CORC  > 自动化研究所  > 中国科学院自动化研究所  > 毕业生  > 博士学位论文
题名多约束条件下末端制导律方法研究
作者李庆春
学位类别工学博士
答辩日期2015-05-27
授予单位中国科学院大学
授予地点中国科学院自动化研究所
导师张文生
关键词末端制导 多约束条件 鲁棒制导律 模糊滑模变结构制导律 角度和时间协同制导 terminal guidance multiple constraints robust guidance law fuzzy sliding mode control guidance law impact angle and time cooperative guidance
其他题名Research on Terminal Guidance Law with Multiple Constraints
学位专业控制理论与控制工程
中文摘要现代战场环境对精确制导武器的发展提出了新的要求,复杂环境多约束条件下末端制导控制是研究热点。由于环境的复杂、飞行器运动的非线性、强耦合、多时变等因素,给飞行器末端制导系统的建模、制导律设计带来许多困难。本文对 多约束条件下末端制导律方法展开研究,目标是设计能够满足终端角度、终端加速度、攻击时间等多约束条件的制导律方法,并具有较高的制导精度,具有重要的理论意义和广泛的应用前景。 本文选择角度约束、角度和加速度约束、角度和时间约束作为三个典型的多约束条件,研究多约束条件下末端制导律设计问题。首先,提出了自适应神经模糊滑模变结构制导律方法以及小波神经网络滑模变结构制导律方法,用以解决带角度约束的鲁棒制导律设计问题;其次,提出了模糊滑模变结构制导律方法,用以解决终端角度和加速度同时约束的制导律设计问题;最后,提出了基于改进的偏置比例被动制导律方法和基于领弹与被领弹策略的多弹协同被动制导律方法,用以解决角度和时间协同的被动制导律设计问题。 本文主要工作和贡献如下: 针对带角度约束的鲁棒制导律设计问题,提出了自适应神经模糊滑模变结构(ANFSMC)制导律方法以及小波神经网络滑模变结构(WNNSMC)制导律方法。两种方法的共同之处在于均选取了同一非线性滑模面,可以在有限时间内系统达到收敛状态,并且无需对飞行器与目标的相对运动模型进行简化。不同之处在于,ANFSMC制导律和WNNSMC 制导律分别应用自适应模糊神经推理系统和小波神经网络来自适应调整增益参数,以削弱系统的高频抖振,并增强对不确定性信息的抗干扰性。将两种鲁棒制导律与滑模变结构制导律进行仿真实验对比,仿真结果表明,两种方法在均可以较好的削弱系统抖振,可以满足过载、脱靶量和终端角度的约束条件,能够用于对固定目标和慢速移动目标的精确制导上,而在收敛时间上WNNSMC制导律优于ANFSMC制导律。 针对终端角度和加速度同时约束的制导律设计问题,提出了模糊滑模变结构(FSMC)制导律方法。考虑飞行器自动驾驶仪的动态特性,利用滑模变结构控制方法设计了针对无滞后系统和一阶滞后系统的滑模变结构(SMC)制导律。为增强SMC制导律对复杂环境的适应性和抗干扰性,利用模糊控制技术对制导参数进行实时在线优化,进而提出FSMC制导律方法。将FSMC制导律进行仿真实验分析,验证了该方法针对无滞后系统和一阶滞后系统可以满足过载、脱靶量、终端角度和加速度的约束条件,并与三种国际先进的制导律进行对比,仿真实验结果表明,FSMC制导律在脱靶量和终端加速度上优于其它方法。 针对角度和时间协同的被动制导律设计问题,提出了基于改进的偏置比例被动制导律(IBPPGL)方法和基于领弹与被领弹策略的多弹协同被动制导律(L-FSCPGL)方法。在仅有角测量信息的被动制导条件下,利用伪线性自适应Kalman滤波理论,对目标相对状态信息进行实时估计,形成仅有角测量信息的机动目标跟踪算法。结合目标跟踪算法,在带角度约束的增广比例被动制导律的基础上,通过附加时间反馈控制项,提出角度和时间同时约束的IBPPGL方法。针对多弹角度和时间协同被动制...
英文摘要Due to the modern battlefield environment, the new and challenging demands on the development of precision-guided weapons are proposed, and the terminal guidance and control problem with multiple constraints and complex environments become a hot issue in the world. However, since the complexity of the environments, nonlinearity, strong coupling and time variability of the flight vehicle's movement, it is difficult to do some research on system modeling and designing the guidance laws. In this work, we research on terminal guidance law with multiple constraints such as terminal angle constraint, terminal acceleration constraint, and attack time et al., which has important theoretical significance and broad engineering values. We choose three typical multiple constraints which are impact angle constraint, impact angle and acceleration constraints, impact angle and time constraints to research on the terminal guidance law designing problem. Firstly, to solve the problem of the robust guidance law with impact angle constraint, we propose an adaptive neuro-fuzzy inference system sliding mode control (ANFSMC) guidance law and a wavelet neural network sliding mode control (WNNSMC) guidance law. Secondly, a fuzzy sliding mode control (FSMC) guidance law is introduced to overcome the issue of the guidance law with impact angle and acceleration constraints. Finally, in order to solve the problem of the passive guidance law with impact angle and time cooperative constraints, we put forward an improved biased proportional passive guidance law (IBPPGL) method and a leader-follower strategy cooperative passive guidance law (L-FSCPGL) method for multi-missiles. The details and contributions are: In order to solve the problem of the robust guidance law with impact angle constraint, we propose two robust guidance law methods, namely, the ANFSMC guidance law and the WNNSMC guidance law. Both of two methods choose the same nonlinear sliding surface which can converge to convergent state in finite time and do not need to simplify the movement model between the flight vehicle and the target. The difference between the ANFSMC guidance law and the WNNSMC guidance law is that: they use the adaptive neuro-fuzzy inference system and the wavelet neural network to adjust the guidance gain, respectively. Thus, the system chattering is weakened and robustness is strengthened highly. Experiments compared with sliding mode control (SMC) guidance law illustrate that the p...
语种中文
其他标识符201218014628007
内容类型学位论文
源URL[http://ir.ia.ac.cn/handle/173211/6706]  
专题毕业生_博士学位论文
推荐引用方式
GB/T 7714
李庆春. 多约束条件下末端制导律方法研究[D]. 中国科学院自动化研究所. 中国科学院大学. 2015.
个性服务
查看访问统计
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

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。


©版权所有 ©2017 CSpace - Powered by CSpace