Method for Generating Cyclic Gait of Biped Robots Based on LSTM Neural Network

YU Jianjun; LI Chen; ZUO Guoyu; RUAN Xiaogang; WANG Yang

doi:10.11936/bjutxb2019060015

Journal of Beijing University of Technology > 2020 > 46(12): 1335-1344. > DOI: 10.11936/bjutxb2019060015

YU Jianjun, LI Chen, ZUO Guoyu, RUAN Xiaogang, WANG Yang. Method for Generating Cyclic Gait of Biped Robots Based on LSTM Neural Network[J]. Journal of Beijing University of Technology, 2020, 46(12): 1335-1344. DOI: 10.11936/bjutxb2019060015

Citation:

PDF (4495 KB)

Method for Generating Cyclic Gait of Biped Robots Based on LSTM Neural Network

Faculty of Information Technology, Beijing University of Technology, Beijing 100124, China

More Information

Received Date: June 25, 2019
Available Online: August 03, 2022
Published Date: December 09, 2020

Graphical Abstract

Abstract

Abstract

To solve the problem that the Kinect field of view cannot simulate the human body walking for a long time, a method based on the long short-term memory (LSTM) neural network prediction model to generating the cyclic gait of the humanoid robot was proposed. The one-dimensional time series of each joint angle was collected by Kinect when the human body walked, and the joint angle driving sequence of the humanoid robot was obtained through the humanoid gait balance model. The C-C method was used to determine the time delay and embedding dimension of the time series, and the phase space reconstruction of the joint angle sequence was used to obtain more eigenvalues of the time series. The joint angle prediction model based on the LSTM neural network was trained and passed by generating a sequence of joint angles for multiple gait cycles. Finally, the generated sequence was used to drive the NAO on WEBOTS platform to complete the walking motion of multiple gait cycles. This method effectively solves the problem of the visual field limitation of the somatosensory camera, enables the humanoid robot to complete the walking simulation of multiple gait cycles.
- LSTM neural network,
- biped robot,
- human walking motion,
- cyclic gait,
- phase space reconstruction,
- sequence of joint angle

FullText(HTML)

References (19)

