State estimation is difficult when the system has multiple modes of operati
on. Modal transitions create discontinuities in the reference point for the
local state variables. The uncertain reference point increases the ambigui
ty in the state measurement. This paper presents an estimation algorithm th
at can be used in multimodal applications. The algorithm is shown to be sup
erior to the Kalman filter when the state measurement is contaminated with
a mode dependent offset. Despite the uncertain reference point in the obser
vation, good estimates of the underlying entire state processes can be gene
rated.