How will we know if a speed reading is accurate?

I have been recording tracks on my GPS and finding anomalies that will show me suddenly sailing about twice or three or four times the actual speed. These anomalies are cast out when I mention my max speed.

The max speed indicator on my gps does not know how to eliminate anomalies, bogus readings. So I find that the max speed feature on my gps is useless. When I upload data tracks to my computer and analyze a track for highest speed, I need to find a series of readings that are somewhat similar. If I see a series of speed registrations that are all within a few percent of one another, then I'll trust them. But when I see 6,9,8,9,7,6,7,7,9,14,7,6,6, .. Well then I am going to cast out that unusually high reading. Too bad the max speed log would not be able to do the same!

If you are reporting your max speed from a single momentary reading, or from a max speed log, then it is very likely that you are believing a reading that is erroneous, an anomaly.

How can we know if a reading is accurate or if it is just erroneous?

GARY

I've been doing another anomaly test. One can really tell that the US army has brought back the build in error. Last time I did the test it moved around very little only 1 to 3 meters in many many minutes. Nowadays it can move about pretty far; and quickly as well. Worse test showed a swing of 20 meters and a max rate of change of 5 meters a second. THAT is pretty drastic. So I have to recall some of my earlier comments on GPS use in speed trails. Two years ago the shift was small enough by far to not influences the tracking too much. Now with the Iraq war it sometimes does; although the same test also showed some stretches of 10-15 minutes in which nothing happened.

So I will go to my back-up plan for GPS tracks. Luckily we in the Dutch F16 class have several GPS units and therefor 1 spare. So what we will do next time is leave 1 GPS tracker in the car in a bag underneath the front window and leave it stationary there. We'll take the others along on the boat. Afterwards we just deduct the swing recorded by the stationary unit from the moving units and cancel out the US army build in error. Actually it is sort of our own differential GPS setup but with a higher accuracy then WAAS or commericial DGPS's. MS-Excel comes in really handy now. Maybe I'll write a short program to do the substraction when I have some spare time left.

So for all you speed freaks out there. Restore your old accuracy by having a stationary tracker during your sailing and do some input file hacking.

The other alternative is indeed to look for smooth average read-outs as you decribe. Calm periods are interchanges with active periods. At least the stationary tests I've done show that. See picture



Coding; block size = 5 seconds = tracking sampling rate.

Blue = 2 km/h or less
light blue = between 2km/h and 5 km/h
red = above 5 km/h


If the record attempt was made in the first 4 minutes than the data would have been undependable. If you did it after 4 minutes than the error speed would be been between 0% and 2% ; so at max 0.4 knots when having a base speed of 20 knots. For a base speed of 30 knots the error would at MAX have been 0.6 knots. With a little luck you would have done the attempt in one of the stationary times and get a 100 % accuracy (0% error) as in those stretches no swing occured.

We can device a ground rule that for all record attempts at stationary GPS tracker must be used so we can see that the swing at the record time was and compensate the measured for the swing. At todays prices this should be easy enough. I think I located another cheap GPS tracker that allows storage of up to 50.000 track points and has a stand-alone battery life of about 8 hours with a wireless blue tooth PC link. This is sufficient for cat speed/course tracking.

Wouter

I don't doubt that this might be a possibility but don't you think that's a pretty big jump to a conclusion?

12.4 nautical miles in one hour.
..................
Woops, it won't let me attach a gps track file.

GARY

Post deleted by hobiegary

I just got the May Car & Driver. They have a test of Performance Meters.

Most, " use accelerometers and an electronic timer to measure performance. An accelerometer is a device about the size of a wristwatcvh battery that produces an electical signal that changes under aceloratrion. Simply put, if you can monitor aceleration in relation to time, you can calculate speed and distance traveled".

These range in price from $90-$960.

Car and Driver uses an $11,800 VBOX GPS based system for their testing.

Who's got a hot rod buddy to share one with?

I don't believe the U.S. ARMY is in charge of when S.A. it's turned on or off. Pres. Bubba ordered it turned off. And since it was turned off my EPE on my GPS always has the same 14'. That's an older model Garmin 48. My 76 with WAAS was about 7'. Before S.A. was turned off the EPE would be in a constant state of change.

I think it's possible that the U.S. military has the ability to choose the area of world where it would want a greater error.

Correct me if I'm wrong here But I remember that pres Clinton turned the applied error OFF, not bubba. Somehow I also remember that Bubba came back on this decision and turned it on again. But this could be just me.

