| A new terrain matching algorithm is presented for use in a passive aircraft navigation system. A monocular sequence of aerial, optical images is matched to a reference digital map of the three-dimensional terrain. Stereo-scopic analysis of each successive pair of images in the sequence yields a recovered elevation map of the observed terrain in the field of view. A two-dimensional "cliff map" is then used as a novel, compact representation of both the recovered elevation map and the reference map. This representation makes it possible to match the sparse, stereoscopic data to the dense reference map. A terrain matching algorithm is presented which achieves its efficiency by matching high-curvature points along the two-dimensional cliff contours. The observed cliff map is found within the reference cliff map, and the associated translation and rotation parameters provide the desired position and heading of the aircraft. The robustness of the terrain matching algorithm is demonstrated by experimental results using real terrain data. A detailed analysis of the performance is presented, including results using both simulated and real stereo implementations. |
