The Study On Attack Helicopter Super-maneuverable And Super-nimble Robust Adaptive Flight Control Using Neural Network

The robust adaptive flight control system design with neural network of attack helicopter is studied in this dissertation and super maneuver nimble of attack helicopter is realized. The work covers the control algorithm with neural network, the hardware realization of neural network and the design of distributed 3D animation real-time flight simulation system applied to attack helicopter.The model of attack helicopter is a highly coupled unstable nonlinear system. Especially, the parameters of the model varies sharply when nimbly maneuver is performed. Firstly, the neural network based nonlinear system identification method is studied. A continuous delta-modification grads learning algorithm based on bounded error online recurrent neural network is presented and the convergence of the algorithm is proved strictly. Then the algorithm is successfully applied to the online identification of the parameters of the model of attack helicopter.At the aspect of the research of neural network control theory in nonlinear SISO system, a neural network based robust adaptive control scheme is presented and then applied to the control system design of collection of attack helicopter. After a deep discussion of the method, it is generalized to the nonlinear MIMO system and the convergence of the algorithm is proved strictly.Subsequently, the control loop of attitude angle is designed with dynamic inversion scheme. The controller is designed at two time scales. In the inner loop the uncoupling of different loop is achieved with dynamic inversion scheme while in the outer loop fuzzy-neural network controllers are designed for every channel and the robustness of the controller is improved with sliding mode control. In the inner loop, adaptive online neural network is introduced to compensate the inversion error resulted from inaccurate model of helicopter. The stability of the close loop system is proved strictly.The controller of trajectory tracking is designed on the basis of attitude controller. The simulation of four typical maneuver shows that the maneuverable of attack helicopter is improved by the control system.Lastly, the hard ware realization of neural network is studied and a DSP based method is presented. Then the distributed 3D animation real-time flight simulation system applied to attack helicopter is designed with the technology of LAN, OpenGL and DSP.