Simulator based adaptive helicopter training using neural networks

Presents an approach to the application of artificial neural networks in adaptive simulator-based helicopter training of novice student pilots. The theory that an experienced helicopter pilot performs in some optimal fashion was used as the underlying basis for the adaptive hover trainer design. This development was based on the hypothesis that a novice can be trained to fly a helicopter automatically if the helicopter system adapts to the learning curve of the student. Based on the hypothesis of adaptive training, a neural network synthesis procedure is developed. The neural network synthesis provides student performance monitoring and helicopter system adaptation using an adaptive neurocontroller