[1]NIU Miaohe,LEI Fei.Motion intention recognition algorithms for lower limb exoskeleton[J].CAAI Transactions on Intelligent Systems,2025,20(2):407-415.[doi:10.11992/tis.202403025]
Copy

Motion intention recognition algorithms for lower limb exoskeleton

CAAI Transactions on Intelligent Systems[ISSN 1673-4785/CN 23-1538/TP] Volume: 20 Number of periods: 2025 2 Page number: 407-415 Column: 学术论文—机器感知与模式识别 Public date: 2025-03-05
Title:
Motion intention recognition algorithms for lower limb exoskeleton
Author(s):
NIU Miaohe1;  LEI Fei2
1. Beijing-Dublin International College, Beijing University of Technology, Beijing 100020, China;
2. School of Information Science and Technology, Beijing University of Technology, Beijing 100020, China
Keywords:
lower limb exoskeleton; assisted walking; surface electromyography signals; human–machine movement; multi-scale; convolutional neural network; perceptual model; intention recognition
CLC:
TP391.4
DOI:
10.11992/tis.202403025
Abstract:
The rapid advancement of artificial intelligence and sensing technology has highlighted the potential of lower-limb exoskeleton technology in assisted walking and motion assistance. Decoding human motion intention using surface myoelectric (sEMG) signals is essential for achieving coordination and unification of human–machine motion. However, owing to the spatiotemporal differences and nonlinear dynamics of sEMG signals, existing methods have limitations, such as single feature capture and low recognition accuracy. A motion intention perception model based on a multiscale convolutional neural network is proposed to solve these problems. The model uses multiple differential convolution blocks to extract the temporal and spatial scale features of sEMG signals. It utilizes a multi-layer deep network to capture the nonlinear dynamic features of sEMG signals. The recognition accuracy of this model for the offline EMG database is 94%, and that of the whole-foot off-ground motion category is 98%. The average recognition accuracy of this model is more than 90% in the experiment of online motion intention recognition using a lower-limb exoskeleton, which verifies its effectiveness in the field of lower limb exoskeleton intention recognition.
References:
[1] 白雪, 梁文渊, 王强, 等. 下肢外骨骼机器人研究进展及趋势[J]. 中国科技信息, 2023(17): 146-149.
BAI Xue, LIANG Wenyuan, WANG Qiang, et al. Research progress and trend of lower limb exoskeleton robot[J]. China science and technology information, 2023(17): 146-149.
[2] 张洋, 谭昕, 程闯. 基于表面肌电信号的下肢运动识别[J]. 机电工程技术, 2023, 52(9): 69-73,151.
ZHANG Yang, TAN Xin, CHENG Chuang. Lower limb motion recognition based on surface EMG signals[J]. Mechanical & electrical engineering technology, 2023, 52(9): 69-73,151.
[3] MILETI I, TABORRI J, TORRICELLI D, et al. Artificial neural network for the identification of postural instability in subject wearing lower limb exoskeleton[C]//2022 IEEE International Conference on Metrology for Extended Reality, Artificial Intelligence and Neural Engineering. Rome: IEEE, 2022: 651-655.
[4] 田双太. 一种可穿戴机器人的多传感器感知系统研究[D]. 合肥: 中国科学技术大学, 2011.
TIAN Shuangtai. Research on multi-sensor sensing system of wearable robot[D]. Hefei: University of Science and Technology of China, 2011.
[5] 张倩. 下肢外骨骼康复机器人的机构设计与分析[D]. 成都: 电子科技大学, 2013.
ZHANG Qian. Mechanism design and analysis of lower limb exoskeleton rehabilitation robot [D]. Chengdu: University of Electronic Science and Technology of China, 2013.
[6] 蒋靖. 下肢助力外骨骼机构设计与研究[D]. 哈尔滨: 哈尔滨工业大学, 2014.
JIANG Jing. Design and research of lower extremity assisted exoskeleton mechanism [D]. Harbin: Harbin Institute of Technology, 2014.
[7] 袁小庆, 邹缓, 吴涛, 等. 基于运动意图识别的上肢助力外骨骼复合控制方法研究[J]. 机械工程学报, 2023, 59(15): 73-82.
YUAN Xiaoqing, ZOU Huan, WU Tao, et al. Compound control method for human-powered augmentation upper exoskeleton based on motion intent recognition[J]. Journal of mechanical engineering, 2023, 59(15): 73-82.
[8] SU Benyue, WANG Jie, LIU Shuangqing, et al. A CNN-based method for intent recognition using inertial measurement units and intelligent lower limb prosthesis[J]. IEEE transactions on neural systems and rehabilitation engineering, 2019, 27(5): 1032-1042.
[9] 丁其川, 熊安斌, 赵新刚, 等. 基于表面肌电的运动意图识别方法研究及应用综述[J]. 自动化学报, 2016, 42(1): 13-25.
DING Qichuan, XIONG Anbin, ZHAO Xingang, et al. A review on researches and applications of sEMG-based motion intent recognition methods[J]. Acta automatica sinica, 2016, 42(1): 13-25.
[10] 佟丽娜, 侯增广, 彭亮, 等. 基于多路sEMG时序分析的人体运动模式识别方法[J]. 自动化学报, 2014, 40(5): 810-821.
TONG Lina, HOU Zengguang, PENG Liang, et al. Multi-channel sEMG time series analysis based human motion recognition method[J]. Acta automatica sinica, 2014, 40(5): 810-821.
