Printable View
I'm having the problem when carrying my smartphone in a thigh pocket and using my loacl GPS app in tracking mode, the sum of vertical gain is calculated wrong. When carrying the device in any other pocket, the sum is correct.
Obviously the constant up&down while walking (and more so when biking) of my leg irritates the device.
Anybody else ever noticed something like this?
Yep, all the time. My wife uses Gaia, I use Backcountry navigator. Some hikes, they are spot-on. Others, they vary widely. Really kinda random, I suppose it's all about how well each phone is picking up the GPS signals, varies a lot depending on which pocket, how much southern-sky exposure we have or whatever.
Since, in the northern hemisphere, the GPS signals are coming from sort of south-ish in the sky, perhaps if we kept out phone in our south-ish pockets all the time, perhaps this might make a slight difference, but who knows. Just a random thought there....
I do know at the end of the day, the track from my smartphone is pretty darn accurate, but elevation is kinda a different animal. Really hard for a phone, or any other GPS device to keep elevation change perfectly accurate, at least from a running total standpoint, as the perceived elevation probably dithers around a bit.
Obviously its a common thing.
What I've noticed so far:
My wife and me carry the same device, and use the same app. She carries the device in a waist pocket, I carry mine in the thigh pocket.
When hiking together, hers shows the correct altitude gain, mine about twice as much. Swapping the device shows the bug being in my way of carrying it.
Within the last week I did the same bike tour twice.
First time, I carried the phone in my backpack, the second time in my thigh pocket.
First trip, the tracking showed the correct Elevation gain (770m), second time it showed 1880m.
Again, the bug is in the way I'm carrying the device.
The long+lat coordinates of the track are always spot-on, as well as the altitude any time I check it against the map or a sign at a summit or other prominent point.
Its just the added altitude of the whole track thats completely off.
Carrying the smartphone in another place would be a poor option, the thigh pocket is just perfect for my everyday use.
In a GPS forum dealing mostly with dedicated GPS devices I read something about filters, that many GPS software has built-in. Unfortunately there is nothing like that in the smartphone apps I'm using, at least not in the docs.
Not sure about a filter, but a "running average" software scheme would probably fix this. As I said below, the GPS is dithering in altitude by who knows, maybe 10 meters at a time when coverage is not perfect, even though the trail is level or only a gradual slope. The software probably just adds up all those dithers hence the big errors. If it did a running average, this problem would smooth out and the errors would drastically decrease.
Really, this is all just an educated guess; This is very analogous to what I did for a living for 30+ years, analyzing flight data from launches. We nearly always had to do running averages on various forms of data (accelerations, positions, etc).
Yes, filter or running average or whatever it is, would be the solution.
My smartphone (as all more recent models) has a barometric sensor, so the altitude data does come from several sources, like GPS (wich is pretty inaccurate), the barometric sensor, and (I assume) from map data. All our country got laser scanned a few years back and the elevation data is down into the fraction of meters accuracy.
I belive the barometric sensor is the one that is detecting the up and down of my legs while walking, and more so while biking, and this gives all the additional altitude gain, based on wrong sensor values that are not filtered or averaged out.
Thanks for confirming that pretty much the same you're observing.