Adaptive autopilot design for guided munitions

Traditional autopilot design for guided munitions requires an accurate aero dynamic model and relies on a gain schedule to account for system nonlinear ities. This paper presents an approach that simplifies the autopilot design process by combining an Inverting controller designed at a single flight c ondition with an on-line neural network to account for errors that arise be cause of the approximate inversion. This eliminates the need for an extensi ve design process and also the requirement for accurate aerodynamic data, w hich can be especially critical at high angles of attack or in other regime s at which the aerodynamics become highly nonlinear. The choice of the inve rsion process itself has been found to be critical in the implementation an d is therefore discussed at length. Finally, results from an application of this approach to a full nonlinear six-degree-of-freedom guided munition si mulation are presented.