摘要
实现远程操作机器人和人类操作者的同步是一项有意义且具有挑战性的任务,其中需要克服的关键问题之一是消除人类手部的生理性震颤。为了解决这个问题,使用两种机器学习方法对人手震颤进行预测以及消除,并且对两种机器学习方法进行性能比较。对震颤消除器的数学模型进行分析并确定模型的输入及输出,建立了一个基于机器学习的消颤器。通过仿真实验,得到两种机器学习方法的性能评估指标,最终确定基于宽度-增量学习系统的消颤器更加有效且性能理想。仿真实验表明:基于宽度-增量学习系统的消颤器展现了很好的回归能力、泛化能力和快速学习能力。
Achieving synchronization between a teleoperated robot and a human operator is a rewarding but challenging task,and one of the key problems to overcome is the elimination of physiological tremors in the human hand.To address this problem,we used two machine learning methods for predicting and eliminating human hand tremor,and then compared the performance of these two methods.Through analyzing the mathematical model of the tremor eliminator and determining the inputs and outputs of the model,we built a defibrillator based on machine learning.Through simulation experiments,the performance evaluation indexes of the two machine learning methods were obtained,and the more effective and ideal performance of the defibrillator based on the broad⁃increment learning system was finally determined.The simulation experiments show that the defibrillator based on the width⁃incremental learning system has good regression ability,generalization ability and fast learning speed.
作者
刘炜桢
赖冠宇
杨伟钧
陈俞强
LIU Wei-zhen;LAI Guan-yu;YANG Wei-jun;CHEN Yu-qiang(School of Automation,Guangdong University of Technology,Guangzhou 510006,China;School of Mechanical and Electrical Engineering,Guangzhou City Polytechnic,Guangzhou 510405,China;School of artificial Intelligence,Dongguan Polytechnic,Dongguan 523186,China)
出处
《广州城市职业学院学报》
2023年第1期88-95,共8页
Journal Of Guangzhou City Polytechnic
基金
广东省教育科学规划课题“校企深度融合下高职工科双创人才培养模式研究”(编号:2021GXJK673)
广东省普通高校重点领域专项“新一代信息技术”“经济型便携式精确测温仪的研究与应用”(编号:2021ZDZX1109)
东莞市社会发展重点项目“基于高精度北斗定位及物联网技术的农产品供应链系统研究”(编号:2020507156156)
东莞市科技特派员项目“基于高精度北斗定位车联网终端系统研究”(编号:20211800500182)。
关键词
遥操作机器人系统
宽度学习系统
增量学习
支持向量机
震颤消除器
teleoperated robotic systems
broad learning systems
incremental learning
support vector machines
tremor eliminators
