Parameter Estimation and Model-free Multi-innovation Adaptive Control Algorithms

Parameter estimation is the basis of adaptive control for dynamic systems. This paper surveys and reviews some parameter estimation methods, including the projection algorithms, stochastic gradient (SG) algorithms and multi-innovation stochastic gradient (MISG) algorithms etc. Further, this paper discusses some adaptive control laws and adaptive control algorithms, including the SG self-tuning control algorithms and MISG weighted self-tuning control algorithms etc. Moreover, this paper presents some model-free adaptive control algorithms, including the compact form model-free adaptive control algorithm, SG partial form adaptive control algorithm and MISG partial form adaptive control algorithm etc. The proposed MISG adaptive control algorithm has better tracking performance than the projection and SG adaptive control algorithms because of using the multi-innovation identification theory.