基于自适应时序对齐的智能下肢假肢运动意图识别

doi:10.3969/j.issn.1006-9771.2025.09.014

Abstract

Abstract:

Objective To address the issue of motion misclassification caused by individual gait differences and fixed time window data extraction in motion intention recognition for intelligent lower limb prostheses, this study proposes a motion intention recognition method based on adaptive temporal alignment.

Methods In lower limb motion analysis, for continuous gait cycle data, inter-class variability across different steady-state modes was utilized to detect gait pattern consistency through inter-cycle frame differencing. For samples identified as single steady-state modes, the dynamic time warping algorithm was introduced to align adjacent motion sequences, thereby reducing individual variability. Haar wavelet 4-level decomposition was applied to extract low-frequency coefficients for feature vector construction, and classification was performed using a support vector machine. The experimental protocol was designed as follows: three inertial measurement units were used to collect lower limb acceleration and angular velocity data from subjects performing thirteen locomotion modes. The test subjects included ten healthy participants and one transtibial amputee. The locomotion modes consisted of five steady-state modes (level walking, stair ascent, stair descent, ramp ascent, and ramp descent) and eight transition modes (mutual transitions between level walking and stair ascent/descent, as well as ramp ascent/descent).

Results Simulation tests on ten healthy individuals and one amputee showed recognition accuracies of 99.24% and 100% for five steady-state modes, and 98.51% and 89.11% for all thirteen motion modes, respectively.

Conclusion This study proposes an adaptive temporal alignment-based motion intention recognition method. The proposed approach effectively reduces the interference of individual gait variability on feature representation, enhances the consistency and discriminability of gait features, and ultimately improves recognition performance.

Key words: motion intention recognition, intelligent lower-limb prostheses, inertial measurement unit, dynamic time warping, adaptive temporal alignment, inter-frame difference, single gait mode

CLC Number:

R496

SU Benyue, LIU Wenyao, ZONG Wenjie, WANG Baoqian, SHENG Min. Adaptive temporal alignment-based motion intention recognition for intelligent lower-limb prostheses[J]. Chinese Journal of Rehabilitation Theory and Practice, 2025, 31(9): 1101-1115.

