传感技术学报2017，Vol.30Issue(8)：1171-1177,7.DOI:10.3969/j.issn.1004-1699.2017.08.007
利用多源信息和极限学习机的人体运动意图识别
The Research of Locomotion-Mode Recognition Based on Multi-SourceInformation and Extreme Learning Machine
摘要
Abstract
The efficient and accurate locomotion-mode recognition is the basis and the key to the flexible prosthesis control.The key is to recognize locomotion-mode(level-ground walking stairs ascent.stairs descent.upslope.downgrade).In order to overcome the problem that a single information source cannot recognize locomotion-mode,human locomotion-mode multi-source information systems were set up.hip joint angle、and acceleration signals and plantar pressure were used as the major source.Human locomotion-mode were decomposed into four segments using the plantar pressure information.hip joint angle features and acceleration signal features were fused into one feature vector according to the user''s locomotion modes characteristics.The features were reduced(at the cost of information content)using principal component analysis(KPCA).These data were initially used to train Extreme Learning Machine(ELM)models,which classify the patterns as upslope,downgrade,stairs ascent,stairs descent or level-ground walking.The average classification accuracy of level-ground walking.stairs ascent.stairs descent.upslope.downgrade is 96.78％.The average recognition time is 0.52 s,The results showed that the method was superior to SVM and BP method.关键词
智能假肢/步态识别/极限学习机/多源信息融合/髋关节运动信号Key words
artificial legs/locomotion-mode recognition/extreme learning machine/multi-source information fusion/hip joint motion signal分类
信息技术与安全科学引用本文复制引用
曹祥红,刘磊,杨鹏,宋寅卯..利用多源信息和极限学习机的人体运动意图识别[J].传感技术学报,2017,30(8):1171-1177,7.基金项目
国家自然科学基金项目(61203323,61503118) （61203323,61503118）
河南省高等学校重点科研项目(16B413006) （16B413006）
河北省自然科学基金项目(F2015202150,F2017202119) （F2015202150,F2017202119）
河北省高等学校科学技术研究项目(QN2015068) （QN2015068）
河南省科技厅重点科研项目(162300410070) （162300410070）
