Human-robot collaboration by intention recognition using deep LSTM neural network

Liang Yan; Xiaoshan Gao; Xiongjie Zhang; Suokui Chang

doi:10.1109/FPM45753.2019.9035907

Human-robot collaboration by intention recognition using deep LSTM neural network

Liang Yan
, Xiaoshan Gao
, Xiongjie Zhang
, Suokui Chang

School of Automation Science and Electrical Engineering

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

25 Scopus citations

Abstract

When a robot is required to perform specific tasks in human-robot-collaboration scenario, it is necessary for the robot to recognize human intention to more effectively and efficiently assist and interact with human. The relationship of the skeleton-based sequence of human motion provides a possible solution for the robot to recognize human intention. In this paper, we present a deep long short term memory (LSTM) neural network to recognize human intention. Considering the characteristics of deep LSTM neural network, it is a typically multiple stacked LSTM model, which combine the advantages of single LSTM layer and overcome the weakness of learning long-range time dependencies for RNN. The experimental results showed that the deep LSTM network with 2-layers have better prediction performance even only 40% of motion sequences is utilized.

Original language	English
Title of host publication	Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1390-1396
Number of pages	7
ISBN (Electronic)	9781728103112
DOIs	https://doi.org/10.1109/FPM45753.2019.9035907
State	Published - Apr 2019
Event	8th IEEE International Conference on Fluid Power and Mechatronics, FPM 2019 - Wuhan, China Duration: 10 Apr 2019 → 13 Apr 2019

Publication series

Name	Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019

Conference

Conference	8th IEEE International Conference on Fluid Power and Mechatronics, FPM 2019
Country/Territory	China
City	Wuhan
Period	10/04/19 → 13/04/19

Keywords

Deep lSTM
Human intention recognition
Recurrent neural networks

Access to Document

10.1109/FPM45753.2019.9035907

Cite this

Yan, L., Gao, X., Zhang, X., & Chang, S. (2019). Human-robot collaboration by intention recognition using deep LSTM neural network. In Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019 (pp. 1390-1396). Article 9035907 (Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/FPM45753.2019.9035907

Yan, Liang ; Gao, Xiaoshan ; Zhang, Xiongjie et al. / Human-robot collaboration by intention recognition using deep LSTM neural network. Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019. Institute of Electrical and Electronics Engineers Inc., 2019. pp. 1390-1396 (Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019).

@inproceedings{142d4ff5f1eb46f3a554bc3f230ccfbc,

title = "Human-robot collaboration by intention recognition using deep LSTM neural network",

abstract = "When a robot is required to perform specific tasks in human-robot-collaboration scenario, it is necessary for the robot to recognize human intention to more effectively and efficiently assist and interact with human. The relationship of the skeleton-based sequence of human motion provides a possible solution for the robot to recognize human intention. In this paper, we present a deep long short term memory (LSTM) neural network to recognize human intention. Considering the characteristics of deep LSTM neural network, it is a typically multiple stacked LSTM model, which combine the advantages of single LSTM layer and overcome the weakness of learning long-range time dependencies for RNN. The experimental results showed that the deep LSTM network with 2-layers have better prediction performance even only 40\% of motion sequences is utilized.",

keywords = "Deep lSTM, Human intention recognition, Recurrent neural networks",

author = "Liang Yan and Xiaoshan Gao and Xiongjie Zhang and Suokui Chang",

note = "Publisher Copyright: {\textcopyright} 2019 IEEE.; 8th IEEE International Conference on Fluid Power and Mechatronics, FPM 2019 ; Conference date: 10-04-2019 Through 13-04-2019",

year = "2019",

month = apr,

doi = "10.1109/FPM45753.2019.9035907",

language = "英语",

series = "Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019",

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

pages = "1390--1396",

booktitle = "Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019",

address = "美国",

}

Yan, L , Gao, X, Zhang, X & Chang, S 2019, Human-robot collaboration by intention recognition using deep LSTM neural network. in Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019., 9035907, Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019, Institute of Electrical and Electronics Engineers Inc., pp. 1390-1396, 8th IEEE International Conference on Fluid Power and Mechatronics, FPM 2019, Wuhan, China, 10/04/19. https://doi.org/10.1109/FPM45753.2019.9035907

Human-robot collaboration by intention recognition using deep LSTM neural network. / Yan, Liang ; Gao, Xiaoshan; Zhang, Xiongjie et al.
Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019. Institute of Electrical and Electronics Engineers Inc., 2019. p. 1390-1396 9035907 (Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019).

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

TY - GEN

T1 - Human-robot collaboration by intention recognition using deep LSTM neural network

AU - Yan, Liang

AU - Gao, Xiaoshan

AU - Zhang, Xiongjie

AU - Chang, Suokui

PY - 2019/4

Y1 - 2019/4

N2 - When a robot is required to perform specific tasks in human-robot-collaboration scenario, it is necessary for the robot to recognize human intention to more effectively and efficiently assist and interact with human. The relationship of the skeleton-based sequence of human motion provides a possible solution for the robot to recognize human intention. In this paper, we present a deep long short term memory (LSTM) neural network to recognize human intention. Considering the characteristics of deep LSTM neural network, it is a typically multiple stacked LSTM model, which combine the advantages of single LSTM layer and overcome the weakness of learning long-range time dependencies for RNN. The experimental results showed that the deep LSTM network with 2-layers have better prediction performance even only 40% of motion sequences is utilized.

AB - When a robot is required to perform specific tasks in human-robot-collaboration scenario, it is necessary for the robot to recognize human intention to more effectively and efficiently assist and interact with human. The relationship of the skeleton-based sequence of human motion provides a possible solution for the robot to recognize human intention. In this paper, we present a deep long short term memory (LSTM) neural network to recognize human intention. Considering the characteristics of deep LSTM neural network, it is a typically multiple stacked LSTM model, which combine the advantages of single LSTM layer and overcome the weakness of learning long-range time dependencies for RNN. The experimental results showed that the deep LSTM network with 2-layers have better prediction performance even only 40% of motion sequences is utilized.

KW - Deep lSTM

KW - Human intention recognition

KW - Recurrent neural networks

UR - https://www.scopus.com/pages/publications/85083024929

U2 - 10.1109/FPM45753.2019.9035907

DO - 10.1109/FPM45753.2019.9035907

M3 - 会议稿件

AN - SCOPUS:85083024929

T3 - Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019

SP - 1390

EP - 1396

BT - Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019

PB - Institute of Electrical and Electronics Engineers Inc.

T2 - 8th IEEE International Conference on Fluid Power and Mechatronics, FPM 2019

Y2 - 10 April 2019 through 13 April 2019

ER -

Yan L , Gao X, Zhang X, Chang S. Human-robot collaboration by intention recognition using deep LSTM neural network. In Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019. Institute of Electrical and Electronics Engineers Inc. 2019. p. 1390-1396. 9035907. (Proceedings of the 8th International Conference on Fluid Power and Mechatronics, FPM 2019). doi: 10.1109/FPM45753.2019.9035907

Human-robot collaboration by intention recognition using deep LSTM neural network

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this