Home Current News News Archive Shop/Advertise Ridecamp Classified Events Learn/AERC
Endurance.Net Home Ridecamp Archives
ridecamp@endurance.net
[Archives Index]   [Date Index]   [Thread Index]   [Author Index]   [Subject Index]

Re: [RC] Newer GPSs and accuracy - Truman Prevatt

The issue with elevation is not elevation it is coverage. A GPS unit will solve for altitude if it is locked to a sufficient number of birds. Where you have a lot of elevation changes you normally have blockage to some available birds. The 3 dimensional fix is quite good since it was designed to provide three dimensional fixes for military aircraft. If not it uses the WGS84 datum to calculate location. This datum is a model for the earth that defines a mathematical surface (ellipsoid) that is the general shape of the earth. Hence if only 3 birds are available it has to force elevation (which is height above the WGS84 ellipsoid) to be used as zero. It has to do this go get any fix at all with 3 birds. Because of this if only three birds are available, the location will be off - hence, any use of that location in a distance will be off. The same is true independent of the reason you don't have sufficient birds - canyons, tree cover, mountains.

The newer GPS units have more sensitive receivers and a better antenna. Some are equipped with the ability to use some form of differential GPS correction, e.g. WAAS, which will help with the accuracy in some places. The receiver sensitivity will go a long way helping provide better accuracy. There are still things one needs to be aware of. First is situations where the receiver can lock onto a reflection of the satellite signal rather than the direct path because you could not see the direct path but could see the reflected path. Radio signals bound off of things just like billiard balls bouncing off the bumper. This could happen in a canyon, mountain valley, etc. This wasn't much a problem for the low end commercial units until the more sensitive receivers came out. This has always been an issue, however, an issue of a active research because of its impact on military use with a lot of papers in the technical literature looking at how to detect it and mitigate it.

For example in the "urban canyons" as well as natural canyons this can present a significant problem to law enforcement using GPS or to our military in the case we have to send in troops into an urban area. By the GPS locking to a reflected path rather than a direct path (which it can't see) the pseudorange estimate used in the location algorithm can be significantly off (light travels at 1 foot per nanosec (billionth of a second) and the difference in the path length of the direct and reflected path can be hundreds to thousands of nanoseconds. That will cause an error in the fix.

Clouds are not a major issue but tree cover of deciduous trees with foliage will always be an issue since the GPS frequency is above the cut off frequency for foliage penetration and leaves on the order of two inches up make dandy reflectors for the signal. Conifer forest probably won't be an issue with the new receivers.

So I would say that if you measured the trail in two directions and compared you would be fine on relatively flat open terrain. In heavy deciduous foliage I'd still measure by hand. I would check canyons and narrow valleys several times at different times a day to see if the results were consistent. If any were more were more than a 1 to 2% difference, I'd measure by hand to be safe. It is probably best to maintain the fix points in memory and download them to a tracking program. I don't know about the lower end commercial units, but some of the better units have the capability to store every NMEA sentence (1 per second) produced by the GPS receiver. That includes the error ellipse (a measure of how good the fix was) and GPS DOP (Dilation of Precision which is a measure of best quality that particular satellite configuration can give you). If the error ellipse (or DOP) suddenly goes from small, e.g., 10 meters to large 200 meters, then that's a good sign that there is a problem in that area.

The thing to always remember about the GPS units is the GPS constellation nor signal structure was never designed for measuring a continuous distance on the earth. It was designed for global all weather navigation and time reference transfer for a global reference time accurate to several nanoseconds for the military. It generates point estimates. Some will be bad - depending on a lot of things. For it's intended use that's fine since for navigation, GPS simply enhances the vehicle INS by removing drift and for time transfer, it enhances the atomic clocks in military systems again to remove the drift to keep all the atomic clocks synchronized with the global time standard. Every GPS point does not have to be accurate to accomplish that - just statistically "on the average" they need to be accurate.
.
GPS was never designed to support any specification of accuracy for distance estimation on a random route. Can it be used for that - yes with some care in understanding were the issues are and how to address them. I've been playing around with location technology for close to 40 years now have worked on at one time or another most space based locations systems from TRANSIT (a.k.a. NAVSAT http://en.wikipedia.org/wiki/Transit_(satellite)) to GPS and personally I've very skeptical of any trail entirely measured with a low end commercial GPS unit. Not saying it can't be done, just saying it needs to be done with great care. The only exception might be on a flat open desert - but that would be a pretty boring trail.


Truman

Sluys Guys wrote:
Hi folks, I would like some opinions from experienced GPS users on the accuracy of the new GPSs. I have a Garmin GPSmap 60cs that I have been using to mileage a trail for a new ride. In the past my old GPSs have had a pretty good margin of error due to elevation, tree cover, mountains etc. and I felt uncomfortable with using the mileage without some adjustments but I am finding this new model to be quite accurate. I have checked the mileage on some steep sections of road against my odometer and it came out exactly the same in cloudy and sunny conditions.. In some steep sections of the trail I have rounded up to the next 10th just to make sure that I wasn't measuring long. I have also measured by tracing my track on the GPS map to make sure it wasn't cutting corners and I feel pretty comfortable with the final mileage as I rode the whole thing last weekend. For those of you who using the new GPSs what is your experience with the accuracy??.

Thanks, Nancy Sluys

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=

Ridecamp is a service of Endurance Net, http://www.endurance.net.
Information, Policy, Disclaimer: http://www.endurance.net/Ridecamp
Subscribe/Unsubscribe http://www.endurance.net/ridecamp/logon.asp

Ride Long and Ride Safe!!

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=





--

“He who fights with monsters might take care lest he thereby become a monster. And if you gaze for long into an abyss, the abyss gazes also into you.” Friedrich Nietzsche, Beyond Good and Evil


=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


Ridecamp is a service of Endurance Net, http://www.endurance.net.
Information, Policy, Disclaimer: http://www.endurance.net/Ridecamp
Subscribe/Unsubscribe http://www.endurance.net/ridecamp/logon.asp

Ride Long and Ride Safe!!

=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=-=


Replies
[RC] Newer GPSs and accuracy, Sluys Guys