中国科学院机构知识库网格
Chinese Academy of Sciences Institutional Repositories Grid
基于在线学习误差反传算法的仿真伺服系统设计

文献类型:期刊论文

作者段海滨 ; 王道波 ; 于秀芬 ; 朱家强
刊名中南大学学报:自然科学版
出版日期2005
卷号36期号:2页码:267-271
关键词误差反传算法 BP神经网络 仿真伺服系统 在线学习 无人机
ISSN号1672-7207
其他题名Design of On2l ine Learning Based Error BackPropagation Algorithm in Simulation Servo System
通讯作者北京8701信箱
中文摘要针对无人机(UAV)仿真伺服系统的驱动模型,提出了一种将误差反传算法用于UAV仿真伺服系统在线学习设计的新方案.在该算法中采用了BP神经网络的基本思想,设计了两输入,单隐层,两输出在线学习策略,输入层分别为给定指令信号和反馈数字解算后的位王信号;隐含层单元数为12个;输出层设为2个输出单元,即经在线学习误差反传算法学习训练后的数字位王和速度,其中位王控制器采用自调节比例-积分-微分(PID)控制,速度通过数字/模拟(D/A)转换后传送到速度控制器,设定精度误差指标为0.05,训练样本数为30.用研制的UAV仿真伺服系统对UAV光纤陀螺传感器进行含实物半物理实时仿真实验,结果表明,该在线学习误差反传算法控制方案的UAV仿真伺服系统具有收敛性好,动态响应快,鲁棒性强的特点.
英文摘要Based on t he driving model of unmanned aerial vehicle (UAV) simulation servo system ,a novel scheme based on on-line learning error back propagation algorithm( EBPA) was proposed in designing UAV simulation servo system. The idea of back p ropagation (BP) neural network was adopted in t he proposed algorit hm. The on-line learning st rategy of EBPA wit h two inp ut s ,single hidden layer and two outputs was applied in this scheme. The inp ut layer included given signal and feedback digital position;t he hidden layer had 12 nerve cells;the outp ut layer had two output nerve cells,which were trained digital position and velocity ,and self-tuning proportional-integral2differential ( PID) cont rol scheme was adopted in the position cont roller;the digital/ana-log(D/A)transformed velocity signal was transmitted to velocity cont roller.The resolution error was 0.05,and t he number of training samples was 30.Finally,Hardware-in-loop real-time simulation experiment s for a type of fiber optic gyro were conducted in the newly designed UAV simulation servo system.Simulation result s illust rate that t he UAV simulation servo system using on2line learning based EBPA has good ast ringency.quick response and st rong robust .Key words:error back propagation algorit hm ; back propagation neural network;simulation servo system;on-line learning;unmanned aerial vehicle
学科主题微波遥感
收录类别CSCD
语种中文
CSCD记录号CSCD:1917413
源URL[http://ir.cssar.ac.cn/handle/122/1939]  
专题国家空间科学中心_微波遥感部
推荐引用方式
GB/T 7714
段海滨,王道波,于秀芬,等. 基于在线学习误差反传算法的仿真伺服系统设计[J]. 中南大学学报:自然科学版,2005,36(2):267-271.
APA 段海滨,王道波,于秀芬,&朱家强.(2005).基于在线学习误差反传算法的仿真伺服系统设计.中南大学学报:自然科学版,36(2),267-271.
MLA 段海滨,et al."基于在线学习误差反传算法的仿真伺服系统设计".中南大学学报:自然科学版 36.2(2005):267-271.

入库方式: OAI收割

来源:国家空间科学中心

浏览0
下载0
收藏0
其他版本

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。