摘要：AbstractThe main contribution of this paper consists in developing a procedure for optimal input design based on the determinant of an interval Gram matrix of sensitivities to improve parameter estimation in a bounded-error context. In this context, the measurement errors are supposed to be bounded, with no other hypothesis on their distribution, contrary to the classical statistical approach. The parameters to be estimated are supposed belong to some prior domains. The developped procedure has been successfully tested on a pharmacokinetic example which highlights its potential for a large class of models.
