Automatica, Vol.36, No.9, 1381-1383, 2000
On the optimality of recursive unbiased state estimation with unknown inputs
For system models with unknown inputs, state estimates may be obtained using linear minimum-variance unbiased estimation. Here, a proof is presented showing that the optimal solution over the class of all linear unbiased estimates may be written in the form of a linear recursive filter, thereby validating previous work in this area.