Chattering-Free LS-SVM Sliding Mode Control for a Class of Uncertain Discrete Time-Delay Systems with Input Saturation

Least squares support vector machine (LS-SVM) classifiers are a class of kernel methods whose solution follows a set of linear equation. In this work, a least squares support vector machine sliding mode control (LS-SVM-SMC) strategy is presented for uncertain discrete time-delay system with input saturation. An equivalent matrix is constructed for input saturation condition in the scheme. Output of LS-SVM is used for replacing sign function of the reaching law in traditional sliding mode control (SMC). Combined LS-SVM-SMC with linear Matrix Inequalities (LMIs), a chattering free SMC algorithm is applied in uncertain discrete time-delay systems with input saturation. The feasibility and effectiveness of the LS-SVM-SMC scheme are demonstrated via numerical examples. As a result, compared with conventional SMC, LS-SVM-SMC is able to achieve the desire transient response with input saturation. And there is no chattering in steady state while unmatched parameter uncertainty exists.