[OSM-talk] Choice of GPS for bicycle use in London

[Andy Robinson]

All of this discussion shows how import it is, for heavy built-up urban
areas, that we (that is OSM) find a way for making some comparisons with a
set of GPS units doing the same thing in the same location. The Mapchester
weekend clearly demonstrated the limited capability of the current
technology (pre sirf star III at least) in the narrow streets of central
Manchester which are bounded by moderately high buildings. We don’t know for
sure whether bounce or the lack of sight was the major issue or whether both
in tandem were enough to confuse the largely eTrex set of units (bearing in
mind we had around a dozen of them in all). There is also the issue of user
experience and perhaps better methods of filtering the track data.

My personal belief is that we should look to map these difficult areas
differently. Much more akin to traditional surveying where you pick your
points and verify position rather than walking a track. I'll certainly be
looking more to the latter for the centre of Birmingham as a small sample
area previously done did not produce good results and was difficult to edit.

As has already been pointed out, this problem affects a tiny proportion of
locations, limited to the very heavy high rise urban landscape found in the
very centre of major cities. Almost all other places are ok with even the
basic entry level gps receivers. Best to say that for those wishing to map,
especially on a limited budget, leave these complex city centres alone for
now.

Cheers,

Andy


Andy Robinson
Andy_J_Robinson at blueyonder.co.uk 

>-----Original Message-----
>From: talk-bounces at openstreetmap.org [mailto:talk-
>bounces at openstreetmap.org] On Behalf Of Nick Hill
>Sent: 22 May 2006 13:43
>To: Erik Johansson
>Cc: Talk Openstreetmap
>Subject: Re: [OSM-talk] Choice of GPS for bicycle use in London
>
>There has been a lot of talk about the sirf star III chipset being so
>much more useful. They can after all pick up GPS signals in situations
>where other chipsets will not receive a signal.
>
>This is not always a good thing. If you are in a concrete canyon with a
>sirfstar III chipset, the fact that it can pick up a GPS signal shows
>the signals it is using are bouncing off buildings. It is receiving
>multipath signals. Such multipath signals are bad for accuracy.
>
>I say it is better to receive no signal rather than receive multipath
>(unless you are stuck down a mountain cravass and need a rough fix).
>
>IMO, a receiver which is only sensitive enough to receive a line of
>sight signal has advantages in urban areas. ie for OSM.
>
>I have seen tracks go way off course in my area, which I beleieve to
>have been captured with a SirfStar III. My Geko, OTOH, rarely lays
>off-course tracks.
>
>The on-screen stated accuracy of my Geko is normally between 15ft-23ft.
>Actual lat/lon accuracy is better than that. It is nearly always clear
>which side of a small urban road I am on, even on different days, so
>long as I keep it still when I turn it on, until it gets a good fix.
>Don't bother turning the Geko on whilst moving.
>
>My tests show Etrex Yellow=Geko 201.
>
>Difference is that Etrex is bigger, has bigger batteries so lasts longer
>on a charge.
>
>I highly recommend the Geko. It is small, light, tends to lay accurate
>GPX traces, and is not sensitive enough to receive most multipath. The
>down side is battery life. In normal mode, with 900mAh Uniross
>rechargeables, I get 9.5Hr. In battery save mode (some loss of accuracy)
>I expect 14 hr.
>
>
>
>
>
>Erik Johansson wrote:
>> On 5/22/06, Christian van den Bosch <cjb at cjb.ie> wrote:
>>
>>> Emil Vaughan wrote:
>>> > Would a "usb mouse" type gps receiver
>>> > connected to my zaurus be a more accurate than a geko? And would it
>>> > work ok under linux?
>>>
>>> I haven't tested it yet, but I believe that a Sirf Star III should be
>>> /much/ more accurate than a Garmin log (because it won't lie,
>>
>>
>> Actually I'm not so sure if  my GPS puck^W mouse is lying. It has just
>> happend once:
>>
>>
>http://www.openstreetmap.org/edit.html?lat=59.33833563680978&lon=18.0867943
>7637329&zoom=16
>>
>>
>> There you have a diagonal line that starts between "Kommendörsgatan"
>> and "Karlavägen", and then drifts up north east. The only explanations
>> I have is that it's extrapolating.
>>
>> This is done with a sirf III and a sirf II the sirf II just lost it's
>> GPS lock. And the SIRF III PDOP is at 50 during that line.
>>
>> /emj
>>
>> _______________________________________________
>> talk mailing list
>> talk at openstreetmap.org
>> http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>>
>
>_______________________________________________
>talk mailing list
>talk at openstreetmap.org
>http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk