Paper presented at Auto-Carto '97
Literature review of map complexity
This project is designed to test new methods for selecting and displaying aeronautical chart data. Vector data, such as the Digital Chart of the World (DCW) or VMAP, is selected according to a ranking system which optimizes the usefulness of the resulting map.
The basis for this system is a cognitive model of object importance. Importance is calculated from a rating system derived from spatial interaction theory. Objects are assigned an initial value and a distance decay function. The relative importance of each object depends upon the value of the distance decay function at a given distance. The distance, R, can be changed to produce different orderings of the relative importance of objects.
The fundamental innovation of this method is that it evaluates the relative importance of objects of different classes, as well as those within a class. Conventional systems of map generalization only analytically evaluate the relative importance of features within a single class.
In practice, R can be changed with scale to produce maps whose emphasis on different classes of objects at different scales. For instance, at small scales such as used for enroute navigation, airports would be emphasized. At low altitude, where terrain and other tactical concerns dominate, elevation data is presented primarily.
Because of rapidly changing conditions and high cockpit workload, tactical aviators are faced with a special need for appropriate map information. This technique is designed to exploit the availability of new data sources while enhancing situational awareness. Experiments conducted on aviators will be used to refine and quantify the value of these techniques.
This project combines applications of geographic information systems, cartography, analytic geography, cognitive modeling and psychological testing in order to improve the design of moving maps.
Each movie presented here consists of 90-100 frames generated in ArcView 3 and recorded using a screen capture program. They are then assembled into an animated .gif file for playback.
Movie 1 presents an earlier version of symbology that I have been developing, and database reselection every frame. Note the instability of certain objects in the display, despite the fact that only one parameter (scale) is being manipulated.
Movie 2 presents newer symbology designed to more clearly indicate which objects are being selected. Database reselection is performed every ten frames. Scale is the only parameter manipulated in this movie also.
Movie 3 also presents the newer symbology and reselection every ten frames. Two parameters are manipulated here, scale and number of selected objects (ranging from 25 to 16 from the beginning to the end of the movie).
Movie 4 also presents the newer symbology
and reselection every ten frames. Two parameters are manipulated
here, scale
and evaluation distance.
The evaluation distance ranges from a large value at the beginning,
which favors selection of airports, to a smaller value at the
end, which favors selection of mountain peaks.
This work is supported by the Naval Research Laboratory, Stennis Space Center, Code 7441
