GPS trip logs. How accurate are they?

[slofox]

General question here...

I was faffing around with the GPS unit and opened up the trip log from this morning's ride. Oh yes, very interesting. You can see where you went on the map and it tells you time and location and also SPEED! Which brings accuracy into question.

According to this morning's log, I exited Homestead Place onto Pelorus Street at 134km/hr...woohoo Mr Rossi...

I don't think so sunshine. It's about twenty metres from my gate and even if I had tried, I doubt I could have got up to that speed in that short space, especially on cold tyres with a cold engine.

In point of fact, I pootle through there very gently after a near stop at the intersection because the road is obscured and it's too easy to be surprised by a demented cager with a killer rage-on.

So what gives? I think the stupid machine is making it up.

There are a few other questionable speed readings as well but we won't go into those just now...

[mashman]

you sure you weren't on an LSD Trip this morning

[oneofsix]

you sure you weren't on an LSD Trip this morning

If he wasn't sounds like the GPS was
A little knowledge is dangerous so watch out cause here it comes. My guess would be the GPS syncing with the satellites as you started your ride, got its initial position wrong then calculated you must have done 134km to get to the corner in that time.

[george formby]

General question here...

I was faffing around with the GPS unit and opened up the trip log from this morning's ride. Oh yes, very interesting. You can see where you went on the map and it tells you time and location and also SPEED! Which brings accuracy into question.

According to this morning's log, I exited Homestead Place onto Pelorus Street at 134km/hr...woohoo Mr Rossi...

I don't think so sunshine. It's about twenty metres from my gate and even if I had tried, I doubt I could have got up to that speed in that short space, especially on cold tyres with a cold engine.

In point of fact, I pootle through there very gently after a near stop at the intersection because the road is obscured and it's too easy to be surprised by a demented cager with a killer rage-on.

So what gives? I think the stupid machine is making it up.

There are a few other questionable speed readings as well but we won't go into those just now...

It would'nt be average speed would it?

[jim.cox]

Under "perfect" conditions, most commercial grade GPS units will be accurate to under 15m 95% or more of the time. Some of the better units will get down to about 5m

But conditions aren't always perfect - as well as a good initialisation, you need a clear view of the sky and to have at least four satelites in good configuration. Furthermore leaves, buildings and mountains can all bounce the radio signals around and confuse the receiver (try getting a good gps signal in the Otira gorge and you will see what I mean)

But also of relevance is the 'space weather' - and in particular solar activity. Sun spots seem to come in an eleven year cycle. We are starting into ta phase of increasing activity. Activity hasn't ramped up as rapidly as expected, its been unexpectdly quiet for the last two or three yeasr. But things are buildiing now, and there has been some significant sun spot activity thie week - maybe that got you...

The speed calc done by the GPS uses tiime and position - or rather the difference between two positions. The time is reliable - those satelites each have at least one atomic clock onboard. But if either position is mis-calced the speed between will be up the pole too

[YellowDog]

The GPS read the make and model of bike and was then overridden by the in-built 'Stereotype Riding Characteristics' for the bike.

The 134kph educated guess had a 98% probability of being correct for that bike

You are the exception that proves the rule Mr Fox

[SMOKEU]

Does anyone here know to roughly what % accuracy most consumer level GPS is in terms of the velocity reading?

[slofox]

you sure you weren't on an LSD Trip this morning

Pretty sure - coulda been a flashback I suppose...

If he wasn't sounds like the GPS was
A little knowledge is dangerous so watch out cause here it comes. My guess would be the GPS syncing with the satellites as you started your ride, got its initial position wrong then calculated you must have done 134km to get to the corner in that time.

Could be...

It would'nt be average speed would it?

Of course not! WAYYYY too low for that!

Under "perfect" conditions, most commercial grade GPS units will be accurate to under 15m 95% or more of the time. Some of the better units will get down to about 5m

But conditions aren't always perfect - as well as a good initialisation, you need a clear view of the sky and to have at least four satelites in good configuration. Furthermore leaves, buildings and mountains can all bounce the radio signals around and confuse the receiver (try getting a good gps signal in the Otira gorge and you will see what I mean)

But also of relevance is the 'space weather' - and in particular solar activity. Sun spots seem to come in an eleven year cycle. We are starting into ta phase of increasing activity. Activity hasn't ramped up as rapidly as expected, its been unexpectdly quiet for the last two or three yeasr. But things are buildiing now, and there has been some significant sun spot activity thie week - maybe that got you...

