Sliding Mode Control for LPV Systems

LPV models are extremely appealing, as they allow describing the dynamics of many physical systems that are of interest in various engineering applications. For such systems, dedicated control approaches have been proposed, which rely on the measurement of the scheduling variables and exploit such information for improving the closed-loop performance with respect to fixed-structure, possibly robust, solutions. Unfortunately, such control techniques are often not so simple to tune and design, especially when also parametric uncertainties affect the system, thus requiring LPV-robust control techniques. In this work we explore the advantages offered by sliding mode (SM) algorithms for the control of LPV systems, showing that a fixed-structure SM approach can outperform genuine LPV solutions in the case of parametric uncertainties on the system model without additional tuning and design needs. A case study considering the control of lateral vehicle dynamics is used to investigate the performance of the different approaches, showing promising results for extending SM controllers to cope with additional uncertainties affecting LPV systems, that may be difficult to act upon with traditional methods.