针对有翼导弹含有未知气动参数时的飞行控制问题,设计了一种带有自适应参数近似的块控反步飞行控制器。分析了块控反步控制器设计的假设条件,并放宽了对指令信号和输入矩阵的部分限制。将导弹动态模型中的未知气动参数转换为待估计参数矩阵,采用自适应参数近似律对未知参数矩阵进行估计,相比使用神经网络逼近未知函数更为简单,易于实现。将滤波器引入控制器设计中,既避免了反复求导虚拟控制律产生的"项数膨胀"问题,又降低了控制器对指令信号的要求。对控制系统跟踪误差动态和Lyapunov稳定性的分析表明系统是稳定且指数收敛的。在某型有翼导弹模型上进行了6自由度(DOF)飞行仿真对比,结果显示所设计的控制器具有良好的指令跟踪能力和较强的鲁棒性。
A flight controller using adaptive parameter approximation block backstepping is designed for a winged missile with unknown aerodynamic parameters.The assumptions for the controller design are analyzed.Moreover,some less stringent assumptions about commands and input matrixes are given.The unknown parameter matrixes,which are formed from the unknown aerodynamic parameters in the missile dynamic model,are estimated by an adaptive parameter approximation method.This method is less complex and easier to implement than neural network approximation.The filters are introduced into the process of controller design,which are used to overcome the "term explosion" problem caused by differentiations of the virtual control law and to reduce the restrictions of commands.The closed-loop system is proved to be stable and converge exponentially through the analysis of tracking errors dynamic and Lyapunov stability.Furthermore,a nonlinear six degree-of-freedom(DOF) flight is simulated on the winged missile model,and the results demonstrate good tracking performance and robustness of the designed flight controller.