I'm lucky enough to hold a few patents, through work (Trimble).   Several have been filed over the years, but so far only three have been approved:

US6021376: Method of displaying connections in the field between like geographical features
This is a real-time feature code processor, running in the field.  Generally, Surveyors are recording point features.  Feature codes are a way they can attach information about relationships between points, for example that a particular point is part of a fenceline or is a tree. The feature code processor examines the codes (a short string containing codes like "FCE" and "TREE") and adds extra details like lines joining the points on a particular fence.  Thus a map display can go from being a meaningless bunch of dots to (after processing) a realistic looking map, with linework and appropriate symbology.  Traditionally, this process is done back in the office after the data has been collected.  The patent has it being done in the field, as the points are recorded.  It's in a real product I worked on.

US6031471: Full alphanumeric character set entry from a very limited number of key buttons
This is about being able to enter alphanumeric text (like feature codes) on a device that has a pretty good display but very few keys.  The basic idea is to display, say, all the letters A through Z and progressively select a smaller and smaller subset of them until a single character is chosen.  The selection is based on the target character's position within the full set - the spatial relationship is mapped to the physical arrangement of the keys.  A major advantage of the idea is that a given character is always the same sequence of keystrokes.

US6072428: Location determination using doppler and pseudorange measurements from fewer than four satellites.
This is a shared one.  My contribution was the idea of using the first and second derivatives (with respect to time) of the Doppler Shift.  GPS replies on signals broadcast from satellites orbiting overhead.  They're moving quite fast, fast enough for there to be an appreciable shift in the apparent broadcast frequency, due to the (relativistic) Doppler Effect.  The idea is that if you look at the rate at which the shift changes (first derivative) and the rate that the change changes (second derivative), you can compute your relationship to the satellite's velocity and acceleration vectors.