INS/GNSS/Fiducial Marker Sensor Fusion for UAM Aircraft Navigation

This paper addresses the sensor fusion of INS, GNSS, and Fiducial Markers for navigation in the Urban Air Mobility environment. The Error-State Kalman filter (ESKF) was adopted for the estimation of position, velocity, attitude, and biases, considering ideal and non-linear GNSS sensors in their tightly and loosely coupled forms. The main contribution of this work is the ESKF formulation for INS/GNSS/Fiducial Marker sensor fusion and validation with synthetic image simulation in the Unreal Engine integrated with Simulink. The Fiducial Marker fusion shows improvements in filter accuracy and corrects the filter when GNSS is not available.