Figures/Tables 18

Table 1

Figure 1

Table 2

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Table 3

Table 4

Table 5

Table 6

Figure 12

References 40

[1]	中国残疾人联合会. 2024年残疾人事业发展统计公报[EB/OL]. (2025-05-08) [2025-06-11]. https://www.cdpf.org.cn/zwgk/zccx/tjgb/1706f34657364af9a52b67f77d8c9f2b.htm.
[2]	ASIF M, TIWANA M I, KHAN U S, et al. Advancements, trends and future prospects of lower limb prosthesis[J]. IEEE Access, 2021, 9: 85956-85977.
[3]	魏艳琴, 曹学军, 杨平, 等. 下肢截肢者穿戴假肢行走能力的评价[J]. 中国康复理论与实践, 2016, 22(7): 855-859. doi: 10.3969/j.issn.1006-9771.2016.07.028
	WEI Y Q, CAO X J, YANG P, et al. Evaluation for walking ability of lower limb amputees with prostheses: a literature analysis[J]. Chin J Rehabil Theory Pract, 2016, 22(7): 855-859.
[4]	顾洪, 李伟达, 李娟. 智能膝关节假肢研究现状及发展趋势[J]. 中国康复理论与实践, 2016, 22(9): 1080-1085. doi: 10.3969/j.issn.1006-9771.2016.09.021
	GU H, LI W D, LI J. State-of-the-art and development of intelligent knee prosthesis (review)[J]. Chin J Rehabil Theory Pract, 2016, 22(9): 1080-1085.
[5]	张意彬, 吕杰, 喻洪流. 基于模糊逻辑算法的智能膝关节假肢步态相位识别[J]. 中国康复理论与实践, 2023, 29(8): 896-902. doi: 10.3969/j.issn.1006-9771.2023.08.005
	ZHANG Y B, LÜ J, YU H L. Gait phase recognition in intelligent above-knee prosthesis based on fuzzy logic algorithm[J]. Chin J Rehabil Theory Pract, 2023, 29(8): 896-902.
[6]	喻洪流. 康复机器人:未来十大远景展望[J]. 中国康复医学杂志, 2020, 35(8): 900-902.
[7]	刘作军, 许长寿, 陈玲玲, 等. 智能假肢膝关节的研发要点及其研究进展综述[J]. 包装工程, 2021, 42(10): 54-63.
	LIU Z J, XU C S, CHEN L L, et al. Key point and progress of intelligent prosthesis knee joint research[J]. Packaging Eng, 2021, 42(10): 54-63.
[8]	王蕾, 王辉, 黄品高, 等. 下肢截肢者行走意图识别方法研究进展[J]. 自动化学报, 2018, 44(8): 1370-1380.
	WANG L, WANG H, HUANG P G, et al. Progress and perspective of recognition methods for walking intention of lower-limb amputees[J]. Acta Automatica Sinica, 2018, 44(8): 1370-1380.
[9]	孙为双, 路知远, 公维军, 等. 表面肌电信号在脑卒中手功能康复机器人中的研究进展[J]. 中国康复医学杂志, 2025, 40(5): 790-794.
[10]	张尧. 智能动力膝关节假肢的人体运动意图识别技术研究[D]. 长春: 吉林大学, 2024.
	ZHANG Y. Research on human locomotion intent recognition technology for intelligent powered knee prosthesis[D]. Changchun: Jilin University, 2024.
[11]	WANG E, CHEN X, LI Y, et al. Lower limb motion intent recognition based on sensor fusion and fuzzy multitask learning[J]. IEEE Trans Fuzzy Syst, 2024, 32(5): 2903-2914.
[12]	MA X, LIU Y, ZHANG X, et al. Real-time continuous locomotion mode recognition and transition prediction for human with lower limb exoskeleton[J]. IEEE J Biomed Health Inform, 2025, 29(2): 1074-1086.
[13]	CARVALHO S P, FIGUEIREDO J, CERQUEIRA J J, et al. Locomotion mode prediction in real-life walking with and without ankle-foot exoskeleton assistance[J]. Appl Intell, 2025, 55: 546.
[14]	黄品高. 智能下肢假肢运动意图的感测与识别关键技术研究[D]. 深圳: 中国科学院大学(中国科学院深圳先进技术研究院), 2020.
	HUANG P G. Study on the key technologies of motion intention sensing and recognition of intelligent lower limb prostheses[D]. Shenzhen: The University of Chinese Academy of Sciences (Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences), 2020.
[15]	陈辉, 任志刚, 冯祖仁, 等. 基于IMU和运动学的四足机器人对角支撑状态估计算法[J]. 控制与决策, 2024, 39(9): 2894-2902.
	CHEN H, REN Z G, FENG Z R, et al. State estimation for diagonal support of quadruped robot based on IMU and kinematics[J]. Control Decision, 2024, 39(9): 2894-2902.
[16]	杨平, 陈浩源, 高睿馨, 等. 基于低成本惯性测量单元和力敏电阻的足部三维运动数据建模[J]. 中国康复理论与实践, 2024, 30(10): 1224-1231. doi: 10.3969/j.issn.1006-9771.2024.10.013
	YANG P, CHEN H Y, GAO R X, et al. Three-dimensional modeling of foot motion based on low-cost inertial measurement unit and force sensing resistor[J]. Chin J Rehabil Theory Pract, 2024, 30(10): 1224-1231.
[17]	SU B Y, WANG J, LIU S Q, et al. A CNN-based method for intent recognition using inertial measurement units and intelligent lower limb prosthesis[J]. IEEE Trans Neural Syst Rehabil Eng, 2019, 27(5): 1032-1042.
[18]	盛敏, 夏安琦, 王可林, 等. 基于几何与物理特征融合的智能下肢假肢运动意图识别[J]. 控制与决策, 2022, 37(4): 953-961.
	SHENG M, XIA A Q, WANG K L, et al. Movement intention recognition of intelligent lower limb prosthesis based on the fusion of geometric and physical features[J]. Control Decision, 2022, 37(4): 953-961.
[19]	KIM H, LEE D, MALDONADO-CONTRERAS J Y, et al. Mode-unified intent estimation of a robotic prosthesis using deep-learning[J]. IEEE Robot Autom Lett, 2025, 10(4): 3206-3213.
