Mailing List Archive

wireless wrote:
> I have not formally researched open source mapping efforts, so
> any feedback on that issue is keenly welcome.

Start with http://openstreetmap.org/


//Peter

On 04/14/12 14:38, Peter Stuge wrote:
> wireless wrote:
>> I have not formally researched open source mapping efforts, so
>> any feedback on that issue is keenly welcome.
>
> Start with http://openstreetmap.org/
> //Peter

EXCELLENT!

OK so to help create maps, I'm think of designing (cobbling together
initially) a gps, and a (water) depth reading output, so in a boat, I
can begin mapping the near shore environment.

How does open street map data get created and are there existing
open source embedded boards one can buy, install in a car and help
extend/enhance the data collected? Or, roll your own?

It seem they already have the land methodologies developed, what
about near-shore marine? Any open source software efforts for that
yet?

In the past I did run across some for when you are out in the
wilderness, but, I did not save the links.....

I just got my account, so I'll be looking into
opensteetmap...


James

On Saturday 14 April 2012 14:56:42 wireless wrote:
> How does open street map data get created and are there existing
> open source embedded boards one can buy, install in a car and help
> extend/enhance the data collected? Or, roll your own?

probably better to ask in their forums
-mike

On Sat, Apr 14, 2012 at 02:56:42PM -0400, wireless wrote:
> On 04/14/12 14:38, Peter Stuge wrote:
> > wireless wrote:
> >> I have not formally researched open source mapping efforts, so
> >> any feedback on that issue is keenly welcome.
> >
> > Start with http://openstreetmap.org/
> > //Peter
>
> EXCELLENT!
>
> OK so to help create maps, I'm think of designing (cobbling together
> initially) a gps, and a (water) depth reading output, so in a boat, I
> can begin mapping the near shore environment.
>
> How does open street map data get created and are there existing
> open source embedded boards one can buy, install in a car and help
> extend/enhance the data collected? Or, roll your own?

Most people around openstreetmap are software / geography people, so
they mostly use what's readilly awailable (GPS loggers, JavaME phones
some time ago, smartphones now). I've seen hardware projects on the
level "how to make a bike holder for XYZ" or "how to get a PDA and
logger work together and with a large battery". There's a hardware page
on the wiki [1] (if you haven't found it already), but it seem pretty
outdated. As osm is opensource centered, there are/were people playing
with open hardware / technologies stuff (openmoko phones, beagle/panda
board with a gps receiver etc...)

There's also a foss-gps mailing list [2], though IIRC there are mostly
software guys who talk about how to work with lower level GPS data.


> It seem they already have the land methodologies developed, what
> about near-shore marine? Any open source software efforts for that
> yet?

There are basically two distinct activities in OSM: the collection of
GPS logs/tracks and then the creation of the map/database itself based
either on those logs or any other freely available sources. For land
based information the elevation doesn't really get entered into the
database, mostly because it would be realy sparse and there is SRTM data
that's much better. I remeber some talk about an initiative to somehow
process the elevation data from the logs, but I'm not sure where it
ended.

Anyway I guess it would be different for marine data, which might be
more usefull... I guess you have to get in contact with people from
openstreet map interested in that. The OpenSeaMap page on osm wiki [3]
mentions even some mailing lists...

yoyo


[1] http://wiki.openstreetmap.org/wiki/Hardware
[2] http://lists.osgeo.org/mailman/listinfo/foss-gps
[3] http://wiki.openstreetmap.org/wiki/OpenSeaMap

On 14/04/2012 19:34, wireless wrote:
> So, I'm looking at options to build/hack a
> GPS device, combined with some sort of
> embedded gentoo friendly Tablet.
>
>
> I'm thinking of multiple channels
> of audio output so it is a 5.1 mp3 player too,
> for a boat or a RV or an car. Mobile so it can
> be moved between vehicles with a quick connect
> to sensors and speakers.

Nokia N900 or new N9 (I have an N9, but the N900 probably has more
hackers around it and cheaper second hand items)

Build something around a gumstix - not that cheap, but I think they have
all the bits ready to plug together. Probably low enough power you could
battery power it

Very easy to hack is a PC Engines Alix + laptop screen. (In some ways
that's called a 1 laptop per child I guess.... Add that to the list)

Good luck finding a Gentoo friendly tablet? I would be interested if
you find something?

Good luck

Ed W

On 04/16/12 08:42, Ed W wrote:
> On 14/04/2012 19:34, wireless wrote:
>> So, I'm looking at options to build/hack a
>> GPS device, combined with some sort of
>> embedded gentoo friendly Tablet.
>>
>>
>> I'm thinking of multiple channels
>> of audio output so it is a 5.1 mp3 player too,
>> for a boat or a RV or an car. Mobile so it can
>> be moved between vehicles with a quick connect
>> to sensors and speakers.
>
> Nokia N900 or new N9 (I have an N9, but the N900 probably has more
> hackers around it and cheaper second hand items)
>
> Build something around a gumstix - not that cheap, but I think they have
> all the bits ready to plug together. Probably low enough power you could
> battery power it
>
> Very easy to hack is a PC Engines Alix + laptop screen. (In some ways
> that's called a 1 laptop per child I guess.... Add that to the list)
>

Great ideas.

