Robust neural network control of flexible-joint robots

Abstract

A robust neural network (NN) controller is proposed for the motion control of rigid-link flexible-joint (RLFJ) robots. No weak elasticity assumption is needed. The NNs are used to approximate three very complicated nonlinear functions. The authors' NN approach requires no off-line learning phase, no persistent excitation conditions, and no lengthy and tedious preliminary analysis to find a regression matrix. Most importantly, the authors can guarantee the uniformly ultimately bounded (UUB) stability of tracking errors and NN weights. The controller can be regarded as a universal reusable controller because the same controller can be applied to any type of RLFJ robots without any modifications.