|本期目录/Table of Contents|

[1]王建敏,董小萌,吴云洁.高超声速飞行器 RBF 神经网络滑模变结构控制[J].电机与控制学报,2016,20(05):103-110.[doi:10. 15938 / j. emc. 2016. 05. 015]
 WANG Jian-min DONG Xiao-meng WU Yun-jie.Hypersonic flight vehicle of sliding mode variable structure control based on RBF neural network [J].,2016,20(05):103-110.[doi:10. 15938 / j. emc. 2016. 05. 015]
点击复制

高超声速飞行器 RBF 神经网络滑模变结构控制(PDF)
分享到:

《电机与控制学报》[ISSN:1007-449X/CN:23-1 408/TM]

卷:
20
期数:
2016年05
页码:
103-110
栏目:
出版日期:
2016-05-02

文章信息/Info

Title:
Hypersonic flight vehicle of sliding mode variable structure control based on RBF neural network
作者:
王建敏12 董小萌3 吴云洁14
 ( 1. 北京航空航天大学 自动化科学与电气工程学院,北京 100191; 2. 中国科学院空间应用工程与技术中心,北京 100094; 3. 中国空间技术研究院 钱学森空间技术实验室,北京 100094; 4. 北京航空航天大学 虚拟现实技术与系统国家重点实验室,北京 100191)
Author(s):
WANG Jian-min DONG Xiao-meng WU Yun-jie
 ( 1. School of Automation Science and Electrical Engineering,Beihang University,Beijing 100191,China; 2. Technology and Engineering Center for Space Utilization,Chinese Academy of Sciences,Beijing 100094,China; 3. Qian Xuesen Laboratory of Space Technology,CAST,Beijing 100094,China; 4. State Key Laboratory of Virtual Reality Technology and Systems,Beihang University,Beijing 100191,China)
关键词:
高超声速飞行器 RBF 神经网络 滑模变结构 控制 抖振
Keywords:
hypersonic flight vehicle RBF neural network sliding mode variable structure control chattering
分类号:
V 448
DOI:
10. 15938 / j. emc. 2016. 05. 015
文献标志码:
A
摘要:
针对高超声速飞行器高度非线性及强耦合的特点,提出了一种基于 RBF 神经网络调参的 滑模变结构控制器。滑模变结构控制器能够使高超声速飞行器稳定飞行,但在系统状态到达滑模 面后会产生剧烈的抖振现象,不利于工程应用。RBF 神经网络在一定条件下可以任意精度逼近非 线性函数,且具有较强的自学习、自适应和自组织能力。将 RBF 神经网络与滑模变结构控制相结 合,一定程度上能够消除滑模控制的抖振问题。在高超声速飞行器的巡航状态下,分别加入高度阶 跃指令和速度阶跃指令进行了仿真。仿真结果表明,所设计的 RBF 神经网络滑模变结构控制器使 高超声速飞行器在保证快速性、鲁棒性和抗干扰性的同时,克服了执行机构的抖振问题。
Abstract:
According to hypersonic flight vehicle of highly nonlinear and strong coupling characteristics, sliding mode variable structure control based on RBF neural network regulating parameters was proposed. Sliding mode variable structure controller makes the hypersonic flight vehicle stably fly,but when the sys-tem states arrived at the sliding mode surface,it will emerge severe chattering,which would influence en-gineering applications. RBF neural networks can approximate nonlinear functions in arbitrary precision un-der certain conditions,in addition it has capacity of strong self-learning,adaptive and self-organizing. The controller that together RBF neural network with sliding mode variable structure can eliminate chattering problem generated by sliding mode variable structure control to a certain extent. Simulation was conducted by giving altitude and velocity command on the cruise condition of hypersonic flight vehicles. Simulation results show that RBF neural network based sliding mode variable structure controller designed here en-sures rapidity,robustness and immunity of the hypersonic flight vehicle,while overcoming the problems of
actuator chattering.

参考文献/References:

相似文献/References:

[1]黄显林,逢洪军. 考虑气动弹性的高超声速飞行器姿态稳定控制[J].电机与控制学报,2014,18(11):97.
 HUANG Xian-lin,PANG Hong-jun. Attitude stabilization control of hypersonic vehicleconsidering aeroelasticity[J].,2014,18(05):97.
[2]王鹏飞,王洁,时建明,等.高超声速飞行器预设性能反演鲁棒控制[J].电机与控制学报,2017,21(02):94.[doi:10.15938/j.emc.2017.02.012]
 WANG Peng-fei,WANGJie SHI Jian-ming,LUO Chang.Prescribed performance back-stepping robustness control of a flexible hypersonic vehicle [J].,2017,21(05):94.[doi:10.15938/j.emc.2017.02.012]

备注/Memo

备注/Memo:
收稿日期: 2014 - 04 - 08
基金项目: 国家自然科学基金( 91216304)
作者简介: 王建敏( 1986—) ,男,博士,研究方向为伺服控制、导弹制导控制、高超声速飞行器制导控制;
董小萌( 1978—) ,男,高级工程师,研究方向为目标跟踪、飞行器控制与建模;
吴云洁( 1969—) ,女,博士,教授,博士生导师,研究方向为智能控制理论、半实物仿真设备及工业过程控制等。
更新日期/Last Update: 2016-06-26