Please wait a minute...
浙江大学学报(工学版)  2021, Vol. 55 Issue (4): 648-657    DOI: 10.3785/j.issn.1008-973X.2021.04.006
计算机技术、电信技术     
基于GWO-SVM的下肢假肢穿戴者骑行相位识别
高新智(),刘作军*(),张燕,陈玲玲
河北工业大学 智能康复装置与检测技术教育部工程研究中心,天津 300130
Bicycle riding phase recognition of lower limb amputees based on GWO-SVM
Xin-zhi GAO(),Zuo-jun LIU*(),Yan ZHANG,Ling-ling CHEN
Engineering Research Center of Intelligent Rehabilitation and Detecting Technology, Ministry of Education, Hebei University of Technology, Tianjin 300130, China
 全文: PDF(1222 KB)   HTML
摘要:

针对下肢假肢穿戴者骑行相位识别的问题,提出基于灰狼算法优化的支持向量机(GWO-SVM)分类模型. 建立下肢多源信息系统,采集膝关节、踝关节的加速度信号以及膝关节角度信号. 应用奇异值分解,对采集到的信号进行降噪处理. 在对信号进行降噪处理之后,为了避免单一信号不确定的影响,从数据冗余角度,选取各信号的特征点,开展归一化处理,组成多维特征向量,作为SVM分类模型的输入. 为了能够进一步提高分类精度,加强全局优化能力,利用GWO算法对核参数进行优化. 通过与PSO-SVM分类模型、GA-SVM分类模型对比表明,基于GWO优化的SVM分类模型对骑行相位的识别率为94%,高于其他方法优化的SVM分类模型.

关键词: 下肢假肢骑行运动相位识别灰狼优化(GWO)支持向量机(SVM)    
Abstract:

An approach based on gray wolf algorithm optimization and support vector machine was proposed aiming at the problem of identifying the riding phases of lower limb prosthetic wearers. A multi-sensor system was constructed for the motion data collection. Then the acceleration signals of the knee joint, ankle joint and the angle signals of the knee joint on the prosthetic side were collected. Singular value noise reduction was used to reduce the noise of the collected signal. Then feature points of each signal were selected and normalized from the perspective of data redundancy. These motion feature point signals formed a multi-dimensional feature vector as the input of SVM classification model, which solved the problem of the uncertain influence of a single signal. The gray wolf algorithm optimized support vector machine kernel parameters, which not only improved the classification accuracy of the recognition, but also enhanced the global optimization ability. The support vector machine model optimized by the gray wolf algorithm has an accuracy rate of 94% for bicycle riding phase recognition, which is higher than the support vector machine model based on particle swarm optimization and the support vector machine model optimized based on genetic optimization algorithm.

Key words: lower prosthesis    bicycle riding    phase recognition    grey wolf optimization (GWO)    support vector machine (SVM)
收稿日期: 2020-07-02 出版日期: 2021-05-07
CLC:  TP 242  
基金资助: 国家自然科学基金资助项目(61703135,61773151);河北省青年自然科学基金资助项目(F2018202279)
通讯作者: 刘作军     E-mail: 982570895@qq.com;liuzuojun@hebut.edu.cn
作者简介: 高新智(1993—),男,博士生,从事智能假肢、模式识别的研究. orcid.org/0000-0001-7644-7667. E-mail: 982570895@qq.com
服务  
把本文推荐给朋友
加入引用管理器
E-mail Alert
作者相关文章  
高新智
刘作军
张燕
陈玲玲

引用本文:

高新智,刘作军,张燕,陈玲玲. 基于GWO-SVM的下肢假肢穿戴者骑行相位识别[J]. 浙江大学学报(工学版), 2021, 55(4): 648-657.

Xin-zhi GAO,Zuo-jun LIU,Yan ZHANG,Ling-ling CHEN. Bicycle riding phase recognition of lower limb amputees based on GWO-SVM. Journal of ZheJiang University (Engineering Science), 2021, 55(4): 648-657.