References

[1]	ZHANG Q, CHAI H, LI Y B, et al. Kinematics analysis of biped robot based on toddler's natural gait extraction[C]//International Conference on Advanced Robotics and Mechatronics. Piscataway: IEEE, 2017: 479-485.
[2]	BULAT G, IGOR D, ILYA A. et al. Toward human-like biped robot gait: biomechanical analysis of human locomotion recorded by Kinect-based Motion Capture system[C]//International Symposium Mechatronics and its Applications. Piscataway: IEEE, 2015: 1-6.
[3]	ILBRHIM A. S, SAMY F M A. Sliding mode control of underactuated five-link biped robot for climbing stairs based on real human data[C]//2017 IEEE International Conference on Industrial Technology. Piscataway, N J: IEEE, 2017: 878-883.
[4]	左国玉, 王鑫鹏, 刘旭.基于混合粒子的仿袋鼠机器人站立平衡控制[J].北京工业大学学报, 2017, 43(7):1009-1016 http://www.cqvip.com/QK/95054X/20177/672563797.html ZUO G Y, WANG X P, LIU X. Optimized balance control for bionic kangaroo robot during stance phase using hybrid particle swarm optimization[J]. Journal of Beijing University of Technology, 2017, 43(7):1009-1016. (in Chinese) http://www.cqvip.com/QK/95054X/20177/672563797.html
[5]	IBRAHIM A S, SAMY F M A, ABDELFATATAH M M. Cyclic gait planning and control of underactuated five-link biped robot during single support and impact phases for normal walking[C]//International Conference on Industrial Technology. Piscataway: IEEE, 2018: 123-128.
[6]	KAO-SHING H, WEI-CHENG J, QIAO-HAN Y. Posture imitation and balance learning for humanoid robots[C]//International Conference on Advanced Robotics and Intelligent Systems. Piscataway: IEEE, 2016: 1-6.
[7]	SEPP H, JURGEN S. Long short-term memory[J]. MIT Press Journals, 1997, 9(8):1735-1780.
[8]	于建均, 吴鹏申, 左国玉, 等.基于RNN机械臂任务模仿系统[J].北京工业大学学报, 2018, 44(11):37-44. doi: 10.11936/bjutxb2017090017 YU J J, WU P S, ZUO G Y, et al. Robot arm task imitation system based on RNN[J]. Journal of Beijing University of Technology, 2018, 44(11):37-44. (in Chinese) doi: 10.11936/bjutxb2017090017
[9]	黄娜, 何泾沙, 孙靖超, 等.基于改进LSTM网络的犯罪态势预测方法[J].北京工业大学学报, 2019, 45(8):742-748. doi: 10.11936/bjutxb2018070018 HUANG N, HE J S, SUN J C, et al. A crime situation prediction method based on improved LSTM network[J]. Journal of Beijing University of Technology, 2019, 45(8):742-748. doi: 10.11936/bjutxb2018070018
[10]	WANG B Y, MING H. Predicting body movement and recognizing actions: an integrated framework for mutual benefits[C]//International Conference on Automatic Face & Gesture Recognition. Piscataway: IEEE, 2018: 341-348.
[11]	CHELLALI R, LI C Z. A multi-variate LSTM based approach for human-robot hand clapping games[C]//International Symposiumon Robot and Human Interactive Communication. Piscataway: IEEE, 2018: 1137-1142.
[12]	DICKSON N T, SAHARI K S M, HU Y H, et al. Multiple sequence behavior recognition on humanoid robot using long short-term memory[C]//Interational Symposium on Robotics and Manufacturing Automation. Piscataway: IEEE, 2014: 109-114.
[13]	于建均, 李晨, 左国玉, 等.仿人机器人步态平衡泛化模型的建立与仿真[J].智能系统学报, 2020(7):1-9. YU J J, LI C, ZUO G Y, et al. Modeling and simulation of humanoid robot gait balance generalization[J]. CAAI Transactions on Intelligent Systems, 2020(7):1-9. (in Chinese)
[14]	马冬玲.相空间重构中时间延迟τ的确定方法[J].硅谷, 2013, 6(19):48-51. http://www.cnki.com.cn/Article/CJFDTotal-GGYT201319087.htm MA D L. Phase space reconstruction method for determining the time delay τ[J]. Silicon Valley, 2013, 6(19):48-51. (in Chinese) http://www.cnki.com.cn/Article/CJFDTotal-GGYT201319087.htm
[15]	LONGKUN T, JIANLI L. C-C method to phase space reconstruction based on multivariate time series[C]//International Conference on Intelligent Control and Information Processing. Piscataway: IEEE, 2011: 438-441.
[16]	于建均, 张远, 左国玉, 等.基于ZMP判据的仿人机器人步态模仿[J].北京工业大学学报, 2018, 44(9):1187-1192. doi: 10.11936/bjutxb2017070025 YU J J, ZHANG Y, ZUO G Y, et al. Humanoid robot gait imitation based on ZMP criterion[J]. Journal of Beijing University of Technology, 2018, 44(9):1187-1192. (in Chinese) doi: 10.11936/bjutxb2017070025
[17]	JUNICHI N, YASUO K, NOZOMU A, et al. Slope-walking of a biped robot with K nearest neighbor method[C]//Fourth International Conference on Innovative Computing, Information and Control. Piscataway: IEEE, 2009: 173-176.
[18]	HAO Z, MUQING D, PENG Y. Human gait recognition based on deterministic learning and Kinect sensor[C]//Chinese Control And Decision Conference. Piscataway: IEEE, 2018: 1842-1847.
[19]	TOMOYA S, SHO S, EIJIRO O, et al. Walking trajectory planning on stairs using virtual slope for biped robots[C]//Transactions on Industrial Electronics. Piscataway: IEEE, 2011: 1385-139.

Cited By

Cited by

Periodical cited type(7)

1.	刘晓铭，华定忠，张静，黄国方，钟亮民. 基于姿态信息的双足移动机器人越障步态控制. 机械与电子. 2024(04): 29-32+38 .
2.	刘小翠. 基于ZMP控制器的仿真机器人偏摆力矩控制设计. 现代工业经济和信息化. 2024(08): 110-112 .
3.	田启磊，王钰. 利用惯性传感器数据实时示教下肢外骨骼步态的研究. 青岛大学学报(自然科学版). 2023(01): 54-59 .
4.	丁玲. 基于神经网络深度强化学习的分拣机器人避障规划技术. 工业仪表与自动化装置. 2023(02): 46-50 .
5.	罗洁，王中训，潘康路，卢中原，刘言. 基于改进人工势场法的无人车路径规划算法. 电子设计工程. 2022(17): 90-94+99 .
6.	谢鸿涛. 仿人机器人的发展进程与实际应用研究. 造纸装备及材料. 2022(12): 56-58 .
7.	徐凌峰，高洪. 基于LSTM神经网络的乘用车能耗预测. 井冈山大学学报(自然科学版). 2021(05): 78-84 .

Other cited types(9)

Get Citation

PDF

XML

Article views (123) PDF downloads (77) Cited by(16)

Turn off MathJax

Article Contents

Abstract

References

Method for Generating Cyclic Gait of Biped Robots Based on LSTM Neural Network

Abstract

References

Cited by

Periodical cited type(7)

Other cited types(9)

Catalog

Export File

Citation

Format

Content