GPS

There are a number of data acquisition systems available that use GPS, and at first it seems like the ultimate technology. By using GPS you can pinpoint the exact location and speed of the vehicle. This gives you accurate track maps; speed and distance data that is unaffected by tire size, wheelspin, lockups, or line; no need for a trackside beacon; easy installation; and the ability to analyze the line taken for each lap. Plus, if you get a chance to take some slow laps on the track you can drive the inside and outside edges of the track and make a map that  shows the actual track boundaries, to see if you are making your apex's and using all of the track. Sounds great!

Unfortunately, GPS doesn't fully deliver on the promise. The problem is accuracy. If you dig deep into the specifications for most GPS based systems, you will find that the typical accuracy of the GPS position is 3 meters. That's 10 feet in any direction. This means that two points shown as the same position in the data could be as much as 20 feet apart in reality. 10 foot accuracy isn't nearly good enough to make any comparison of driving lines.

It isn't quite as bad as it sounds, however. The 3 meter accuracy is mostly due to a slow drift in the position, caused by changes in the atmosphere and other factors that affect the GPS signal. The fact that the drift in position is slow means that vehicle speed can be accurately measured, because the samples are taken so close together (multiple times per second) that the drift doesn't affect the speed measurement. It also means that position measurements taken during a short time period will have much improved relative accuracy. To illustrate this, below are a few GPS traces taken from the same car and track with a Race Technology DL1. The traces colored black are from 5 consecitive laps, and the traces colored red are 5 consecutive laps from the next day. As you can see, the traces from consecutive laps appear to have reasonable accuracy, but the traces from different days have a significant drift.

It should be noted that there is a way (an expensive way) to improve accuracy of a GPS system by using an external, fixed base station. This base station can measure the GPS error and communicate it back to the vehicle system, which uses this information to correct the signal. With the right equipment, accuracy can be improved to as little as 2cm. The following is the published accuracy of some GPS based systems. "CEP" means "Circle Error Probable", and is the circle in which the measurement error is, with a given % confidence. The prices are included to illustrate the cost differences needed to make even small improvements in accuray.

Assuming that you are using a typical 3m accuracy system, what does all this mean to the user? First, it means that the speed data taken from a GPS system will be accurate, but the position data may not be. Second, it means that the ability to analyze line is restricted to laps that are close together. If you want to compare two sequential laps, the GPS data will give you a good indication of what was done differently. If you want to compare different laps from the same session, perhaps 20 minutes apart, you will be able to pick out large differences, but anything subtle will be questionable. If you want to compare laps that are a few hours, days, or months apart… forget it. Also, the idea of mapping the inside and outside boundaries of a track has limited use. It would work OK if you were able to drive the boundaries then immediately run your laps, but not for later sessions. Finally, if you want the capability to compare lines, make sure that the sofware even has the capability to do it. Some software uses the GPS data to create a track map, but doesn't allow multiple laps to be compared.

That said, GPS does have some advantages:

• It doesn't require a trackside beacon. It can be a hassle to put out and retrieve your beacon, and make sure it always has good batteries.
• A basic installation is very simple, and a GPS system can be moved from car-to-car easily. The caveat here is that the "basic" installation is only going to give you speed, position, and lateral acceleration. To get any extra inputs, such as RPM, throttle position, etc. requires the same wiring as a non-GPS system, and once you've done all of that adding a wheelspeed sensor (for a non-GPS install) is not much more work.
• The speed and position signal is not affected by tire size, wheelspin, lockup, off course excursions, etc.

If you are looking for a system that will provide basic data and can be easily moved to many different cars, a GPS system is worth considering. In most permanently installed applications, the cost of GPS outweighs the benefits. Even if you wanted to swap the system between two vehicles, you could wire both vehicles for wheelspeed for much less than the cost of GPS. In the end, it is important to evaluate the total system—hardware (GPS, inputs, display, etc.) and software, to decide which is best for you.