 |
Mitsubishi 4WD Owners Club of Qld (Inc) |
|
|
GPS - Part 5 |
|
|
|
|
|
|
| |||||||
|
GPS-101 Part 5 (Maps on GPS units) Hi again for another exciting instalment of GPS-101. This month I’m going to discuss the different types of maps that are available on GPS units. Unless you’ve been living under a rock, or been busy hunting for sabre tooth tigers, you’ll have seen satellite navigation systems advertised. These typically have maps of gazetted roads and can work out how to get from one place to another and back again, and will even give you audible instructions of where to turn. These units use ‘vector’ maps in conjunction with a GPS receiver. The other types of maps that are often used in conjunction with GPS units are ‘raster’ maps. So, what’s the difference? Vector maps are defined by points, lines, line strings, curves, ellipses, polygons, text, and other objects. In simple terms: to define a point you only need one coordinate; a line needs two coordinates (each end of the line); line strings, polygons and curves have multiple coordinates; text has a coordinate, an angle and the text. As well as the location information, each of these objects can have multiple attributes, such as colour, size, line type (e.g. solid, dots, dashes), etc. The vector map is drawn by the computer using software that understands the defined format for the vector map and the attributes associated with it. The computer could be a PC, PDA, or GPS Unit. Some advantages of vector maps are: - They take up a lot less space than raster maps (I’ll explain why later) - You can include other type of attributes about the objects, such as road names, speed limits, if they are toll roads or not, direction of traffic flow (e.g. one way). - You can include connectivity information between the objects (i.e. roads, roundabouts, etc) It is the connectivity information that is used by the satellite navigation systems to calculate your route for you. If all you want to do is get directions from one address to another, then this type of sat-nav system is all you need. But, if you want to get off the beaten track, like most 4WD club members want to do, then these type of systems won’t get you very far. Raster maps are defined by pixels (picture elements), just like your digital camera. They are effectively just digitised photos of the paper maps. As an example of the difference, consider the diagram below:
A vector map can define this with very little information. There are three lines, each with two coordinates. The software knows how to plot the coordinates and make the raster image that you see on the computer screen, by joining the coordinates with a straight black line. A raster map has to split this up into many hundreds of horizontal lines of information (depends on the resolution of the map). Let’s just consider the small section of this diagram in the dotted box and enlarge this section
A raster map would describe this as follows – where w = white, and b = black: Line 1 = w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,b,b,b,w Line 2 = w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,b,b,b,w Line 3 = w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,w,b,b,b,w,w Etc. A simple raster compression format uses Run-Length-Encoding, which would look like this: Line 1 = 24w,3b,1w Line 2 = 24w,3b,1w Line 3 = 23w,3b,2w This saves a fair bit of space, but with hundreds or thousands of horizontal lines to describe, it is still a lot more information than the vector equivalent. This is exacerbated if you start using hundreds of different colours as well. In general, the more complex a map becomes, the greater the difference in size between a vector map and a raster map. Raster maps are used in software packages like OziExplorer, which I discussed last month. The beauty of using raster maps, is that you can scan any paper map and calibrate it to the geodetic coordinates – providing that the map has been produced on a valid geodetic projection. I’ll explain that statement below. The bottom line is that unless you know how the curve of the earth’s surface is represented on the map, then the coordinates from the GPS won’t mean anything. A map is a flat representation of the surface of the earth, which – believe it or not - is not flat. However, the earth is not a perfect sphere either. We’ll discuss more about this next month. Quite a few people have asked if I can calibrate the maps that we get from the 4WD parks. The short answer is usually, no. This is because of the way that they are created. These maps are often produced by drawing tracks on top of topographic maps. But, the problem is that they have usually been enlarged or reduced on a photocopier, which has not enlarged or reduced evenly in each direction, let alone from the middle to the edges. Typically these photocopiers use lenses, which distort the edges slightly. Borrow somebody’s glasses and have a look through them. You’ll see the distortion at the edges. So the problem is, that unless you know exactly how the image was distorted (or projected) you can’t un-distort it accurately. Next month I’ll take a look at map projection, datum and coordinate systems. Happy Navigating. Greg Conlon |
|
|
