Robust Activation Strategy of Scanning Sensors via Sequential Design in Parameter Estimation of Distributed Systems

摘要

We study the problem of determining an activation strategy of discrete scanning sensors so as to maximize the accuracy of parameter estimation of a distributed system defined in a given multidimensional domain. Usually, when designing an identification experiment for nonlinear models, the uncertainty of nominal parameters has to be taken into account. To this end, an approach involving sequential design techniques is proposed, where there is no particular information about the parameter distribution. A scalar measure of the Fisher information matrix is used as the design criterion. The setting examined here corresponds to situations where there are many sensors and only some of them are activated during a given time interval. Additionally, the routine based on the extension of Fedorov's idea of directly constrained design measures is proposed and verified by a computer simulation regarding air pollution process.