TABLE OF CONTENTS Feature-matching techniques determine the user's position by measuring features of the environment, such as terrain height or roads, and comparing them with a database in the same way that a person would compare landmarks with a map.
Feature-matching systems must be initialized with an approximate position solution in order to determine which region of the database to search. Limiting the database search area minimizes the computational load and the number of instances where there is more than one possible match between the measured features and those in the database. Most systems also require a velocity solution, usually from an INS or other dead-reckoning sensor, to determine the relative positions of the features they measure. Thus, feature matching is not a stand-alone navigation technique; it is only used as part of an integrated navigation system. Also, all feature-matching systems provide the occasional wrong fix, either because the database is out of date or due to selecting the wrong match where there are multiple possibilities. This must be handled by the integration algorithm (see Section 14.4.3).
Section 11.1 describes terrain-referenced navigation, Section 11.2 reviews image matching and Section 11.3 describes map-matching techniques. Finally, Section 11.4 discusses stellar navigation, gravity gradiometry, and magnetic field variations.
Terrain-referenced navigation determines position by comparing a series of terrain height measurements with a database. It is also known as terrain-aided navigation (TAN), terrain-contour navigation (TCN), and terrain contour matching (TCM), while the term terrain-referenced navigation is sometimes used to describe a broader...
