., pp. 149-154, 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018, Tsu, Japan, 18/9/10. https://doi.org/10.1109/MECATRONICS.2018.8495698

Furuta D, Kutsuzawa K, Sakaino S, Tsuji T. LSTM Learning of Inverse Dynamics with Contact in Various Environments. In Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018. Institute of Electrical and Electronics Engineers Inc. 2018. p. 149-154. 8495698. (Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018). https://doi.org/10.1109/MECATRONICS.2018.8495698

Furuta, Daichi ; Kutsuzawa, Kyo ; Sakaino, Sho ; Tsuji, Toshiaki. / LSTM Learning of Inverse Dynamics with Contact in Various Environments. Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018. Institute of Electrical and Electronics Engineers Inc., 2018. pp. 149-154 (Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018).

@inproceedings{82cd0e86d9c140a99a3bbd3c8ce941e8,

title = "LSTM Learning of Inverse Dynamics with Contact in Various Environments",

abstract = "A machine learning method has been introduced to solve the problem of inverse dynamics with contact in various environments. Conventional methods need multiple contact models to switch according to situations, while such methods have a difficulty in dealing with different environments. We propose a machine learning method that can handle various environments with a single learning model. We use long short-term memory as a learning model with high expression ability. From the verification, the proposed method showed higher performance than Gaussian processes. In addition, the performance of the model was improved by using training data collected under various environmental conditions.",

author = "Daichi Furuta and Kyo Kutsuzawa and Sho Sakaino and Toshiaki Tsuji",

year = "2018",

month = oct,

day = "17",

doi = "10.1109/MECATRONICS.2018.8495698",

language = "English",

series = "Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "149--154",

booktitle = "Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018",

note = "12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018 ; Conference date: 10-09-2018 Through 12-09-2018",

}

TY - GEN

T1 - LSTM Learning of Inverse Dynamics with Contact in Various Environments

AU - Furuta, Daichi

AU - Kutsuzawa, Kyo

AU - Sakaino, Sho

AU - Tsuji, Toshiaki

PY - 2018/10/17

Y1 - 2018/10/17

N2 - A machine learning method has been introduced to solve the problem of inverse dynamics with contact in various environments. Conventional methods need multiple contact models to switch according to situations, while such methods have a difficulty in dealing with different environments. We propose a machine learning method that can handle various environments with a single learning model. We use long short-term memory as a learning model with high expression ability. From the verification, the proposed method showed higher performance than Gaussian processes. In addition, the performance of the model was improved by using training data collected under various environmental conditions.

AB - A machine learning method has been introduced to solve the problem of inverse dynamics with contact in various environments. Conventional methods need multiple contact models to switch according to situations, while such methods have a difficulty in dealing with different environments. We propose a machine learning method that can handle various environments with a single learning model. We use long short-term memory as a learning model with high expression ability. From the verification, the proposed method showed higher performance than Gaussian processes. In addition, the performance of the model was improved by using training data collected under various environmental conditions.

UR - http://www.scopus.com/inward/record.url?scp=85057186097&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85057186097&partnerID=8YFLogxK

U2 - 10.1109/MECATRONICS.2018.8495698

DO - 10.1109/MECATRONICS.2018.8495698

M3 - Conference contribution

AN - SCOPUS:85057186097

T3 - Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018

SP - 149

EP - 154

BT - Proceedings - 2018 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 12th France-Japan and 10th Europe-Asia Congress on Mechatronics, Mecatronics 2018

Y2 - 10 September 2018 through 12 September 2018

ER -

LSTM Learning of Inverse Dynamics with Contact in Various Environments

Abstract

Publication series

Conference

ASJC Scopus subject areas

UN SDGs

Access to Document

Other files and links

Cite this