链接本文:

http://www.zjujournals.com/eng/CN/10.3785/j.issn.1008-973X.2021.04.006        http://www.zjujournals.com/eng/CN/Y2021/V55/I4/648

图 1  骑行相位划分
图 2  加速度、角度测量仪器
图 3  假肢穿戴者骑行图片
受试者编号 性别 年龄/岁 身高/cm
A1 42 173
A2 27 175
A3 25 162
表 1  受试者情况
图 4-1  
图 4  运动信号预处理
受试者 θp /(°) θl /(°) θr /(°) θu /(°)
A1 89.27 113.72 100.51 80.39
A2 86.75 108.26 96.86 75.24
A3 85.73 103.62 96.83 78.57
表 2  受试者各相位对应角度
图 5  二分类法示意图
核函数 C=2,γ=1 C=1,γ=1 C=2,γ=0.5 C=3,γ=2
polynomial 75% 73% 48% 83%
sigmoid 31% 28% 82% 3%
linear 77% 83% 80% 85%
RBF 79% 85% 85% 83%
表 3  不同核函数的测试结果
图 6  骑行相位分类结果
图 7  GWO优化SVM分类模型算法流程
图 8  优化SVM识别结果
优化方法 参数设置 寻优结果 A /%
GA-SVM np = 20,
PC = 0.9,
Pm = 0.01
C = 44.602 7
γ = 25.110 6
87
PSO-SVM npar = 20,
c1 = 1.5,
c2 = 1.7
C = 53.835 4
γ = 22.018 6
90
GWO-SVM nw = 20 C = 20.246 2
γ = 35.184 2
94
表 4  优化SVM的识别结果
算法 nr pr /%
用力区 下缓冲区 放松区 上缓冲区 总和
BP 19 15 21 17 72 72
SVM 21 22 17 19 79 79
GA-SVM 22 20 22 23 87 87
PSO-SVM 22 24 23 21 90 90
GWO-SVM 24 22 23 25 94 94
表 5  各分类模型的识别结果
图 9  3种优化算法的适应度
1 ZIEGLER-GRAHAM K, MACKENZIE E J, EPHRAIM P L, et al Estimating the prevalence of limb loss in the United States: 2005 to 2050[J]. Archives of Physical Medicine and Rehabilitation, 2008, 89 (3): 422- 429
doi: 10.1016/j.apmr.2007.11.005
2 DAVIES B, DATTA D Mobility outcome following unilateral lower limb amputation[J]. Prosthetics and Orthotics International, 2003, 27 (3): 186- 190
doi: 10.1080/03093640308726681
3 MASSALIN Y, ABDRAKHMANOVA M, VAROL H A User-independent intent recognition for lower-limb prostheses using depth sensing[J]. IEEE Transactions on Biomedical Engineering, 2017, 65 (8): 1759- 1770
4 宣伯凯, 刘作军, 陈玲玲, 等 膝上型假肢的运动意图识别与控制[J]. 东南大学学报: 自然科学版, 2017, 47 (6): 1107- 1116
XUAN Bo-kai, LIU Zuo-jun, CHEN Ling-ling, et al Motion intention recognition and control of above knee prosthesis[J]. Journal of Southeast University: Natural Science Edition, 2017, 47 (6): 1107- 1116
5 CHEN Bao-jun, WANG Qin-ning, WANG Long Adaptive slope walking with a robotic transtibial prosthesis based on volitional EMG control[J]. IEEE/ASME Transactions on Mechatronics, 2015, 20 (5): 2146- 2157
doi: 10.1109/TMECH.2014.2365877
6 赵晓东, 刘作军, 陈玲玲, 等 下肢假肢穿戴者跑动步态识别方法[J]. 浙江大学学报: 工学版, 2018, 52 (10): 1980- 1988
ZHAO Xiao-dong, LIU Zuo-jun, CHEN Ling-ling, et al Approach of running gait recognition for lower limb amputees[J]. Journal of Zhejiang University: Engineering Science, 2018, 52 (10): 1980- 1988
