Identification and control of Sandwich system with backlash

The identification and control of Sandwich systems are surveyed for the systems which involve backlash nonlinearity. The focus is made on the way of constructing the identification model and the convergence rate of the identification algorithm. Based on a parameterization of backlash, the Sandwich system with backlash is constructed as a regression model in which the backlash parameter and linear subsystems parameters are separated by using the key-term separation principle. In order to improve the convergence rate, a multi-innovation least squares algorithm is applied to identify the parameter of the regression equation by modifying the scalar single innovation as multi-innovation matrix. Finally, a traditional PID controller achieves a reference signal tracking based on the result of the parameter estimation.