Model-free tuning solution for sliding mode control of servo systems

This paper suggests a model-free tuning solution for a sliding mode control system (SMCS) structure dedicated to servo systems. The new SMCS structure is viewed in the framework of reference trajectory tracking using a first-order nonlinear dynamic system as a local approximation of the process model. The sliding mode control signal augments the control signal specific to a model-free PI control system (CS) structure in order to compensate for the estimation errors which affect the systematic design and performance. The derivatives in the local approximation of the process model are estimated numerically using a Savitzky-Golay filter to carry out both differentiation and smoothing. A simple design approach is proposed for the SMCS structure. The real-time experimental results concerning the speed control of a laboratory nonlinear DC servo system prove the performance improvement of the SMCS structure against a model-free PI CS structure.