[11] SONG Ge, WANG Yucheng, WANG Mingming, et al. Lower limb movement intent recognition based on grid search random forest algorithm[C]//Proceedings of the 3rd International Conference on Robotics, Control and Automation. Chengdu: ACM, 2018: 225–229.
[12] WEI Yue, GENG Yanjuan, YU Wenlong, et al. Real-time classification of forearm movements based on high density surface electromyography[C]//2017 IEEE International Conference on Real-time Computing and Robotics. Okinawa: IEEE, 2017: 246-251.
[13] ZHOU Congcong, YANG Lilin, LIAO Heng, et al. Ankle foot motion recognition based on wireless wearable sEMG and acceleration sensors for smart AFO[J]. Sensors and actuators a: physical, 2021, 331: 113025.
[14] SHEN Shu, GU Kang, CHEN Xinrong, et al. Gesture recognition through sEMG with wearable device based on deep learning[J]. Mobile networks and applications, 2020, 25: 2447-2458.
[15] REN Xina, SOO Y G, ODAGAKI M, et al. sEMG-based hand motion recognition system using RMSR and AR model[C]//2017 36th Chinese Control Conference. Dalian: IEEE, 2017: 5410-5415.
[16] ZENG Min, GU J, FENG Ying. Motion prediction based on sEMG- transformer for lower limb exoskeleton robot control[C]//2023 International Conference on Advanced Robotics and Mechatronics. Sanya: IEEE, 2023: 864-869.
[17] WEI Baichun, YI Chunzhi, ZHANG Qi, et al. ActiveSelfHAR: incorporating self-training into active learning to improve cross-subject human activity recognition[J]. IEEE internet of things journal, 2024, 11(4): 6833-6847.
[18] ANDERSON F C, PANDY M G. Individual muscle contributions to support in normal walking[J]. Gait & posture, 2003, 17(2): 159-169.
[19] AYACHI M, SEDDIK H. Overview of EMG signal preprocessing and classification for bionic hand control[C]//2022 IEEE Information Technologies & Smart Industrial Systems. Paris: IEEE, 2022: 1-6.
[20] DE LUCA C J, KUZNETSOV M, GILMORE L D, et al. Inter-electrode spacing of surface EMG sensors: reduction of crosstalk contamination during voluntary contractions[J]. Journal of biomechanics, 2012, 45(3): 555-561.
[21] 杜妍辰, 孙洁, 汪晓铭, 等. 一种基于表面肌电信号映射人体下肢运动意图的方法[J]. 生物医学工程学进展, 2023, 44(2): 158-162.
DU Yanchen, SUN Jie, WANG Xiaoming, et al. A method of human lower limb motion intention mapping based on surface electromyography[J]. Progress in biomedical engineering, 2023, 44(2): 158-162.
[22] WU Yi, FOURN E, BESNIER P. A planar quad-band band-pass filter employing dual-mode band-stop resonators[C]//2021 IEEE MTT-S International Microwave Filter Workshop. Perugia: IEEE, 2021: 1-3.
[23] DENG Yaping, JIA Hao, LI Pengcheng, et al. A deep learning method based on long short term memory and sliding time window for type recognition and time location of power quality disturbance[C]//2018 Chinese Automation Congress. Xi’an: IEEE, 2018: 1764-1768.
[24] BAO Tianzhe, ALI RAZA ZAIDI S, XIE Shengquan, et al. Inter-subject domain adaptation for CNN-based wrist kinematics estimation using sEMG[J]. IEEE transactions on neural systems and rehabilitation engineering, 2021, 29: 1068-1078.
[25] SAITO K, WATANABE K, USHIKU Y, et al. Maximum classifier discrepancy for unsupervised domain adaptation[C]//2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition. Salt Lake City: IEEE, 2018: 3723-3732.
[26] WANG Kun, HUANG Yining, XU Lichao, et al. Decoding of pre-movement EEG patterns using cross-entropy loss-based ensemble learning[C]//2022 16th ICME International Conference on Complex Medical Engineering. Zhongshan: IEEE, 2022: 260-263.
[27] CHO J W, KIM D J, JUNG Y, et al. MCDAL: maximum classifier discrepancy for active learning[J]. IEEE transactions on neural networks and learning systems, 2023, 34(11): 8753-8763.
[28] KONAR J, KHANDELWAL P, TRIPATHI R. Comparison of various learning rate scheduling techniques on convolutional neural network[C]//2020 IEEE International Students’ Conference on Electrical, Electronics and Computer Science. Bhopal: IEEE, 2020: 1-5.
[29] IIDUKA H. Appropriate learning rates of adaptive learning rate optimization algorithms for training deep neural networks[J]. IEEE transactions on cybernetics, 2022, 52(12): 13250-13261.
[30] LLUGSI R, EL YACOUBI S, FONTAINE A, et al. Comparison between Adam, AdaMax and AdamW optimizers to implement a weather forecast based on neural networks for the Andean city of Quito[C]//2021 IEEE Fifth Ecuador Technical Chapters Meeting. Cuenca: IEEE, 2021: 1-6.
[31] RAY N T, KNARR B A, HIGGINSON J S. Walking speed changes in response to novel user-driven treadmill control[J]. Journal of biomechanics, 2018, 78: 143-149.
[32] RAFFALT P, GUUL M K, NIELSEN A, et al. Economy, movement dynamics, and muscle activity of human walking at different speeds[J]. Scientific reports, 2017, 7(1): 43986.
Similar References:

Memo

-

Last Update: 2025-03-05

Copyright © CAAI Transactions on Intelligent Systems