OPTICAL NAVIGATION FILTER BASED ON ROAD INFORMATION

Kurzfassung

The present paper shows an approach to measure and fuse optical orientation data of sensors during an airplane flight for traffic data acquisition. This orientation data is necessary for geo-registration of any aerial images and extracted features. Since an inertial measurement unit (IMU) is a very costly component, alternative methods for orientation determination should be considered. This can be done by fusion of different optical information supported by a GPS sensor and a road network database with additional elevation information. It is shown that the extracted roads are matched with given database segments. To adjust the rotation angles a simplex optimization algorithm is used. In addition, an interest point tracking has been implemented to determine the relative orientation between adjacent images. Further the relative orientation is fused with the road orientation in a Kalman Filter to support outlier detection. The approach yields rotation errors of around 0.16 degrees for pitch and roll. This is only 1/10 of the used IMU’s accuracy (Applanix POS-AV 410) but 3 - 10 times more accurate than current low-cost systems.