熊鹏文,林虹,宋爱国,胡凌燕,陈大鹏.基于随机森林回归的手臂末端力的软测量方法[J].仪器仪表学报,2017,38(10):2400-2406
基于随机森林回归的手臂末端力的软测量方法
Soft measurement method of end of arm force based on random forest regression
  
DOI:
中文关键词:  随机森林  软测量  手臂末端力  集成学习
英文关键词:random forest  soft measurement  end of arm force  ensemble learning
基金项目:国家自然科学基金(61663027, 81501560, 61563035)、国家杰出青年科学基金(61325018)项目资助
作者单位
熊鹏文 1.南昌大学信息工程学院南昌330031;2. 东南大学仪器科学与工程学院南京210096 
林虹 国网江西省电力公司信息通信分公司南昌330077 
宋爱国 东南大学仪器科学与工程学院南京210096 
胡凌燕 南昌大学信息工程学院南昌330031 
陈大鹏 2. 东南大学仪器科学与工程学院南京210096;4. 北卡罗来纳大学夏洛特分校计算机工程系夏洛特28223 
AuthorInstitution
Xiong Pengwen 1. School of Information Engineering, Nanchang University, Nanchang 330031, China; 2. School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China 
Lin Hong Information and Communication Branch, State Grid Jiangxi Electric Power Company, Nanchang 330077, China 
Song Aiguo School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China 
Hu Lingyan School of Information Engineering, Nanchang University, Nanchang 330031, China 
Chen Dapeng 2.School of Instrument Science and Engineering, Southeast University, Nanjing 210096, China; 4. Department of Computer Science, University of North Carolina at Charlotte, Charlotte 28223, USA 
摘要点击次数: 2051
全文下载次数: 1673
中文摘要:
      针对手臂康复训练后仍缺乏准确力觉的康复病人提出了一种手臂末端力的软测量方法。采用肌电信号(EMG)传感器与手臂姿态传感器获取的数据综合描述手臂的综合状态信息,并作为随机森林回归的输入,将手臂末端力作为随机森林回归的输出。依据康复训练的基本动作单元,针对性的设计了“推拉”和“提放”两组试验,在离线状态下,利用力传感器测量得到的实际末端力与手臂的综合状态信息作为样本集,并通过大量样本数据训练随机森林回归算子得到稳定可靠的回归算子,最后通过在线预测手臂末端力与真实末端力输出的比较,验证了该方法的有效性。
英文摘要:
      In order to make the arm rehabilitation patients accurately feel the end of arm force, a soft measurement method of end of arm force is proposed based on random forest regression. The data from the electromyography (EMG) sensors and the arm gesture sensors are integrated to describe the comprehensive feature information of the arm. The comprehensive feature information is used as the input of Random Forest Regression with the end of arm force as the output. According to the basic actions of arm rehabilitation, two groups of experiments ‘push pull’ and ‘lift drop’ are specifically designed. The Random Forest Regression is trained offline with the comprehensive feature information and the end of arm force measured with a force sensor, then prediction of end of arm force is conducted and compared with the real force. The method is verified by the experimental results and the effectiveness of the method is proved.
查看全文  查看/发表评论  下载PDF阅读器