A Neural Network Parallel Adaptive Controller for Fighter Aircraft Pitch-Rate Tracking

A fighter aircraft pitch-rate command-tracking controller based on a neural network parallel controller is proposed. The scheme consists of an online radial basis function neural network (RBFNN) in parallel with a model reference adaptive controller (MRAC) and uses a growing dynamic RBFNN to augment MRAC. Updating the RBFNN width, the center and weight characteristics are performed such that the error reduction and improved tracking accuracy are accomplished. The RBFNN architecture adapts its centers and radii and tunes the relevant parameters, dynamically addressing the issues related to initial error and dimensional growth inherent in static neural network design. The total control signal is used to change the elevator deflection, keeping the other control surface deflections at random values, even when the aircraft operates at different maneuvers. Moreover, a suitable reference model structure is used for all aircraft operating modes, and the system is then fully tuned by the parallel controller. The strength of the proposed scheme is in its ability to effectively perform, even when plant mode swings and functional changes occur. Theoretical results are validated by conducting simulation studies on a nonlinear F16 fighter aircraft model operating at different modes created by a randomly changing parameter set.