[20]	唐易, 陈奕希, 喻洪流, 等. 一种面向下肢假肢的运动意图识别方法及验证[J]. 信息与控制, 2023, 52(5): 598-606.
	TANG Y, CHEN Y X, YU H L, et al. Research and verification of a motion intention recognition method for lower limb prosthesis[J]. Inform Control, 2023, 52(5): 598-606.
[21]	CHEN Z, ZHANG T. Evaluation of basic sports actions for students based on DTW posture matching algorithm[J]. System Soft Comput, 2025, 7: 200196.
[22]	HAQUE M R, ISLAM M R, SAZONOV E, et al. Swing-phase detection of locomotive mode transitions for smooth multi-functional robotic lower-limb prosthesis control[J]. Front Robot AI, 2024, 11: 1267072.
[23]	ZHENG E, WAN J, GAO S, et al. Adaptive locomotion transition recognition with wearable sensors for lower limb robotic prosthesis[J]. IEEE ASME Trans Mechatron, 2024, 29(1): 279-289.
[24]	王启宁, 郑恩昊, 陈保君, 等. 面向人机融合的智能动力下肢假肢研究现状与挑战[J]. 自动化学报, 2016, 42(12): 1780-1793.
	WANG Q N, ZHENG E H, CHEN B J, et al. Recent progress and challenges of robotic lower-limb prostheses for human-robot integration[J]. Acta Automatica Sinica, 2016, 42(12): 1780-1793.
[25]	盛敏, 刘双庆, 王婕, 等. 基于GMM-HMM模型的智能下肢假肢运动意图识别[J]. 仪器仪表学报, 2019, 40(5): 169-178.
	SHENG M, LIU S Q, WANG J, et al. Motion intent recognition of intelligent lower limb prosthesis based on GMM-HMM[J]. Chin J Sci Instrum, 2019, 40(5): 169-178.
[26]	邵企能, 王禾, 胡天羿, 等. 年龄与性别对人体步态生物力学特征的影响[J]. 应用力学学报, 2022, 39(6): 1193-1202.
	SHAO Q N, WANG H, HU T Y, et al. Effects of age and gender on biomechanical characteristics of human gait[J]. Chin J Appl Mechan, 2022, 39(6): 1193-1202.
[27]	WINTER D A. Biomechanics of normal and pathological gait[J]. J Mot Behav, 1989, 21(4): 337-355.
[28]	BERNDT D J, CLIFFORD J. Using dynamic time warping to find patterns in time series[C]. Seattle:Proceedings of the 3rd International Conference on Knowledge Discovery and Data Mining, 1994.
[29]	苏本跃, 张利, 何清旋, 等. 基于小波特征匹配的短时人体行为识别[J]. 系统仿真学报, 2023, 35(1): 158-168. doi: 10.16182/j.issn1004731x.joss.22-0176
	SU B Y, ZHANG L, HE Q X, et al. Short-time human activity recognition based on wavelet features matching[J]. J System Simul, 2023, 35(1): 158-168.
[30]	CHANG C C, LIN C J. LIBSVM: a library for support vector machines[J]. ACM Trans Intell Syst Technol, 2011, 2(3): 1-27.
[31]	周绍磊, 廖剑, 史贤俊. 基于Fisher准则和最大熵原理的SVM核参数选择方法[J]. 控制与决策, 2014, 29(11): 1991-1996.
	ZHOU S L, LIAO J, SHI X J. SVM parameters selection method based on Fisher criterion and maximum entropy principle[J]. Control Decision, 2014, 29(11): 1991-1996.
[32]	AU S K, WEBER J, HERR H. Powered ankle-foot prosthesis improves walking metabolic economy[J]. IEEE Trans Robot, 2009, 25(1): 51-66.
[33]	苏本跃, 王婕, 刘双庆, 等. 惯性动捕数据驱动下的智能下肢假肢运动意图识别方法[J]. 自动化学报, 2020, 46(7): 1517-1530.
	SU B Y, WANG J, LIU S Q, et al. An improved motion intent recognition method for intelligent lower limb prosthesis driven by inertial motion capture data[J]. Acta Automatica Sinica, 2020, 46(7): 1517-1530.
[34]	夏安琦. 基于融合特征及分层策略的下肢假肢运动意图识别[D]. 安庆: 安庆师范大学, 2021.
	XIA A Q. Motion intention recognition of lower limb prosthesis based on fusion feature and stratification strategy[D]. Anqing: Anqing Normal University, 2021.
[35]	WANG N, AMBIKAIRAJAH E, LOVELL N H, et al. Accelerometry based classification of walking patterns using time-frequency analysis[C]. Lyon:2007 29th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, 2007.
[36]	HE H, TAN Y, ZHANG W. A wavelet tensor fuzzy clustering scheme for multi-sensor human activity recognition[J]. Eng Appl Artif Intell, 2018, 70: 109-122.
[37]	苏本跃, 宗文杰, 刘文瑶, 等. 基于足底动力相数据和小波变换自适应分解的下肢意图识别方法[J]. 控制与决策, 2025, 40(10): 3005-3018.
	SU B Y, ZONG W J, LIU W Y, et al. Lower limb prosthesis intention recognition method based on powered plantarflexion phase data and wavelet transform adaptive decomposition[J]. Control Decision, 2025, 40(10): 3005-3018.
[38]	SHENG M, WANG W J, TONG T T, et al. Motion intent recognition in intelligent lower limb prosthesis using one-dimensional dual-tree complex wavelet transforms[J]. Comp Intell Neurosci, 2021, 2021: 563173.
[39]	LIU Y, AN H L, MA H X, et al. Novel feature extraction and locomotion mode classification using intelligent lower-limb prosthesis[J]. Machines, 2023, 11(2): 235.
[40]	ZHANG P, ZHANG J, ELSABBAGH A. Lower limb motion intention recognition based on sEMG fusion features[J]. IEEE Sens J, 2022, 22(7): 7005-7014.

