Adaptive H2 optimal internal model control based on least squares identification

Based on the internal model control structure, the design and analysis method for a discrete-time H ₂ optimal adaptive controller is proposed in this paper. The complicated identification techniques of robustness could be avoided by measuring the noise with dead zone methods of online identification and using feedback filtering method to deal with the robustness problem. In accordance with the Certainty Equivalence principle, a discrete-time H ₂ optimal internal model controller and a discrete-time least squares with covariance resetting adaptive law are combined with the adaptive controller referred in this paper, therefore the adaptive system can work well when the plant is slow time-variant. The research of the system simulation is done while there is error in the system modeling. All the theoretical analysis and simulation results indicate the better control effect of the algorithm.