Skip to main content

I received the following inquiry from a certifier:

quote:
On a an application for a calibration course certification, the applicant had hired an engineering firm to measure the course (half mile). The method of measurement was listed as 'Survey Grade GPS Receiver.' I hadn't heard of this device, and don't want to plunge forward with the certifications if you feel there's any question about the method. Do you know if this is acceptable?


I did some digging around and found the following:

quote:
Survey-grade GPS units are used for surveying tasks that require very high accuracy (within 1 cm), such as bridge construction. Survey-grade GPS units use differential correction for highest accuracy.


Global Positioning Systems

These things aren't cheap; $20K is listed as the price.

The question remains: is this close enough for government (i.e. RRTC) work? For a course measurement, I'd say yes; cal course, maybe?
Original Post

Replies sorted oldest to newest

I know nothing about this method. Why not ask the outfit that did the work to supply some info? I think there is a good chance that the method is OK. My son Tom works for an outfit that uses GPS for precise layout of things. This is not your ordinary off-the-shelf hand-held unit. It's a unit which has an associated base unit located nearby, permitting really good accuracy.

Perhaps Tom can shed light.

If a professional surveyor has signed off on the accuracy I'd be inclined to accept the work at face value.
Gentlemen,

Accuracy with surveyors' GPS falls within two ranges, more or less. Sub-meter positional accuracy, and centimeter level. The methods and equipment used determine the level of accuracy. Centimeter level would yield a maximum error of 2cm over the course. Maximum acceptable in theory, but in practice when we re-acquire a benchmark it's on the order of a millimeter or two difference from the previous reading.

The additional accuracy is obtained by having two receivers. Each receiver gets a signal from the satellite grid, and from the other receiver. This forms a trianglular loop from the two receivers to each satellite acquired, and many samples are averaged instantaneously.

To certify a boundary-line survey, an accuracy of one part in ten thousand is required (0.26' for 1/2 mile). Most of the large boundaries we close are within a couple hundredths on rough ground using traditional equipment. I seriously doubt the surveyor would certify anything less than this.

Since this is a level-projection, or straight-line distance it will vary slightly from the distance on the pavement, but if it's even relativly flat the error won't be significant. A 20' hill in the middle yields only a 4" length difference from plane. One count on the counter.

I'd be inclined to go with it.

Best,

Tom
Tom,

Doesn't this mean that distance measurements between the two receivers would be much more accurate with a system like this, but absolute position accuracy wouldn't be improved much? I think to have improved absolute position accuracy you need an additional receiver at a known position. Of course for a calibration course you would just need the accurate distance measurement.
Mark,
The distance accuracy in this case is the same as the position accuracy. Leica (they make our GPS rig) advertises positional accuracy as +/-5mm +/- 5/10,000,000 of the baseline, or distance between receivers.
Random errors tend to cancel each other out in repetative measurements so the error between two points, on the average,is expected to be the same as the positional accuracy.
Horizontal data is more precise than elevation data, due to the geometry of the triangles formed by the baseline/satellite. Longer triangles from horizon to horizon are formed than from surface to orbit.
Once the receiver has been set up for a while, it becomes a known point because of the number of samples it has averaged. Doing base mapping, we may let the units sit for half a day just logging data. Setting up on a known point eliminates the wait.
There's a huge amount of work crunching the numbers after collecting the data. GPS fixes your position relative to the earth's center of gravity. This needs to be translated into ground coordinates using the "ellipsoidal height" of the earth at that point. If you thought spherical geometry was hairy....
GPS is very accurate, and repeatable. In the hands of a qualified field crew there's few better locating systems. And none that can perform over such a wide area. Our work involved locating every manhole/valve/fire hydrant etc. in Augusta County, involving hundreds of thousands of points. We will commonly go back to things like a water meter to use as a benchmark. One shot is as good as the rest. We were recently called up to locate a water valve that had been paved over. Because of the huge number of metal objects in the pavement, a metal locator was useless. GPS, right on the spot.
quote:
Any sufficiently advanced technology is indiscernible from majic
.
Last edited by tomriegel
Message from Michael Franke:

Thanks to you all for getting back to me. I did the obvious, as Pete suggested, and simply called the fellow at Engineering Inc., in Billings, MT, who did the measurement. I had a nice conversation with Michael Sanderson, who said that the precision of the device (he mentioned a price of $60,000) was at least +/- 0.01 meters. That seems pretty good. If the measurement data for the Montana Marathon had only been relatively as good.


Michael

Add Reply

Post
×
×
×
×
Link copied to your clipboard.
×