I want to make a comment on Personal GPS units that everyone claims they are using for MPH.

I work with high end computer controlled GPS units almost on a daily basis....at least until this project I am on gets over with.

I read alot of posts where riders compare radar guns to GPS units....claiming that a gps is the best way to come up with MPH....

On that topic, I would agree and disagree..... I imagine that those of you that are claiming to come up with MPH are using a $15000 GPS unit with differential correction and a few other goodies.
Most of the $300-500 units are only accurate to within 30 ft. some only 300 ft.
In order for a GPS to be anywhere near accurate, a reciever has to latch on to at least 3 satellites out of the approximatly 24-30 that are available to begin with. Idealy, a gps has to latch on to 5 or more for realistic readings.
Most store bought GPS units are forced to take advantage of what is called c/a....which basically is reserved for civialian access to the satellites. This restriction forces the "home" user with the hand-held gps to have a unit that has some serious correction ability.
also...the time required for a gps unit to calibrate and lock in is often not feasable for these kind of calculations...In order for a person to come up with MPH the GPS is assuming that you are maintaining a constant speed from point a to point b (such as in an aircraft). If you start at point "A" and start to get yourself up to speed...By the time you reach point "B" , you have nothing better than a computer generated estimate on how fast you went.....(since satellites are geostationary there could be a problem if the unit tries to latch on to a new satellite)
This was not meant to be a point of argument, just an eye opener....Because if you read the ads and paper work for most of the "cheaper" hand held units...they can be very misleading. and I rarely doubt that most of the people who make these claims are toting a military grade GPS and a huge Battery to power it.
The handheld units are great....don't get me wrong....just not very accurate for finding a needle in a haystack. I would still like to own one...They are accurate enough to save your life if your lost.

 

While non-differential GPS receivers sold over the counter to casual users have an error budget regarding ABSOLUTE positioning as described, these units are quite accurate regarding indicated speed.

The speed indication technique compares successive positions between each other with respect to time; a constant error of position does not degrade the accuracy of indicated speed.

In other words, a consistent position error over the time interval when a speed reading is taken does not affect the speed reading accuracy.

The position readout is absolute; the speed reading is realtive (to the previous position readout).

GPS MPH is pretty good.

Tree Farmer

 

I had a gps in an a car, and the speed is not right on, it's close, but it's not gonna tell you if your quad is faster than the next guys, considering the small amount of speed that most quads are faster than others.

 

Tree Farmer,

I have to disagree, sorry.

For example, the GPS takes the first reading and the position is off, by say 100 feet. If all subsequent position readings are also off by 100 feet, in the same direction, then the ground speed will be accurate, you are right here. This is not the case however. Every position reading from the GPS is subject to the (up to) 300 foot "fuzz factor" injected by the US Government. Each reading is not off by the same factor (distance or direction). Thus ground speed cannot be accurate without correction data being fed into the GPS.

In other words, if the subsequent position reading has a different error from absolute (distance and direction), than the previous reading, the speed cannot be accurate.

Just spent a few months designing an interface that straps to the back of a Garmin GPS 12 receiver. The project was to prove the feasability of feeding correction data into an inexpensive hand held receiver. We sent local correction data to the PageNet satellite. It was then picked up by a PageNet pager strapped to the back of the GPS and fed into the GPS (via an interface card). We managed to improve the accuracy of the receiver from 300 feet to under 3 feet. During the project we monitored actual position versus what the GPS reported. Let me say that off-the shelf, the Garmin GPS 12 was surprisingly accurate, without the correction data, most of the time.

Your statement indicating the GPS receivers are "pretty good" is the best we can say about them. I would still put radar on top of the list if a hand held GPS is the other contender.

DJ

 

I'm not sure you disagree, deej!

Rather, you confirm my premise.

If, over the time interval of the MPH reading, the position error remains constant (x,y,z "bar" deltas unchanged), rather accurate speed readings emerge.

Integration programs smooth out the "jitter" largely from time-varying position errors otherwise producing trivial results.

In terms of position accuracy, 30-300 feet error is not uncommon, depending upon geometry, etc., factors influencing accuracy. Considering, however, we're talking baselines 22,300 MILES long, fairly accurate speed readings are possible from successive time-difference-of-arrival measurements.

Is good, calibrated radar more accurate? With probable certainty. But GPS ain't bad, from someone who's intersected a few hyperboloids over time.

By the way, Al "Pinnochio" Gore did NOT invent GPS (as he claims he invented the Internet); instead, one Mr. McDonald and his team at the Defense Advanced Research Projects Agency (DARPA) did; DARPA advances a competeing claim about the Internet, also.

Anyway, Mr. McDonald (now in private industry) sold me my Garmin GPS II+.

Tree Farmer

 

Well, I certainly have no credentials with which to dispute any of these statements. however I did call the tech dept. at Garmin to ask what the margin of error was regarding the GPS 12's measurement of mph when used on a quad. He told me that it was very small..about .5mph, which would certainly put it in the realm of most radar or vascar guns & certainly as accurate as most speedometers.
Anyway, for what it's worth.

 

I use my Garmin GPS unit with a laptop containing "Street Atlas, USA" in my van while travelling. The readings from the Garmin and the van speedometer are very close, so I believe the Garmin statement that the speed is within 0.5 mph. Also, the handheld GPS units have already "latched" the satellites prior to giving a speed indication, and typically 5 or more satellites are being received in these 12 channel units. Speed indications can't be believed if the user is moving slowly, as the "Selective Availability"/scrambling makes the speed indication in error.

 

TreeFarmer,

You are right. I think we are on the same page here. I am agreeing with you without realizing it.

Can the position error be considered constant over the measurement interval? Yah it probably can, given the length of the interval (and as you point out, the relatively short interval given the big picture). Thus the speed reported by the GPS can be quite good.

The 0.5 MPH accuracy figure quoted by Garmin is interesting.

Good banter! (and good night!)

DJ

 

For what it is worth, I'll take the results from a laser-transit and a couple of photo-eyes hooked to a stopwatch any day.

 

Kevin, your measurement techniques (laser ranging and photo-electric timing over precisely-measured intervals) are undeniably more accurate than consumer-level GPS equipment and techniques. (Doppler radar gives more accurate speed readings, also.)

However, Chrondek setups are rare, mostly on racetracks or dragstrips; GPS provides the advantage of practically universal availability; road, trail, woods.

Within its range of precision, GPS provides usable MPH figures; yet, as someone said, the guy who gets to the finish line first wins!

Tree Farmer