编号	性别	年龄/岁	身高/cm	体质量/kg
HV01	男	24	183	83
HV02	男	27	180	70
HV03	男	26	162	55
HV04	男	27	173	78
HV05	男	30	175	69
HV06	女	26	170	60
HV07	女	24	165	52
HV08	女	24	178	62
HV09	女	24	163	49
HV10	女	18	155	40

模式类别	运动模式	运动模式描述
转换模式	行走—上楼	平地行走到上楼转换
	行走—下楼	平地行走到下楼转换
	行走—上坡	平地行走到上坡转换
	行走—下坡	平地行走到下坡转换
	上楼—行走	上楼到平地行走转换
	下楼—行走	下楼到平地行走转换
	上坡—行走	上坡到平地行走转换
	下坡—行走	下坡到平地行走转换
稳态模式	行走	平地行走
	上楼	稳步上台阶
	下楼	稳步下台阶
	上坡	稳步上坡
	下坡	稳步下坡

数据提取与处理	特征	准确率	精确率	召回率	F1得分
摆动相前45帧	均值方差最值	97.06	98.55	97.42	97.93
摆动相前45帧	小波4层低频系数	97.98	99.02	98.51	98.73
完整摆动相+自适应时序对齐	均值方差最值	98.02	99.89	98.10	98.96
完整摆动相+自适应时序对齐	小波4层低频系数	98.51	99.58	98.31	98.91

分类器	地形识别准确率/%	最终结果
分类器	地形识别准确率/%	准确率/%	精确率/%	召回率/%	F1得分/%	时间消耗/s
LDA	84.74	95.16	99.07	98.08	98.54	0.0291
KNN	93.17	94.32	96.98	95.30	96.05	0.0011
SVM	94.27	98.51	99.58	98.31	98.91	0.0102
RF	90.19	95.49	98.30	98.21	98.20	0.1211
GNB	76.15	77.88	93.76	87.32	90.08	0.1399

插值帧数	准确率/%	精确率/%	召回率/%	F1得分/%	时间消耗/s
16帧	95.45	98.71	98.01	98.33	0.0104
32帧	98.06	98.95	98.86	98.89	0.0140
64帧	98.51	99.58	98.31	98.91	0.0102
128帧	98.19	99.54	98.03	98.75	0.0181
256帧	98.04	99.49	97.91	98.66	0.0353

Adaptive temporal alignment-based motion intention recognition for intelligent lower-limb prostheses

RichHTML

PDF (PC)

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 18

References 40

Related Articles 3

Metrics

Comments

Recommended 0

文献	传感器		特征	运动模式种类/n		识别精度/%
文献	类型	位置	特征	稳态	转换	识别精度/%
苏本跃等^[33](2020)	IMU	健侧	均值方差最值	5	8	95.10
Sheng等^[38](2021)	IMU	健侧	DTCWT的5层低频系数	5	8	97.27
Liu等^[39](2023)	IMU 地面反作用力鞋垫	健侧	大腿相图形状膝关节角轨迹等	6	-	99.16
Zheng等^[23](2024)	IMU 压力传感器	患侧	大腿倾斜角足部倾斜角步态相位等	7	18	98.04
本研究方法	IMU	健侧	Haar小波的4层低频系数	5	8	98.51

[1]	YANG Ping, CHEN Haoyuan, GAO Ruixin, WANG Xinping, Philip ROWE. Three-dimensional modeling of foot motion based on low-cost inertial measurement unit and force sensing resistor [J]. Chinese Journal of Rehabilitation Theory and Practice, 2024, 30(10): 1224-1231.
[2]	Meng-lin CAO,Yu-hao CHEN,Jue WANG,Tian LIU. Advance in Human Motion Intention Recognition Based on Surface Electromyography (review) [J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2021, 27(5): 595-603.
[3]	YANG Bo;FU Cheng-long. Inertial Measurement Unit and Fall Risk Assessment in the Elderly (review) [J]. 《Chinese Journal of Rehabilitation Theory and Practice》, 2015, 21(07): 780-784.