Adaptive critic based neurocontroller for autolanding of aircraft with varying glideslopes

In this paper, adaptive critic based neural networks have been used to design a controller for a benchmark problem in aircraft autolanding. The adaptive critic control methodology comprises successive adaptations of two neural networks, namely `action´ and `critic´ networks until closed loop optimal control is achieved. The autolanding problem deals with longitudinal dynamics of an aircraft which is to be landed in a specified touchdown region in the presence of wind disturbances and gusts using elevator deflection as the control for glideslope and flare modes. The performance of the neurocontroller is compared to that of a conventional PID controller. Neurocontroller´s capabilities are further explored by making it more generic and versatile in the sense that the glideslope angle can be changed at will during the landing process. Flight paths (trajectories) obtained for a wide range of glideslope angles in presence of wind gusts are compared with the optimal flight paths which are obtained by solving the linear quadratic regulator formulation using conventional optimal control theory