查看文章

This paper is about bounded-error state estimation for models described by a continuous-time state equation from discrete-time measurements. A guaranteed solution to this problem is proposed, based on Müller's theorems and interval analysis. This technique allows, at any time instant, the characterization of the set of all state vectors that are consistent with the error bounds, measurements and model structure. Joint state and parameter estimation is also possible. The resulting methodology is illustrated on the estimation of the state of a compartmental model.