Sliding mode neurocontrol with applications

In this study the tracking problem for a class of nonlinear uncertain systems is tackled. A new sliding mode neurocontroller is suggested to solve this problem. The designing of this controller includes the construction of online state estimates and the corresponding tracking control based on sliding mode approach using obtained state estimates. We apply a special sliding mode technique during the "offline training" to estimate the right-hand side of the given dynamics in finite-time and then to use these estimates for the best (in LQ-sense) nominal weights selection in the designed neuro observer. A switching (sign) type term is incorporated in to the observer structure to correct the current state estimates using only available and on-line measurable output data supplied with a new learning procedure with a relay term. The illustrative example dealing with a real water ozonation process is presented.