关键词：光机电；传感器；调度；设计
摘 要：There is no doubt that preparation of experimental conditions in the sense of choosing a suitable observational strategy has great influence on the expected accuracy of parameter estimates, especially in the context of DPSs. The first part of the chapter contains a brief overview of the parameter identification problem and optimal experimental design for linear and nonlinear lumped dynamic systems. In particular, the notion of continuous designs is intro-duced, which dramatically reduces the problem dimensionality. Furthermore, generalization of some classical results from experimental design theory to multi-response systems with possibly correlated outputs is presented for the purpose of providing fundamental results which characterize the solutions and thus indicate possible numerical procedures. The inherent nonlinearity of DPSs with respect to the estimated parameters is certainly a problem of paramount importance and one of main complications in the delineated approach. However, under some assumptions, the results derived for systems linear in parameters can be extended to the class of nonlinear models based on suitable approximations of the FIM. Such a generalized classical theory paves the way to numerous applications of DPSs and, being crucial for specific problems, constitutes close connection with the chapters which follow.