The speed calc done by the GPS uses tiime and position - or rather the difference between two positions. The time is reliable - those satelites each have at least one atomic clock onboard. But if either position is mis-calced the speed between will be up the pole too

Sounds plausible to me...

The GPS read the make and model of bike and was then overridden by the in-built 'Stereotype Riding Characteristics' for the bike.

The 134kph educated guess had a 98% probability of being correct for that bike

You are the exception that proves the rule Mr Fox

Voted "best reply". Bling duly awarded.

[george formby]

Under "perfect" conditions, most commercial grade GPS units will be accurate to under 15m 95% or more of the time. Some of the better units will get down to about 5m

But conditions aren't always perfect - as well as a good initialisation, you need a clear view of the sky and to have at least four satelites in good configuration. Furthermore leaves, buildings and mountains can all bounce the radio signals around and confuse the receiver (try getting a good gps signal in the Otira gorge and you will see what I mean)

But also of relevance is the 'space weather' - and in particular solar activity. Sun spots seem to come in an eleven year cycle. We are starting into ta phase of increasing activity. Activity hasn't ramped up as rapidly as expected, its been unexpectdly quiet for the last two or three yeasr. But things are buildiing now, and there has been some significant sun spot activity thie week - maybe that got you...

The speed calc done by the GPS uses tiime and position - or rather the difference between two positions. The time is reliable - those satelites each have at least one atomic clock onboard. But if either position is mis-calced the speed between will be up the pole too

Er, yup. I noticed they don't work in European tunnels neither & it's an absolute sod when you come into day light again, the Sat nav is sorting itself out & you have to get into one of those 6 lanes NOW! I was moments away from spending my holiday in Germany & all my accommodation was booked in France.

I think the speed is pretty accurate though, the sat nav matched the ticket I got perfectly.

[slofox]

Does anyone here know to roughly what % accuracy most consumer level GPS is in terms of the velocity reading?

http://gpsinformation.net/main/gpsspeed.htm


Read that. There's heaps more on the net as well...

I also found mention of the fact that GPS units measure speed in 2D - as though you are on level ground. So if you are going up or down a steep hill, your actual speed may well be more than the GPS readout.

[jim.cox]

[
I also found mention of the fact that GPS units measure speed in 2D - as though you are on level ground. So if you are going up or down a steep hill, your actual speed may well be more than the GPS readout.

Wont make bugger all difference

Take one of NZ steepest roads, the Crown Range, as an example - the difference measuring the distance from the Arrow Turnoff to the Summit is less than 10% depending on whether you do it in 2 or 3 D

And most NZ roasds aren't that steep

[Dogboy900]

I have had the same thing out hunting where I have a speed of 100'skms/hr
I think it's because it loses reception for a while and then you suddenly re-appear further along.

Of course mine is a crappy old Etrex "yellow Peril" and loses reception almost as much as it has it!

[Gremlin]

I also found mention of the fact that GPS units measure speed in 2D - as though you are on level ground. So if you are going up or down a steep hill, your actual speed may well be more than the GPS readout.

GPS should also be able to tell you your elevation. You should be able to open the track in Mapsource and get a graph showing a summary of the track, ie, up and down, as you climbed and descended hills.

Your GPS may also tell you its approximate current accuracy... well, mine does...

If you want accurate tracks, if you have the option, turn off "Lock on road", as it assumes you're close to the nearest road... My zumo doesn't have that option

[Jantar]

Most modern GPS units do measure speed in 3D. Mine even computes glide angle when I'm gliding, and to do that it must calculate in 3D.

The apparent sudden high speed is when it loses lock on one or more satelites, then reaquires a new satelite. It should only appear in the track log as a single anomalous reading.

[mashman]

solar flares can affect GPS signals, or so i've read... and there just so happens to have been a couple in the neighbourhood recently ... just saying is all...