Vision-aided navigation for aircrafts based on road junction detection

In this paper, a vision-aided navigation method for aircraft is proposed. The road network is detected from ground scenes provided by down-looking camera. The road junctions are extracted and referenced. Based on the principle of camera calibration, the position of the aircraft can be estimated by the road junction matching pairs in image and digital map. It is then employed to fix the accumulative error of INS. In order to improve the matching efficiency, the property of road junction and the orientation and the height output of INS are used to reduce the searching space. The simulation results have verified its feasibility.