Anyways. I'm looking at the about the same skies as the arabs so ...

Wouter

Yep, I think I'll have to admit that you say a true thing here. It is only a possibility at this time.

In principe a normal GPS without S.A. degrading should be able to pin-point your position to within a radius of 3 to 5 meters. And thus compete closely with Differential GPS applications. In my last test I got accuracy of about 15 meter radius over a time span of 10 minutes; with a swing speed of 14.5 km./h at one time and about 9 km/h at another time. This is pretty drastic and I hadn't seen that before.

But yes there can be other reasons as well. The view of the sky in the test was not perfect and maybe the losing of 1 or 2 satellates shortly after one another could also have caused this rapid swings. Typically these are not present when on the water as there you have a prefect 360 degree sky and often I get signals of 5 satellites or more. This really helps.

Will try to do more tests when I have spare time; this time in an open plane and for longer periods of time

Wouter

To that point, increasing the sample rate (say from 0.5 seconds to 2 seconds) my help dampen the peaks a little.

Wouter,
I don't think it is fair to compare the accuracy of DGPS to GPS w/o SA. Once SA is gone the largest single source of error is atmospheric distortion and DGPS is able to compensate for this but of course these systems are pretty spendy. However DGPS can get you sub-meter and even sub-cm accuracy which is way better than just GPS without SA. Now WAAS is a poor mans version of DGPS but is a good alternative given the cost.

Your idea of using a fixed reciever to subtract out the atmospheric error and get poor mans DGPS is interesting, it would be an interesting expt to set 2 GPS recievers next to each other, log the data, subtract them and see how much error you get. If you do this please let us know how accurate this is. BTW this is the way they used to do DGPS back in the old days...

Wow now that I think about it for a minute it has been 10 years since I worked in this field, sure doesn't feel like it. But some of my details might have become fuzzy with time...

-Eric
SC 20 without GPS

At times we have used GPS for survey work when we have had to do "non critical" coverage over large areas of very "rugged" land, and to ensure that the GPS transmitted, variable error could be eliminated as much as possible we would set up a radio beacon (line of sight accuracy only) on a prominent high point of known XYZ coordinates. By taking a continuous stationary series of downloaded GPS readings at the beacon (with the recorded time of each reading) then we could take both readings from the beacon, when in line of sight and readings from the GPS, and when out of line of sight, the GPS readings only (and at a minimum 5 GPS readings at each point) With the time of each reading known then it is relatively simply to reduce the error factor to an acceptable level in the final result calculations.
Even with this system the accuracy could only be calculated to cm accuracy, suitable for something like a gravity survey but a totally unacceptable accuracy for cadastral surveying for areas like CBD where a 1 mm error can mean thousands of dollars difference in “rental” for a multi storied building, or even out of boundary residential work.
With the fluctuating (anything but “constant” ) errors recorded using this sort of an accurate “checking” system, our results showed that for use as an accurate measurement for speed the GPS is “indicative” only having a low degree of accuracy.

There is also the error of the signal passing from its satellite source through the atmosphere (as well as any built in error). Don't underestimate the amount of variation that the passage through the atmosphere creates on the readings, it is “significant, AND it is not something that can be cancelled out, many people have tried in many ways and, to date, no one has succeeded.

You know,I'm going to guess that any deviation the atmosphere would cause is compensated for.

I'm fairly sure it was BUBBA......of course his next title may be the first lady....er....man????

can't resist.......Maybe if you were a tax payer of the U.S. you'd have a better signal.

It wasn't, do an internet search and see that it was Clinton.

Wouter

The atmospheric variation when using GPS is at all times variable, and although some allowance can be made for it, it is impossible to cancel it out completely, or for that matter even to cancel out the majority of its effects. Its effect is the same as for veiwing the "heavens" through a land based telescope, the picture is distorted, the "signal" bent, and the light frequency speeds are distorted.

Wouter:
"Bubba" is a nickname for Bill Clinton.

18.3 nautical miles in
01:22:54 hours
13.2 Nautical Miles per Hour AVERAGE for 83 minutes.

28 statute miles in 2 hours dead - 14 mph moving average from GPS trip counter reset when leaving beach. Broad Reach without spin in 15-20 mph winds, 4-5ft side seas. Includes time spent cleaning boards and rudders (approximatly 15 times). The (highly unreliable) instant GPS readout varied 15-18 mph most of the time.

the biggest problem I have with relying on gps for speed readings averaged over time is the fact that on the T500, my gps would constantly lose reception even when out in the open.

Very annoying.