基于系统辨识的小型无人直升机建模及控制器设计

发布时间：2018-08-29 20:29

【摘要】：小型无人直升机因其具有随时起降、低空飞行等优异性能使其在军事和民用上都被广泛应用。但由于直升机是一个内部结构复杂、非线性、强耦合的飞行器,在飞行过程中极易受到温度变化、风力变化等环境因素的干扰,因而开发一套理想的无人直升机飞控系统具有诸多难题。首先,直升机本身具有的非线性以及高度复杂性,使其很难建立一套完整的机理模型。其次,直升机本身的不稳定性导致对输入的反应非常敏感,需要加入高频的反馈信号使其稳定。再次,直升机在飞行过程中横纵向耦合以及高度与偏航间耦合严重,这增加了控制器设计的难度。最后直升机飞行过程中存在高危险,这影响飞行实验的安全。针对以上的小型无人直升机的特点,本文首先以直升机飞行力学为基础,建立小型无人直升机的机理模型。深度分析了小型无人直升机内部的力学关系及运动方程,得到小型无人直升机的各个输入输出通道的传递函数原始结构模型,然后基于CIFER系统辨识建模,以建立的传递函数原始结构模型为基础,提出了一种改进的系统辨识方法,将最小二乘法应用到搜索状态空间模型最小代价函数中,加快了代价函数收敛的速度。通过仿真对比,改进的辨识方法进一步提升了系统辨识的精度,从而建立得到了具有良好跟踪效果的小型无人直升机传递函数模型。本文利用自抗扰控制方法来建立小型无人直升机的控制系统,实现了对无人直升机在俯仰、偏航、滚转、高度四个通道的控制器设计,从而克服了飞行试验中的各种扰动因素,使辨识得到的模型能够更好的运用于工程实践中。本文使用的飞控仿真平台是MATLAB中的SIMULINK,利用该平台对建立的传递函数结构模型进行了仿真测试,在模型中分别加入了自抗扰控制器以及传统的PID控制器,对两个控制器的抗干扰效果进行了对比。仿真结果表明所设计的自抗扰控制器在抗干扰效果上要优于传统的PID控制器,从而证明自抗扰控制器具有很高的实用性与有效性。
[Abstract]:Small unmanned helicopter is widely used in military and civilian because of its excellent performance such as taking off and landing at any time and flying at low altitude. However, the helicopter is a complex, nonlinear and strongly coupled aircraft, which is easily disturbed by environmental factors such as temperature change, wind force change and so on. Therefore, it is difficult to develop an ideal unmanned helicopter flight control system. Firstly, the nonlinear and high complexity of the helicopter makes it difficult to establish a complete mechanism model. Secondly, the instability of the helicopter itself leads to a very sensitive response to the input, and high frequency feedback signals are needed to stabilize the helicopter. Thirdly, the horizontal and longitudinal coupling and the coupling between altitude and yaw are very serious, which increases the difficulty of controller design. Finally, there is a high risk in the helicopter flight, which affects the safety of the flight experiment. In view of the characteristics of the above small unmanned helicopter, the mechanism model of the small unmanned helicopter is established on the basis of helicopter flight mechanics. The mechanical relations and motion equations of the small unmanned helicopter are deeply analyzed, and the original structure model of the transfer function of each input and output channel of the small unmanned helicopter is obtained, and then the model is established based on CIFER system identification. Based on the original structure model of the transfer function, an improved system identification method is proposed. The least square method is applied to the minimum cost function of the search state space model, which accelerates the convergence of the cost function. The simulation results show that the improved identification method can further improve the accuracy of the system identification and thus establish the transfer function model of the small unmanned helicopter with good tracking effect. In this paper, the control system of small unmanned helicopter is established by using the ADRC method, and the controller design of pitch, yaw, roll and altitude of unmanned helicopter is realized, which overcomes all kinds of disturbance factors in flight test. The identified model can be better used in engineering practice. The flight control simulation platform used in this paper is SIMULINK, in MATLAB. The transfer function structure model is simulated and tested by using the platform. The ADRC controller and the traditional PID controller are added to the model, respectively. The anti-interference effect of the two controllers is compared. The simulation results show that the designed ADRC is superior to the traditional PID controller in anti-jamming effect, which proves that the ADRC has high practicability and effectiveness.
【学位授予单位】：南昌航空大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：V279;V249.1

【参考文献】