Integration of GPS with Other Sensors and Network Assistance - Understanding GPS: Principles and Applications - page 504

temperature information, not all do. The second issue that must be considered is the
stability of the underlying sensitivity itself. Can the temperature compensation
curve, without adjustment, be used for several months or even several years? The
answer to such a question may not be known by the manufacturer, so it is therefore
advisable to at least monitor the curves for stability. This subject is addressed further
in Section 9.3.3.
The use of gyros and accelerometers based upon MEMS technology is currently
receiving much attention for military systems [30] as a result of expected cost,
weight, size, and power savings. Due largely to their use as sensors for air bag
deployment in cars, accelerometer development is more mature than that for gyros.
Until fairly recently, however, performance has been a limiting factor. In [31-35],
accelerometer developments are described that achieve navigation grade accuracies
(i.e., with bias errors as low as 20 g, and scale factor errors approaching 50 ppm).

Gyro technology is not there yet, however; [36-38] summarize key developments.
Although 1º/hour gyro bias performance is predicted, reported performance levels
are limited to 10º/hour, with scale factor errors approaching 500 ppm. MEMS
applications in the commercial world appear very promising [39]. Automotive navi-
gation systems could certainly make use of MEMS technology, given the perfor-
mance improvements expected, as the gyros and accelerometers are cost competitive
with existing inertial sensors. Characterization of MEMS-based sensors for land
vehicle applications is treated in [40]. Like the existing sensors (e.g., the vibrational
gyros), MEMS gyros, as well as accelerometers, are expected to have significant
temperature sensitivities that must be compensated to realize their full performance
potential.

µ

9.3.2.2 Map Databases
As mentioned in Section 9.3.1, the emergence of high-quality, affordable, digital
maps has been a significant factor in the wider acceptance of automotive navigation.
Digital road maps are not only an essential component for pathfinding and route
guidance in navigation systems, but also a high-value addition to the positioning
subsystem.
At the time of this writing, two primary companies were developing digital road
map databases for vehicle navigation: Navteq and TeleAtlas. TeleAtlas acquired
Etak in 2000 and acquired Geographic Data Technology in 2004. Both Navteq and
TeleAtlas have extensive databases covering most of the United States, Canada,
Europe, some countries in Asia and the Middle East, and other emerging markets
worldwide. The accuracy of these databases, as determined by comparing road cen-
terline vectors to ground truth, ranges from under 12m in urban areas to 50m or
more in rural areas. New initiatives are underway to map road centerlines to better
than 5-m accuracies and to include vertical information for use in advanced driving
systems. Over time, both the positional and topological accuracies are being
improved through GPS surveying, photogrammetry, and other data acquisition
methods [41].
Even before GPS became a viable positioning system for use in commercial
products, digital road maps were used as a component in the positioning subsystem
of navigation systems. The Etak Navigator, introduced in 1984, consisted of a cas-