Research On The Modeling And Control To A Quadrotor Helicopter Simulator

Quadrotor is a kind of simple rotorcraft drived by four propellers, which could vertically take-off and land. The modern quadrotors are mostly minimized Unmanned Aerial Vehicle (UAV) for both military and civilian usages. It is also a typical complex control system with the feature of underactuation, instability, multivariable, uncertain nonlinearity and strong coupling. Modeling and control to a quadrotor is a hot topic in recent controlling field.Based on a summary of the research status quo, the key technologies and the future applications of the quadrotor, this paper concentrates on its special characteristics, mainly researched the problems on modeling and the ways of controlling. Some important theoretical analysis and research results are as follows:Firstly, the physical model of the four-rotor flight simulator are established by dynamics equations. Then comes the design of a controller by LQR method. According to the state equation, selecting proper weight coefficient matrixes, the system's state feedback is constructed, with which the control of the quadrotor is achieved. The validity of the controller is proved by simulations and real-time controllings, then the input and output data of the system is gotten under this controlling.Secondly, the RBF-ARX modeling theory as a global model is adopted for the nonlinearity of quadrotor. The structure, the identification and parameter optimization based on RBF-ARX model are discussed strongly. The predictive outputs and modeling errors between the RBF-ARX model and a global linear ARX model are compared respectively. The results show the superiority of the RBF-ARX model for the modeling of nonlinear systems.Finally, adopt a MIMO RBF-ARX model to represent the nonlinear dynamics of a quadrotor, and then based on a locally linearized ARX model that is obtained from the RBF-ARX model at a working-point, the state-feedback control law with LQR approach is proposed to control the quadrotor. The validity of the controller is proved by simulations. From thr results we conclude the LQR controller works well. It not only can get to any legal gesture, but also stabilizes quickly.