[OSM-talk] Vehicle Tracking Systems

Robert (Jamie) Munro rjmunro at arjam.net
Mon Apr 2 19:02:12 BST 2007


-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Robert Scott wrote:
> On Monday 02 April 2007 12:34, Phil Winstanley wrote:
>> Just curious, anyone out there working on vehicle tracking systems?
> 
> Depends what you mean by vehicle tracking systems.
> 
> Coincidentally enough, just this minute I've been walking around, scratching 
> my head and considering what would be involved in building a generic position 
> deduction framework which could accept input from any source of 
> positioning 'hints', take into account any sources of error and give the best 
> possible guess it can on your position.
> 
> Such hint sources might include GPS, car odometer hookup, accelerometers, 
> gyroscopes, cameras with image recognition algorithms...
> 
> The obvious first step would be attaching an error value to every single 
> number in the simulation and calculating those in parallel to the positioning 
> calculations. But then you'd possibly need quite complex heuristics for what 
> to do when several of your sources disagree to a degree greater than the 
> error margin.
> 
> I imagine papers have been written on this subject, but I have yet to find 
> them. Any input on this ugly brain-dump of mine would be welcome. 

Take a look at interval arithmetic.
http://research.sun.com/minds/2004-0527/
http://research.sun.com/projects/dashboard.php?id=164

There's some really good ideas there.

Robert (Jamie) Munro
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.6 (Darwin)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org

iD8DBQFGEUUhz+aYVHdncI0RArmAAJ9brgF0RTY+pm3m6EoNKgqPQpe4VACg6t6Z
opx5yORRraV0NEjawPcgJtA=
=FvH5
-----END PGP SIGNATURE-----




More information about the talk mailing list