摘要
考虑高超声速飞行器再入过程中存在气动舵低效的问题,提出了质量矩/气动舵复合控制方式,研究了这两类执行机构的复合控制分配问题,并针对高超声速飞行器强非线性和不确定性的对象特性,基于神经网络方法设计了自适应动态逆姿态控制系统。首先给出了质量块配置原则以及质量矩/气动舵复合控制模型;其次,为获得良好的控制分配精度并保证较小的执行机构能耗,基于二次规划方法设计了质量矩/气动舵复合控制分配策略;再次,利用神经网络权值的自适应调整来逼近系统中存在的不确定性,补偿动态逆误差,设计了基于神经网络的自适应动态逆控制器。最后,通过仿真验证了文中控制分配策略和自适应动态逆方法的有效性。
Considering the low efficiency of aerodynamic fins for the reentry hypersonic vehicle, the compound control actuators with moving masses and aerodynamic fins are introduced and the control allocation problem between them is researched. Besides, against strong nonlinearity and uncertainty of the vehicle, the adaptive dynamic inverse attitude control system based on the neural network （NN） is designed. Firstly, the principle of the configuration of masses and the compound control-oriented model are given. Secondly, in order to obtain a good control allocation accuracy and low energy consumption of actuators, a control allocation strategy is provided on the basis of the quadratic programming method. Thirdly, to approximate the system uncertainty and compensate the dynamic inversion error, the nonlinear dynamic inversion attitude control system based on NN with weights updating is designed. Finally, the simulation results show the effectiveness of the control allocation strategy and the adaptive dynamic inverse method in their application of attitude control of the hypersonic vehicle.
出处
《宇航学报》
EI
CAS
CSCD
北大核心
2013年第7期955-962,共8页
Journal of Astronautics
基金
中央高校基本科研业务费专项资金(HIT.NSRIF 201163)
航天科技创新基金(CASC2011021)
航空科学基金(20110777006)
航天支撑技术基金(2013-HT-HGD-15)
国家自然科学基金(61174202)
关键词
高超声速飞行器
质量矩
气动舵
姿态控制
控制分配
神经网络
动态逆
Hypersonic vehicle
Moving mass/aerodynamic fin
Attitude control
Control allocation
Neuralnetwork
Dynamic inversion