doi: 10.3785/j.issn.1008-973X.2018.10.018
7 赵晓东, 刘作军, 苟斌, 等 下肢假肢斜坡路况运动控制策略分析[J]. 控制与决策, 2019, 34 (6): 1160- 1168
ZHAO Xiao-dong, LIU Zuo-jun, GOU Bin, et al Analysis of slope motion control strategy of lower limb prostheses[J]. Control and Decision, 2019, 34 (6): 1160- 1168
8 CHAICHAOWARAT R, GRANADOS D F P, KINUGAWA J, et al. Passive knee exoskeleton using torsion spring for cycling assistance [C]// IEEE International Conference on Intelligent Robots and Systems. Vancouver: IEEE, 2017: 3069-3074.
9 LAWSON B E, LEDOUX E D, GOLDFARB M A robotic lower limb prosthesis for efficient bicycling[J]. IEEE Transactions on Robotics, 2017, 33 (2): 432- 445
doi: 10.1109/TRO.2016.2636844
10 KIM H W, AN J W, YOO H D, et al. Balancing control of bicycle robot using PID control [C]// International Conference on Control, Automation and Systems. Gwangju: IEEE, 2013: 145-147.
11 赵丽娜, 刘作军, 苟斌, 等 基于隐马尔可夫模型的动力型下肢假肢步态预识别[J]. 机器人, 2014, 36 (3): 337- 341
ZHAO Li-na, LIU Zuo-jun, GOU Bin, et al Gait pre-recognition of dynamic lower limb prosthesis based on hidden Markov model[J]. Robot, 2014, 36 (3): 337- 341
12 王永雄, 陈晗, 尹钟, 等 基于惯导信息的人体动作和路况识别[J]. 生物医学工程学杂志, 2018, 35 (4): 621- 630
WANG Yong-xiong, CHEN Han, YIN Zhong, et al Human action and road condition recognition based on the inertial information[J]. Journal of Biomedical Engineering, 2018, 35 (4): 621- 630
13 GHAMISI P, BENEDIKTSSON J A Feature selection based on hybridization of genetic algorithm and particle swarm optimization[J]. IEEE Geoscience and Remote Sensing Letters, 2015, 12 (2): 309- 313
doi: 10.1109/LGRS.2014.2337320
14 彭亮, 侯增广, 王晨, 等 康复辅助机器人及其物理人机交互方法[J]. 自动化学报, 2018, 44 (11): 2000- 2010
PENG Liang, HOU Zeng-guang, WANG Chen, et al Physical interaction methods for rehabilitation and assistive robots[J]. Acta Automatica Sinica, 2018, 44 (11): 2000- 2010
15 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 Qi-ning, ZHENG En-hao, CHENG Bao-jun, et al Recent progress and challenges of robotic lower-limb prostheses for human robot integration[J]. Acta Automatica Sinica, 2016, 42 (12): 1780- 1793
16 YOUNG A J, HARGROVE L J, KUIKEN T A The effects of electrode size and orientation on the sensitivity of myoelectric pattern recognition systems to electrode shift[J]. IEEE Transactions on Biomedical Engineering, 2011, 58 (9): 2537- 2544
doi: 10.1109/TBME.2011.2159216
