GPS Accuracy

[+FabFranks]

Oddly enough I was on a cruise last month from Sydney to LA. We crossed the equator and my Garmin accuracy was 8'. I asked the first officer about the ships GPS accuracy and he said about 3 meters (10'). I suspect he was not quite candid as I can not believe my little GPS is better than a commercial navigators.

[+StarBrand]

Keep in mind the "accuracy" reading you see is known as an EPE. Estimated Postional Error. It is a low confidence estimate of the error currently being observed by the current sat constellation and signal strength.

 

Generally you should read it something like: "I think the current reading might be within xx feet of the actual coordinates about 60% of the time under similar conditions". The actual error could easily have been 2 or 3 or 4 times that estimate. The unit just doesn't know. If it did - don't you think they would correct for it??

 

Also, we are talking a rather large ocean going vessel here not a micro cache - I think being within 10 feet for a vessel that size is easily "close enough"

[Motorcycle_Mama]

Moving to the GPS and Technology forum.

[+splashy]

This 10 feet will be the bridge of the ship, so it doesn't really matter so much.

Edited December 5, 2009 by splashy

[+boda]

When you think about, if you are navigating from Sydney to LA, 10 feet will get you pretty close.

[+TotemLake]

You might also want to check the specs on your Garmin. I'll warrant it states to 3m accuracy as well.

[+Chrysalides]

The actual error could easily have been 2 or 3 or 4 times that estimate. The unit just doesn't know. If it did - don't you think they would correct for it??

It can correct for it only if it knows in which direction the error lies. Which it doesn't.

[oakenwood]

Oddly enough I was on a cruise last month from Sydney to LA. We crossed the equator and my Garmin accuracy was 8'. I asked the first officer about the ships GPS accuracy and he said about 3 meters (10'). I suspect he was not quite candid as I can not believe my little GPS is better than a commercial navigators.

Don't underestimate your little Garmin. Their high-sensitivity chipsets let them see more satellites for better accuracy. As others have pointed out, a ship's GPS doesn't have to be all that accurate. Even ten meters of error would still fall within the hull.

[+Team CowboyPapa]

Oddly enough I was on a cruise last month from Sydney to LA. We crossed the equator and my Garmin accuracy was 8'. I asked the first officer about the ships GPS accuracy and he said about 3 meters (10'). I suspect he was not quite candid as I can not believe my little GPS is better than a commercial navigators.

Don't underestimate your little Garmin. Their high-sensitivity chipsets let them see more satellites for better accuracy. As others have pointed out, a ship's GPS doesn't have to be all that accurate. Even ten meters of error would still fall within the hull.

Actually, that might be worthy of a plural, might it not?

It would not surprise me if they have two of them, aft and stern.

How else would they now the direction of the axis of the ship?

[+naviguesser74]

I had a chance to follow a surveyor who was marking points with an $80,000 GPS. I shot the same points using a Palm Treo 700P/Global Sat BT GPSr that I know to be pretty accurate and an iPhone using MobileX. For 9 pts., TREO avg diff for LAT was 5.63' and for LONG was 6.19'. For iPhone, 18' and 8.76', respectively. It was an open area. The iPhone is not as sensitive -- it takes awhile to update, where the Treo updates with every step -- but the accuracy isn't too bad.

 

By the way, where I am (NW Ohio 41N and 83W), one degree Lat = 69 miles, Long. 53 miles. Using decimal degrees, the 4th decimal point, I calculate to be 36' and 28' for a one digit change in lat. and long., respectively. As the fifth digit moves up and down by one, it should be a 2'-3' change.

[NordicMan]

Our little "consumer grade" GPSr's are darn accurate! I've heard many stories about how military personnel would often bring a consumer grade GPSr with them into the field, because they're accurate enough for "non-critical uses"

Edited December 7, 2009 by NordicMan

[+Sky King 36]

Keep in mind the "accuracy" reading you see is known as an EPE. Estimated Postional Error. It is a low confidence estimate of the error currently being observed by the current sat constellation and signal strength.

 

Generally you should read it something like: "I think the current reading might be within xx feet of the actual coordinates about 60% of the time under similar conditions". The actual error could easily have been 2 or 3 or 4 times that estimate. The unit just doesn't know. If it did - don't you think they would correct for it??

 

Well said. Further to what StarBrand said, there are a number of error sources in GPS calculation, typically referred to "DOP" or "dilution of precision". Some sources of dilution, such as atmospheric (ADOP, or differences in the propogation speed of the radio wave through the atmosphere) and TDOP, which are clock slippage timing errors, are generally minimized (not eliminated, just reduced) by WAAS. There are other dilution sources that are not easily corrected. The greatest of these is GDOP, or geometric dilution. This is the decrease in precision caused by the geometry of the visible satellites. In open skies the angles between visible satellites is large and so the accuracy of position based on triangulation* is high. In urban terrain, hilly terrain, or in heavy foliage, satellites near the horizon are no longer visible and GDOP climbs rapidly. WAAS can't really help with this. And the receiver can't work this out. All the receiver knows is that "the satellites are aligned with each other in a way that makes me less confident about my solution, and I will reflect this lower confidence to my user in the form of a higher estimated accuracy." This estimated accuracy is easier for the lay user to understand than displaying CEP (circular error probability) variables.

 

*(Geek disclosure... GPS engineering purists are likely to reply that GPS is a rho-rho calculation, not theta-theta, so it isn't technically triangulation... but the way in which GDOP dilutes precision is best illustrated this way)

[+Minimike2]

He was on a boat. Rho-rho rho your boat.

Theta way he had sufficient accuracy.

Edited December 9, 2009 by Minimike2

[+TotemLake]

He was on a boat. Rho-rho rho your boat.

Theta way he had sufficient accuracy.

[oakenwood]

If you're in a valley, holding your GPSr over your head just might raise it enough to see a satellite low on the horizon, improving accuracy.