Unknown piecewise constant parameters identification with exponential rate of convergence

The scope of this research is the identification unknown piecewise constant parameters linear regression equation under finite excitation condition. Compared to known methods, make computational burden lower, only one model identify all switching states used in developed procedure with following two-fold contribution. First all, we propose a new truly online estimation algorithm based on well-known DREM approach detect time and preserve alertness adjustable detection delay. Second, despite fact that signal function unknown, adaptive law derived provides global exponential convergence estimates their true values case regressor finitely exciting somewhere inside interval between two consecutive switches. robustness proposed influence external disturbances analytically proved. Its effectiveness demonstrated via numerical experiments, which both abstract regressions second-order plant are used.