重复使用助推飞行器大调姿段控制器设计  

Controller Design of the RBV Large Attitude Adjusting Phase

作　　者：许江涛[1] 陈阳阳[1] 吴限德 王佩臣[2] 

机构地区：[1]哈尔滨工程大学航天工程系,黑龙江哈尔滨150001 [2]黑龙江工程学院数学系,黑龙江哈尔滨150001

出　　处：《控制工程》2013年第3期493-496,共4页Control Engineering of China

基　　金：中国博士后科学基金(20110491028);黑龙江省青年自然科学基金(QC2011C020);中央高校基金科研业务费(HEUCFZ1126

摘　　要：针对常规飞行控制方法难以适用于重复使用助推飞行器(RBV)大调姿转弯飞行段强非线性、强耦合、大角度的问题,提出一种基于神经网络鲁棒自适应非线性动态逆的重复使用助推飞行器飞行控制策略。首先,充分考虑气动力系数随马赫数、攻角及侧滑角等变化的非线性特点,建立完整的重复使用助推飞行器非线性六自由度模型;其次,利用非线性动态逆控制实现重复使用助推飞行器系统的伪线性解耦;然后,利用单隐层神经网络来逼近系统模型不确定性所带来的动态逆误差,最后,利用鲁棒自适应控制来抑制外加干扰以及网络逼近误差。非线性仿真验证了该控制策略在大调姿转弯段的可行性和有效性,同时控制策略的设计方法物理概念清晰,易于工程实现。For conventional flight control strategy cannot adapt to the large attitude adjusting phase of RBV with strong nonlinearity, strong coupled and large angle, the flight control strategy of RBV based on the neural network robust adaptive inverse is proposed. Firstly, considering the nonlinear characteristics of aerodynamic coefficients varying with mach numbers, attack angle and sideslip angle, the complete nonlinear 6-DOF model of RBV is established. Secondly, using the nonlinear dynamic inverse control strategy achieves the pseudo linear decoupling of RBV. And then, using the neural network with single hidden layer approximates the dynamic inverse error for system model uncertainty. Finally, the external disturbance and network approximating error is suppressed by robust adaptive control. The nonlinear simulation on the large attitude adjusting phase verifies the feasibility and validity of this control strategy. Meanwhile, the physical concept of the design method is clear and the control strategy is realized in engineering conveniently.

关 键 词：重复使用助推飞行器 大调姿转弯段 控制策略 鲁棒自适应 非线性动态逆 

分 类 号：TP273[自动化与计算机技术—检测技术与自动化装置]