ओपनस्ट्रीटमॅप संकेतचित्र OpenStreetMap

On plotting points

GerardSharpने 18 August 2010 ला English भाषेत लिहीले

Okay, the accuracy of my GPS unit is "reasonable". It can take a very long time to start locking and it occasionally gets weird along the way, but it's got most things about where I would expect to find them.
In a car, it's nowhere accurate enough to plot bus stops.
Bus Stops in Hamilton do not have any identifying numbers on them indicating route either. I guess I can mark where the pole is...

Looks like the most fun I can have is going to be on foot, walking around plotting where things are.

Email icon Bluesky Icon Facebook Icon LinkedIn Icon Mastodon Icon Telegram Icon X Icon

Discussion

dcpकडून 18 August 2010, 11:04 लाला अभिप्राय

Hi,Gerard.
First welcome to the OSM community. I bought my GPS-gear 2 years ago and I too was very uncertain of its accuracy (even when it said +/- 3meters). It took me a very long time before I became confident that my plots were good enough to match with others. I suppose the accuracy may be dependant on the GPS-Receiver one buys: I just don't know!
What I did was to map streets many times, backwards and forewards. My plots were hardly ever exactly over oneanother, but you do get a feeling for these inaccuracies. You are probably using Potlach which everyone starts with. Try to graduate to JOSM and after you have got use to it then download the user tracks with the OSM-map data. You will then see how inaccurate the others are compared with yourself. Then you don't worry about it anymore
dcp

GerardSharpकडून 18 August 2010, 13:07 लाला अभिप्राय

Hi/thanks.
I have seen from my traces how accurate the unit is in general - most of the road lines are roughly where the road is; a systematic offset of 0-2m I can expect and fudge out; but what I'm thinking more of is the sudden unexplained leap 10 metres off to the side of the road - I don't recall twirling the wheel and driving through anyone's house! so it must be my cellphone/GPS thing. Probably it's getting a bad reading through interference or weak signal and so it's just doing the best it can with what it's got!
Solution is to move slower (walk!) and travel the same route multiple times to get more data points.

I started on JOSM and I think I prefer it to Potlach.. JOSM is offline and outside the browser; Potlach just irks me with their method of entering ways. I like/miss my right mouse button.

Alright, see you on the edit lists! :)

Harry Woodकडून 18 August 2010, 13:37 लाला अभिप्राय

The radio waves bounce off buildings, particularly reflective glass fronted buildings. Things like this will cause your location to jump, as the GPS unit corrects itself later on. The device will tend to extrapolate a little bit from your previous movements, so guessing which direction you're going and how fast, all based on a fuzzy picture it builds up from signal timings it receives as input.

That's GPS for you. Sometimes you have to mix in a bit of guesswork, dead reckoning. Don't be assuming that you GPS trace is more accurate than anyone elses, but by the same token, don't be afraid to edit and add OpenStreetMap data even if you're a little unsure of the accuracy.

marscotकडून 18 August 2010, 14:30 लाला अभिप्राय

Welcome,, is that Hamilton Lanarkshire, if so I am just up the road from you, while you mark off the bus stops grab POI too like banks ect.

Opkकडून 18 August 2010, 16:08 लाला अभिप्राय

Most GPS devices use the doppler effect to measure speed in addition to triangulation for your position. Typically, the doppler data is actually the more reliable. Often the position your GPS reports comes from the result of applying your velocity to your last known position. The effects of atmospheric conditions can also often result in an error that remains consistent for a GPS track. So when interpreting GPS results it is reasonable to apply relative offset adjustments.

davespodकडून 18 August 2010, 16:09 लाला अभिप्राय

Some reviews of my own GPS logger (iBlue 747A+, based on MTK chipset) say that it is not well calibrated for walking. If this is true (and I've no idea whether it is), some units may actually get worse results walking, but no doubt others might get better results. I use mine while walking in spite of this, but I seem to get the best results of all cycling. I'd echo what others have said about not worrying too much, though.

However, your final statement is true in general - much more fun doing your surveys on foot (or bicycle) than driving.

marscotकडून 18 August 2010, 17:49 लाला अभिप्राय

I agree its more fun on foot.

GerardSharpकडून 18 August 2010, 23:08 लाला अभिप्राय

@marscot: Nope, Hamilton New Zealand. But I've started marking Supermarkets and Fast Food places (big and easy to find from Yahoo's lowres pics) so if you ever come for a visit, you can find food!
@everyone else: I expect that as the GPS radios are coming from space, there's probably a fair few atmospherics and multipath and reflection that play with it. Indeed, I'm surprised it works at all. :)

टिप्प्णी लिहिण्यासाठी सनोंद-प्रवेश करा