Deadzone compensation in motion control systems using neural networks

A compensation scheme is presented for general nonlinear actuator deadzones of unknown width. The compensator uses two neural networks (NN's), one to estimate the unknown deadzone and another to provide adaptive compensation in the feedforward path, The compensator NN has a special augmented form co ntaining extra neurons whose activation functions provide a "jump function basis set" for approximating piecewise continuous functions. Rigorous proof s of closed-loop stability for the deadzone compensator are provided and yi eld tuning algorithms for the weights of the two NN's, The technique provid es a general procedure for using NN's to determine the preinverse of an unk nown right-invertible function.