GPS (not a/c question!)

[PAXboy]

Trusting that some GPS savvy person is to hand ... a friend wonders about her car's Sat-Nav. "It's tells me the speed I'm doing but when I'm going up a steep hill - how doe sit work out the speed?"

My reply was, "The distance between you and the satellites changes so little (relatively) that it makes no difference. Also, the device will be triangulating with three or more signals."

Am I on the right line?

Cheers.

--------------------
"I tell you, we are here on Earth to fart around, and don't let anybody tell you any different." Kurt Vonnegut, Jr.

[Paul Wilson]

Could be using loads of signals, my antique Garmin GPS12 uses up to ..... er 12 ;) , also the GPS unit does work out height, although this is the least accurate function. So knowing speed between A + B in plan view, plus increase in height it can generate an accurate speed.

Saying all that, unless the hill is VERY steep for a long time the effect of going up a hill is marginal on speed readings.

i.e imagine a 1 in 10 hill (10%) a kilometre long, as against a flat piece of road a kilometre long.

car covers each distance in 1 minute

Flat distance = 1 km => speed = 60km/h

Hill distance using pythagorus (sp x10 see me:hmm: )

height of hill = 100m length = 1000

=> (100 x 100) + (1000 X 1000) = square root distance travelled

=> 10000 + 1000000 = square root distance travelled

=>1010000 = square root distance travelled

=> distance travelled = 1004.987m

call it 1005 metres = 60.3km/h

or 1.005 km as you can see practically no differance in distance travelled, and well within the error tolerances of speedometers, so you would never notice the differance even if the GPS was just telling the speed over "flat" ground, and not allowing for hills.

-----

repeated for a 1 in 4 hill (25%) -- seriously landrover steep

gives figures of

flat distance 1000m = 60mk/h

hill distance 1030.776m still only a 3% differance = 61.84mk/h

not noticable
-----

repeated for a 1 in 1 hill (45degree angle suicidal nutters on scramblers only)

flat = 1000m

hill = 1414.21m you'd probably notice that differance, but that would be the least of your problems

:D

[PAXboy]

Totally wonderful, many thanks PW. :ok: Good to know I was on the right road (Huh-Huh-Huh) :rolleyes:

[twistedenginestarter]

I think car sat-navs have wheel sensors. GPS alone would not be accurate enough or reliable (think of shielding in built-up areas)

[farqueue]

If a GPS has a good set of sats, it generates a 4D solution to its position. The terms of this are X, Y, Z, deltaX, deltaY, deltaZ and t. The 3 deltas
can be easily used to compute a speed or velocity. Note you also get an
acurate time.

On height, because someone left a sodding great lump of rock littering up
the signal path of all the sats `down there', you can only get signals from sats that are above you, as opposed to being able to see sats anywhere around the horizontal. This is why height is harder to get good accuracy for.

[Paul Wilson]

They could well also use speed sensors, but I wouldn't have thought it was necessary, as you can buy "standalone" sat-nav that just sits on top of the dash, and the delay on the speed system on my sat-nav has a 1-2 sec refresh rate, which would imlply that it was not getting "outside" help.

Just wish there was a mute button for the voice, as I don't go the way she says for the first bit of any journey (because I know my home town better than she does).

[BEagle]

Thinking of getting a Blaupunkt TravelPilot R52 fitted to the current set of wheels, together with a hands-free for my phone plus a CD changer. Biggest problem is where the heck to put the aerial(s) on a SLK!

These in-car nav systems are getting very clever. They now contain gyros and use the wheelspeed feed which runs the 'auto volume speed compensation' to assist with the navigation algorithm.

Remember signposts and milestones?

[Sirius Flying]

Actually, GPS receivers usually use the doppler frequency shift to determine speed (or velocities, whatever..). This gives ---very--- precise speed measurements (around +/- 0.01 meters/second), provided that enough satellites are available.