A Precise Neural-Disturbance Learning Controller of Constrained Robotic Manipulators

An adaptive robust controller is introduced for high-precision tracking control problems of robotic manipulators with output constraints. A nonlinear function employed to transform the constrained objective new free variables that are then synthesized using a sliding-mode-like as an indirect mission. signal derived ensure boundedness main without violation physical The performance improved by adopting neural-network model conditioned learning laws deal uncertainties and disturbances inside system dynamics. disturbance-observer-based additionally properly injected into neural eliminate approximation error achieving asymptotically accuracy. Performance overall validated intensive theoretical proofs comparative simulation results.