基于隐马尔可夫模型的动力型下肢假肢步态预识别-机器人.pdfVIP

第36 卷第3 期 机器人 ROBOT Vol.36, No.3 2014 年5 月 May, 2014 DOI ：10.3724/SP.J.1218.2014.00337 基于隐马尔可夫模型的动力型下肢假肢步态预识别 1 12 1 12 赵丽娜 ，刘作军 ，苟 斌 ，杨 鹏 （1. 河北工业大学控制科学与工程学院，天津 300130； 2. 智能康复装置与检测技术教育部工程研究中心，天津 300130 ） 摘 要：为使动力型假肢膝关节协调配合人体的运动，关键是对人体行走步态进行有效预识别．本文利用安 装在假肢接受腔上的加速度传感器和安装在足底的压力传感器采集人体的运动信息，根据人体运动的规律性和重 复性特点，通过将隐马尔可夫模型引入到所获得的运动信息中来分析并预识别人体的运动步态．实验表明，基于 隐马尔可夫模型的动力型下肢假肢的步态预识别方法是有效并且准确的． 关键词：动力型下肢假肢；加速度传感器；步态预识别；隐马尔可夫模型 中图分类号：TP391 文献标识码：A 文章编号：1002-0446(2014)-03-0337-05 Gait Pre-recognition of Dynamic Lower Limb Prosthesis Based on Hidden Markov Model ZHAO Lina1 ，LIU Zuojun12 ，GOU Bin1 ，YANG Peng12 (1. School of Control Science and Engineering, Hebei University of Technology, Tianjin 300130, China; 2. Engineering Research Center of Intelligent Rehabilitation, Ministry of Education, Tianjin 300130, China) Abstract: Effective pre-recognition of human gait is one of the key points to make the dynamic prosthetic knee joint coordinate with the body movement. Acceleration sensor installed in the prosthetic socket and pressure sensor installed in the plantar are used to acquire body’s motion information. According to regularity and repeatability characteristics, hidden Markov model is adopted to analyze the acquired motion information and performing gait pre-recognition. The experiments show that the gait pre-recognition of dynamic lower prosthesis based on hidden Markov model is effective and accurate. Keywords: dynamic lower prosthesis; accelerometer sensor; gait pre-recognition; hidden Markov model 1 引言（Introduction ） 国内从20 世纪80 年代初开始