17 CHEN Bao-jun, WANG Xue-gang, HUANG Yan, et al A foot-wearable interface for locomotion mode recognition based on discrete contact force distribution[J]. Mechatronics, 2015, 12 (32): 12- 21
18 JURI T, EDUARDO P, STEFANO R, et al Gait partitioning methods: a systematic review[J]. Sensors, 2016, 16 (1): 66- 82
doi: 10.3390/s16010066
19 YANG Bin, YU Chun-yang, DONG Yong-gui Capacitively-coupled electrocardiogram measuring system and noise reduction by singular spectrum analysis[J]. IEEE Sensors Journal, 2016, 10 (16): 3802- 3810
20 MONDAL A, SAXENA I, TANG Hong, et al A noise reduction technique based on nonlinear kernel function for heart sound analysis[J]. IEEE Journal of Biomedical and Health Informatics, 2017, 22 (3): 775- 784
21 张晓涛, 李伟光 基于奇异值和奇异向量的振动信号降噪方法[J]. 振动. 测试与诊断, 2018, 38 (3): 553- 558
ZHANG Xiao-tao, LI Wei-guang Noise reduction method of vibration signal based on singular value and singular vector[J]. Journal of Vibration, Measurement and Diagnosis, 2018, 38 (3): 553- 558
22 WEI Qiang, WU Bin, XU De-wei, et al Natural sampling SVM-based common-mode voltage reduction in medium-voltage current source rectifier[J]. IEEE Transactions on Power Electronics, 2017, 32 (10): 7553- 7560
doi: 10.1109/TPEL.2016.2632442
23 WU Jian-xin, YANG Hao Linear regression-based efficient SVM learning for large-scale classification[J]. IEEE Transactions on Neural Networks and Learning Systems, 2017, 26 (10): 2357- 2369
24 YIN Ying-jie, XU De, WANG Xin-gang, et al Online state-based structured SVM combined with incremental PCA for robust visual tracking[J]. IEEE Transactions on Cybernetics, 2015, 45 (9): 1988- 2000
doi: 10.1109/TCYB.2014.2363078
25 严秀权, 吴洪涛, 李耀, 等 压电作动器的支持向量机迟滞模型[J]. 仪器仪表学报, 2018, 39 (9): 228- 235
YAN Xiu-quan, WU Hong-tao, LI Yao, et al Support vector machine-based hysteresis model of piezoelectric actuator[J]. Chinese Journal of Scientific Instrument, 2018, 39 (9): 228- 235
26 刘忠宝, 秦振涛, 罗学刚, 等 利用融合数据分布特征的模糊双支持向量机对恒星光谱分类[J]. 光谱学与光谱分析, 2019, 39 (4): 1307- 1311
LIU Zhong-bao, QIN Zhen-tao, LUO Xue-gang, et al Stellar spectra classification by support vector machine with spectral distribution properties[J]. Spectroscopy and Spectral Analysis, 2019, 39 (4): 1307- 1311
27 LIANG Gao-bo, HONG Hui-chao, XIE Wei-fang, et al Combining convolutional neural network with recursive neural network for blood cell image classification[J]. IEEE Access, 2018, 7 (6): 36188- 36197
