Control of pitch angle of a flexible launch vehicle with model predictive control approach using neural networks

In this paper, we investigate the problem of model predictive control of pitch angle of a flexible launch vehicle using neural networks. We simulate motion of launch vehicle on the pitch plane by considering effects of flexibility. Aeroelstic effects are also modeled by adding elastic terms of the structure to state space control equations which completes the vehicle model. To control the vehicle, we use a model predictive controller which is based on a neural network. Basic notions of this controlling approach are presented and we explain how a neural network can be used for this purpose. Experimental results on synthetic data in Simulink environment demonstrates our approach is useful and effective and neural networks can be used to control a vehicle in a nonlinear, adaptive control setting.