SIA OpenIR  > 机器人学研究室
基于遥控数据学习的小型直升机控制方法研究
Alternative TitleRemote control learning based small-scale helicopter control
赵新刚; 伊嘉瑞; 张婵; 韩建达
Department机器人学研究室
Source Publication仪器仪表学报
ISSN0254-3087
2010
Volume31Issue:S2Pages:72-76
Indexed ByEI
EI Accession number20111113746619
Contribution Rank1
Funding Organization国家自然科学基金(No.60705028);辽宁省自然科学基金;辽宁省博士启动基金资助
Keyword遥控学习 神经网络 直升机控制
Abstract本文提出了利用神经网络学习人手遥控直升机特点的控制方法,针对单一网络进行训练时存在权值相互影响而造成的控制效果不稳定的情况,结合直升机的固有特点,提出了4神经网络的网络结构,降低了训练数据的干扰,改善了控制效果。直升机悬停控制仿真结果表明,该方法在可以获得更小的悬停位置偏差,更平稳的直升机姿态变化。  
Other AbstractHuman Romote based control method which is realized by neural network is proposed. When training a single network, there is interaction between the weight value which may effect instability in helicopter control. Combined with the inherent characteristics of the helicopter, a kind of Four Neural Network structure, which reducing the interference of training data to improve the control effect, is proposed. Helicopter hover control simulation results show that the method can get smaller position error, a more stable helicopter attitude changes.
Language中文
Document Type期刊论文
Identifierhttp://ir.sia.cn/handle/173321/7216
Collection机器人学研究室
Affiliation1.中国科学院沈阳自动化研究所机器人学国家重点实验室
2.沈阳建筑大学
Recommended Citation
GB/T 7714
赵新刚,伊嘉瑞,张婵,等. 基于遥控数据学习的小型直升机控制方法研究[J]. 仪器仪表学报,2010,31(S2):72-76.
APA 赵新刚,伊嘉瑞,张婵,&韩建达.(2010).基于遥控数据学习的小型直升机控制方法研究.仪器仪表学报,31(S2),72-76.
MLA 赵新刚,et al."基于遥控数据学习的小型直升机控制方法研究".仪器仪表学报 31.S2(2010):72-76.
Files in This Item:
File Name/Size DocType Version Access License
ZWQK000916.pdf(525KB)期刊论文出版稿开放获取ODC PDDLView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[赵新刚]'s Articles
[伊嘉瑞]'s Articles
[张婵]'s Articles
Baidu academic
Similar articles in Baidu academic
[赵新刚]'s Articles
[伊嘉瑞]'s Articles
[张婵]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[赵新刚]'s Articles
[伊嘉瑞]'s Articles
[张婵]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: ZWQK000916.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.