28 李镇, 樊炳辉, 王鑫, 等 基于BP神经网络的仿生眼定位追踪算法[J]. 机器人, 2017, 39 (1): 63- 69
LI Zhen, FAN Bing-hui, WANG Xin, et al Locating and tracking algorithm of biomimetic eye based on backpropagation neural network[J]. Robot, 2017, 39 (1): 63- 69
29 焦晓璇, 景博, 李娟, 等 基于ADEGWO-SVM的机载燃油泵寿命预测研究[J]. 仪器仪表学报, 2018, 39 (8): 43- 52
JIAO Xiao-xuan, JIN Bo, LI Juan, et al Research on remaining useful life prediction of fuel pump based on adaptive differential evaluation grey wolf optimizer-support vector machine[J]. Chinese Journal of Scientific Instrument, 2018, 39 (8): 43- 52
30 戚晓利, 王振亚, 吴保林, 等 基于ACMPE、ISSL-Isomap和GWO-SVM的行星齿轮箱故障诊断[J]. 航空动力学报, 2019, 34 (4): 744- 755
QI Xiao-li, WANG Zhen-ya, WU Bao-lin, et al Planetary gearbox fault diagnosis based on ACMPE, ISSL-Isomap and GWO-SVM[J]. Journal of Aerospace Power, 2019, 34 (4): 744- 755
31 李麟玮, 吴益平, 苗发盛 基于灰狼支持向量机的非等时距滑坡位移预测[J]. 浙江大学学报: 工学版, 2018, 52 (10): 1998- 2006
LI Lin-wei, WU Yi-ping, MIAO Fa-sheng Prediction of non-equidistant landslide displacement time series based on grey wolf support vector machine[J]. Journal of Zhejiang University: Engineering Science, 2018, 52 (10): 1998- 2006
doi: 10.3785/j.issn.1008-973X.2018.10.020
[1] 黄杰,王东,王新晴,殷勤,邵发明. 液压挖掘机作业循环状态智能识别方法[J]. 浙江大学学报(工学版), 2019, 53(9): 1663-1673.
[2] 都明宇, 鲍官军, 杨庆华, 王志恒, 张立彬. 基于改进支持向量机的人手动作模式识别方法[J]. 浙江大学学报(工学版), 2018, 52(7): 1239-1246.
[3] 赵晓东, 刘作军, 陈玲玲, 杨鹏. 下肢假肢穿戴者跑动步态识别方法[J]. 浙江大学学报(工学版), 2018, 52(10): 1980-1988.
[4] 李麟玮, 吴益平, 苗发盛. 基于灰狼支持向量机的非等时距滑坡位移预测[J]. 浙江大学学报(工学版), 2018, 52(10): 1998-2006.
[5] 袁红, 王波, 王丽, 许睦旬. 以轮廓为对象的体态特征情绪分类与预测[J]. 浙江大学学报(工学版), 2018, 52(1): 160-165.
[6] 尤海辉, 马增益, 唐义军, 王月兰, 郑林, 俞钟, 吉澄军. 循环流化床入炉垃圾热值软测量[J]. 浙江大学学报(工学版), 2017, 51(6): 1163-1172.
[7] 廖苗, 赵于前, 曾业战, 黄忠朝, 张丙奎, 邹北骥. 基于支持向量机和椭圆拟合的细胞图像自动分割[J]. 浙江大学学报(工学版), 2017, 51(4): 722-728.
[8] 钟崴, 彭梁, 周永刚, 徐剑, 从飞云. 基于小波包分析和支持向量机的锅炉结渣诊断[J]. 浙江大学学报(工学版), 2016, 50(8): 1499-1506.
[9] 赵凌, 黄平捷, 刘宝玲, 赵树浩, 侯迪波, 张光新. 多层导电结构内部状态脉冲涡流检测分析方法[J]. 浙江大学学报(工学版), 2016, 50(4): 603-608.
[10] 潘翔,童伟淮,张三元,郑河荣. 结合语义本体与泊松方程的动画角色模型分割[J]. 浙江大学学报(工学版), 2015, 49(9): 1634-1641.
[11] 施锦河, 沈继忠, 王攀. 四类运动想象脑电信号特征提取与分类算法[J]. J4, 2012, 46(2): 338-344.
[12] 刘晨彬,潘颖,张海石,黄峰平,夏顺仁. 基于磁共振图像的脑瘤MGMT表达状况检测算法[J]. J4, 2012, 46(1): 170-176.
[13] 佘青山, 孟明, 罗志增, 马玉良. 基于多核学习的下肢肌电信号动作识别[J]. J4, 2010, 44(7): 1292-1297.
[14] 胡志坤, 王美铃, 桂卫华, 阳春华, 丁家峰. 基于支持向量机的时序周波波形分类方法[J]. J4, 2010, 44(7): 1327-1332.
[15] 马玉良, 徐文良, 孟明, 罗志增, 杨家强. 基于神经网络的智能下肢假肢自适应控制[J]. J4, 2010, 44(7): 1373-1376.