高超声速飞行器多速率神经高度控制  

Multi-rate neural altitude control for hypersonic vehicle via flight path angle tracking

作　　者：王士星[1,2] 孙富春[1] 田亮[3] 

机构地区：[1]清华大学计算机科学与技术系,北京100084 [2]海军航空工程学院控制工程系,山东烟台264001 [3]河北师范大学数学与信息科学学院,河北石家庄050024

出　　处：《中南大学学报（自然科学版）》2013年第S1期47-51,共5页Journal of Central South University:Science and Technology

基　　金：国家自然科学基金资助项目(61004002);航空科学基金资助项目(20115184005)

摘　　要：针对高超声速飞行器的纵向动力学的特点,对于快变动力学的高度子系统,通过变换将高度指令转换为航迹角跟踪问题,虚拟控制量通过标称量反馈并采用神经网络逼近不确定项。速度子系统被看作慢变动力学,为有效利用发动机信息,速度子系统应用了不同于高度子系统的采样速率。研究结果表明:控制器能够保证所有相关信号半全局最终一致有界;仿真结果验证了算法的有效性。The altitude control was analyzed for the longitudinal dynamics of a generic hypersonic flight vehicle(HFV).By transforming altitude command into the tracking of flight path angle with fast dynamics,the system design was focused on the control of the attitude subsystem.The virtual control was designed with nominal feedback and neural network(NN) approximation via back-stepping.To use the information of throttle setting,the multi-rate sampling method was employed for the two subsystems where the velocity subsystem was considered as slow dynamics.Under the proposed controller,the semiglobal uniform ultimate boundedness(SGUUB) stability was guaranteed.The slow dynamics were transformed into the pa-rameter estimation problem and the update law was designed.The simulation results show the effectiveness of the pro-posed control approach.

关 键 词：高超声速飞行器 神经网络 多速率采样 离散控制 

分 类 号：V448[航空宇航科学与技术—飞行器设计]