ROBUST NEURAL-NETWORK CONTROL OF RIGID-LINK ELECTRICALLY DRIVEN ROBOTS

A robust neural-network (NN) controller is proposed for the motion con trol of rigid-link electrically driven (RLED) robots. Two-layer NN's a re used to approximate two very complicated nonlinear functions. The m ain advantage of our approach is that the NN weights are tuned on-line , with no off-line learning phase required. Most importantly, we can g uarantee the uniformly ultimately bounded (WB) stability of tracking e rrors and NN weights. When compared with standard adaptive robot contr ollers, we do not require lengthy and tedious preliminary analysis to determine a regression matrix, The controller can be regarded as a uni versal reusable controller because the same controller can be applied to any type of RLED robots without any modifications.