Low sensitivity filters for state estimation in the presence of large parameter uncertainties

Three minimax sensitivity criteria are used to synthesize filters for estimating the state of a simple first-order plant when large uncertainties in dynamic and/or statistical parameters of the plant and measurement system are present. Two design examples involving unknown plant bandwidth and unknown plant input noise covariance are worked out in detail. Graphical comparison of the error performance of the three minimax sensitivity filters with that of the optimal filter is presented for each case.