> Good luck finding a Gentoo friendly tablet? I would be interested if you
> find something?

First generation, I thinking Arm9, maybe a pandaboards with some LED
screen mounted together with 2" nylon standoffs. That way, when
I figure out how to add a board that receives and processes signals
from the depth finder, I can use 12 or 14 bit AD to input into
the panda board.

I may also rig a temperature sensor to the boat hull, or via a tether.

What concerns me about openstreetmap, is I had envisioned a system where
the database is downloaded to the operator's embedded system depending
on the anticipated area of local/regional travel, thus not having to
depend on constant/reliable internet access.

Sort of a GPS/nautical navigation system, under the control of the
operator. I'm still learning about openstreetmap so I have moved those
issues (as suggested by our Mike) to that forum, for now.

From my initial research so far, nothing exist for a marine
environment, so I'll most like roll-it, including some epoxy
coatings and daughter board on the first-gen effort.

My reason for using this list, is to flush out any gentoo-ish
common ground or any hardware suggestions; not an expose
on openstreetmap....

> Good luck
> Ed W

It's liable to be a few months to cmplete this research, but,
I'll post back, or drop you some private email, if you are the
only one interested in what I learn.


James

wireless wrote:
> What concerns me about openstreetmap, is I had envisioned a system where
> the database is downloaded to the operator's embedded system depending
> on the anticipated area of local/regional travel, thus not having to
> depend on constant/reliable internet access.

You can do that as well. I think you need to spend more time with the
project than one day.


> Sort of a GPS/nautical navigation system, under the control of the
> operator. I'm still learning about openstreetmap so I have moved
> those issues (as suggested by our Mike) to that forum, for now.

It's a good idea.


> From my initial research so far, nothing exist for a marine
> environment, so I'll most like roll-it, including some epoxy
> coatings and daughter board on the first-gen effort.

Do AD as close to the sensor as possible. I would use a digital bus
back to the pandaorwhatever.


> My reason for using this list, is to flush out any gentoo-ish
> common ground or any hardware suggestions;

I guess not much there.


> It's liable to be a few months to cmplete this research,

Yes.


//Peter

> I may also rig a temperature sensor to the boat hull, or via a tether.

If you want this to scale to potentially commercial end users, I would
consider an NMEA based off the shelf sensor? Although the prices are
quickly > £100, ie non trivial, NMEA 0183 sensors are plentiful and
NMEA2K is getting rapidly available (add an Actisys bridge to get data
into the computer)

> What concerns me about openstreetmap, is I had envisioned a system where
> the database is downloaded to the operator's embedded system depending
> on the anticipated area of local/regional travel, thus not having to
> depend on constant/reliable internet access.

Can you not download openstreetmap?

Your bigger issue is that it's not really a nautical mapping system?

In the USA you can have all the raster nautical and aviation maps for
free (cool). The argument goes that the tax payer already paid for
them, so the data is made completely freely available


> My reason for using this list, is to flush out any gentoo-ish
> common ground or any hardware suggestions; not an expose
> on openstreetmap....

I think you should consider some other options also:

- Small embedded server which collects the information, with a handheld
system then connecting via http or similar. I like the PC Engines Alix
boards.
- OpenCPN customised for your needs

The top one is similar to something I'm working on. We
(http://www.mailasail.com) are building a small embedded router for
connecting all the communications bits on a boat together and we can
easily take in NMEA and repeat it via GPSD. I plan to either write, or
pay someone to write, a fancy javascript based instrument console
something along the lines of the ipad "NMEARemote" app. The goal would
be to use all the responsive magic that's easy to get these days to make
it scale from a phone to tablet screen.

I think with a decent browser on your tablet you can fairly easily build
decent web apps these days, especially if you don't need to target IE.
To my eye this dramatically simplifies the problem because you can use
off the shelf components and focus on the software

As an aside if anyone here is interested in extra work I have a fairly
big backlog of web based work. Probably need backbone+jquery skills
with a Rails/Mojolicious backend to talk to

Good luck

Ed W

On 04/17/12 05:15, Ed W wrote:

> I think with a decent browser on your tablet you can fairly easily build
> decent web apps these days, especially if you don't need to target IE.
> To my eye this dramatically simplifies the problem because you can use
> off the shelf components and focus on the software

> Ed W

Ed,

You have given me lots to think about. No doubts I do not want to
reinvent the wheel. I have some (consulting) deliverables to finish up
before I can devote myself to this pursuit.....

I'll be in touch.


James