A real time neural adaptive trajectorytracking controller is developed for a class of nonlinear systems. Theneural networks are employed for feedforward compensation of unknown systemdynamics. The parameters of neural network are tuned online, in real time,while overall stability is guaranteed.
The neural networks for this controlapplication are Orthonormal Activation Function based Neural Networks (OAFNN). These networks were developed as a part of this research and haveattractive properties for direct adaptive control. The controller designis based on multiple OAFNNs for computationally efficient control.