How accurate is dis...
 

  You don't need to be an 'investor' to invest in Singletrack: 6 days left: 95% of target - Find out more

[Closed] How accurate is distance calculated using GPS?

20 Posts
17 Users
0 Reactions
189 Views
Posts: 1510
Free Member
Topic starter
 

I am aware that you can calculate distance between two lat/long coordinates, but although it takes in to account the curvature of the earth how does it create an accurate measurement given the altitude difference between the two lat/long points?

eg, say you you have 2 lat / long points close by, but you have climbed 3 miles to get from one to the other, then surely the distance that the GPS says you have travelled must be incorrect?

Thanks


 
Posted : 04/02/2012 11:15 pm
Posts: 0
Free Member
 

Mine tells me how much climbing I have done so presumably they do take into account these sort of things to a certain extent.


 
Posted : 04/02/2012 11:22 pm
Posts: 0
Free Member
 

GPS also uses 3rd or 4th satellite for the altitude component. Any more than that is just trigonometry!


 
Posted : 04/02/2012 11:30 pm
Posts: 0
Free Member
 

It's ignored. The difference is too small to be of concern for the casual/amateur user.

eg, say you you have 2 lat / long points close by, but you have climbed 3 miles to get from one to the other,
How often do you climb a 15,000ft mountain?


 
Posted : 04/02/2012 11:30 pm
Posts: 5559
Free Member
 

GPS can use interpolation iirc but they largely ignore it for the reason Druid notes.


 
Posted : 04/02/2012 11:38 pm
Posts: 1510
Free Member
Topic starter
 

The reason I ask is that I have a requirement for a project I am working on to accurately capture distance travelled. So for example, if I was doing a 40 mile bike ride over lots of steep hills, would my GPS give me an accurate distance travelled measurement compared to a distance measurement based off of a wheel revolution measurement ?

I could try it, just wanted to see if anyone had done this ?


 
Posted : 05/02/2012 12:09 am
Posts: 0
Free Member
 

The wheel revolution method is likely to be more accurate - if properly calibrated. Every GPS has a sampling interval and this can result in an under-recording of distance. Furthermore, it is not unknown for a GPS signal to be interrupted/corrupted resulting in both over- and under-recording of distance.

In practice, for a 40 mile ride it is likely to make little difference.


 
Posted : 05/02/2012 12:14 am
Posts: 33325
Full Member
 

I would have thought that over 40 miles, using a conventional cycle 'puter, the cumulative error caused by any slight measurement flaw in the wheel circumference, like tyre pressure altering, could add up to just as much of a difference as that using a GPS. If it's that crucial, use both together and average any difference in the final measurement. A cycle computer isn't [i]that[/i] expensive, after all.


 
Posted : 05/02/2012 12:23 am
Posts: 2
Free Member
 

If I ever take a GPS file and put it into tracklogs or similar, it always claims more than the GPS device does both for climbing and distance.

Not sure if this helps.


 
Posted : 05/02/2012 1:00 am
Posts: 0
Free Member
 

what device is it?


 
Posted : 05/02/2012 6:46 am
Posts: 0
Free Member
 

what device is it?

Cruise missile 😉


 
Posted : 05/02/2012 7:17 am
Posts: 0
Free Member
 

It's an interesting question one I've been meaning to ask. Last week I did two runs over the same route, first time it was measured at 3.06 miles, second time at 3.11 , tracklogs has it at 3miles. Not the biggest difference I grant you.

I was also wondering if different phone gps \ app combos made a difference? Me and my mate riding together measured a route at 14.74 and 15.25 miles respectively, tracklogs had it at 14.97


 
Posted : 05/02/2012 8:52 am
Posts: 0
Free Member
 

If you need accurate distance measument you need to attach a 3rd wheel .

Any methode will have some error in it.

Probably the biggest error for GPS would be loss of signal now and then even if it is only for a few seconds at a time.


 
Posted : 05/02/2012 8:58 am
Posts: 0
Free Member
 

Not all GPS is created equal. I don't understand the mechanics as to why, but we have kit at work accurate to +/- 3cm and stuff accurate to +/- 50cm. We all know how random mapping on phones can be too.


 
Posted : 05/02/2012 9:06 am
Posts: 0
Free Member
 

[i]The reason I ask is that I have a requirement for a project I am working on to accurately capture distance travelled. So for example, if I was doing a 40 mile bike ride over lots of steep hills, would my GPS give me an accurate distance travelled measurement compared to a distance measurement based off of a wheel revolution measurement ?[/i]

For extra accuracy you really need to have a separate height database and cross-reference that to your lat-lon. There's freely available ones, but I can't recollect if the free ones have the resolution that you require.
But as others have mentioned the wheel reading may also be off for a variety of reasons.


 
Posted : 05/02/2012 9:16 am
Posts: 4675
Full Member
 

Not all GPS is created equal. I don't understand the mechanics as to why, but we have kit at work accurate to +/- 3cm and stuff accurate to +/- 50cm. We all know how random mapping on phones can be too.

That could well be differential gps. There's a base station that the position is very accurately known and by comparing the gps position to the real position you can work out the error in the gps measurement, which you apply to your unknown location.It removes errors due to atmospheric effects. (But I could be wrong, it's a long time since I was into gps, and things may have moved on.)


 
Posted : 05/02/2012 9:20 am
Posts: 0
Free Member
 

The really accurate units are differential.


 
Posted : 05/02/2012 9:23 am
Posts: 19
Free Member
 

what device is it?
Cruise missile 😉

Only if you can disable the [url= http://blog.makezine.com/2011/07/25/gps-units-disable-themselves-if-they-go-faster-than-1200-mph/ ]COCOM Limits[/url]


 
Posted : 05/02/2012 9:30 am
Posts: 0
Free Member
 

Differential gps is still used and the principles haven't changed.

What is really clever is using post processing for applying corrections.

Basically, you can collect positional info and correct it back at the office when processing it. The corrections required are known for the exact time and location.

If your sad and really care Trimble give some excellent tutorials on their site. [url= http://www.trimble.com/gps/dgps.shtml ]Trimble[/url]


 
Posted : 05/02/2012 10:29 am
Posts: 17
Free Member
 

normally see them both line up quite well. Road vs GPS is closest off road wheels tend to leave the ground or bikes pushed or carried.

However have managed some incredible GPS feats including hitting 500+mph a couple of times when is suddenly decides I have crossed a valley in 5s or near some steep edges assumes I am bungee jumping


 
Posted : 05/02/2012 10:37 am
Posts: 0
Free Member
 

You can get very good real time differential solutions using RTK & satellite broadcast corrections from the like of veripos, Fugro & c-nav, although these corrections cost several hundred pounds a day, but will give you sub 10cm positions anywhere in the world.

Although in Europe & us, you can use SBAS services like WAAS or EGNOS (wich are free) mostly to tighten up the vertical accuracy in the solution, which with standard gps is not as accurate as the horizontal positioning.

Yep, I'm a giant GPS geek.... Although I blame my job for that!


 
Posted : 05/02/2012 1:10 pm

6 DAYS LEFT
We are currently at 95% of our target!