Gaussian mixture filtering for data fusion with switching observation models: Application to aircraft relative altimetry

The relative altitude sensing function onboard aircraft is currently supported by the radio-altimeters system along with specific thresholding and voting logics. In order to improve its accuracy and integrity a multi-sensor data fusion is proposed that may involve radio altimeters, GPS information, terrain database, barometer, lidar and inertial measurements. Because these sensors are likely to be affected by multiple malfunctions we are faced to a linear state estimation problem with switching observation models.