Are GPS devices accurate enough to be used as evidence in accidents?

[Cyclopathic]

If both or all parties in an accident were operating gps could the boffins use all of the data to discern who was where and when and who hit who first from which direction.
I am guessing the answer is "no" because I haven't heard about it, but why not?

 

[T.M.H.N.E.T]

Yes. No idea.

 

[Risex4]

I'm sure I've read in the past that although GPS technology is capable of centimetre grade accuracy, this is routinely reserved for military purposes, and that hardware released to the general public is purposefully "offset" or "blurred". Publicly available technology is normally only certified to 30 feet or so.

Also, GPS track files can be modified artificially, so all in all, I'd say probably not.

 

[T.M.H.N.E.T]

I'm sure I've read in the past that although GPS technology is capable of centimetre grade accuracy, this is routinely reserved for military purposes, and that hardware released to the general public is purposefully "offset" or "blurred". Publicly available technology is normally only certified to 30 feet or so.

Also, GPS track files can be modified artificially, so all in all, I'd say probably not.

I might be alone in this, but the last thing I'd think about in an accident is getting my satnav or indeed Edge800 to a computer so I could modify files.

 

[ColinJ]

They have taken the 'dithering' off the GPS signals now but the kind of devices we use are typically accurate to 5 metres or so. There are devices used in surveying etc. that are accurate to cms rather than metres but we don't use them for cycling.

I have found that when I ride an out and back route, my GPS tracklog shows me about 3 or 4 metres away from my outward journey so the device may not be accurate enough to say exactly where I was, but it is consistent enough to say where I was relative to where I had been earlier!

 

[T.M.H.N.E.T]

It's also worth actually looking at a log of a GPS. Only the most guilty of nerd would even think about editing enough of a file like this to make yourself appear slower or in a different place.

</Trackpoint>
<Trackpoint>
<Time>2013-06-02T09:44:42.000Z</Time>
<Position>
<LatitudeDegrees>54.796053199097514</LatitudeDegrees>
<LongitudeDegrees>-6.154615161940455</LongitudeDegrees>
</Position>
<AltitudeMeters>101.0</AltitudeMeters>
<DistanceMeters>56588.578125</DistanceMeters>
<HeartRateBpm>
<Value>155</Value>
</HeartRateBpm>
<Cadence>79</Cadence>
<Extensions>
<TPX xmlns="http://www.garmin.com/xmlschemas/ActivityExtension/v2">
<Speed>7.105999946594239</Speed>
</TPX>
</Extensions>
</Trackpoint>
<Trackpoint>
<Time>2013-06-02T09:44:43.000Z</Time>
<Position>
<LatitudeDegrees>54.79612042196095</LatitudeDegrees>
<LongitudeDegrees>-6.154677774757147</LongitudeDegrees>
</Position>
<AltitudeMeters>101.4000015258789</AltitudeMeters>
<DistanceMeters>56596.0</DistanceMeters>
<HeartRateBpm>
<Value>155</Value>
</HeartRateBpm>
<Cadence>83</Cadence>
<Extensions>
<TPX xmlns="http://www.garmin.com/xmlschemas/ActivityExtension/v2">
<Speed>7.276000022888184</Speed>
</TPX>
</Extensions>
</Trackpoint>
<Trackpoint>
<Time>2013-06-02T09:44:44.000Z</Time>
<Position>
<LatitudeDegrees>54.796184627339244</LatitudeDegrees>
<LongitudeDegrees>-6.154734352603555</LongitudeDegrees>
</Position>
<AltitudeMeters>102.0</AltitudeMeters>
<DistanceMeters>56603.69140625</DistanceMeters>
<HeartRateBpm>
<Value>156</Value>
</HeartRateBpm>
<Cadence>86</Cadence>
<Extensions>
<TPX xmlns="http://www.garmin.com/xmlschemas/ActivityExtension/v2">
<Speed>7.83300018310547</Speed>
</TPX>
</Extensions>
</Trackpoint>
<Trackpoint>
<Time>2013-06-02T09:44:45.000Z</Time>
<Position>

261,726 lines to edit over the 70mile ride I plucked that from.

 

[fossala]

You can check how accurate your garmin 800 is at any given time.

 

[Cyclopathic]

Going by what has been said then I'd say the answer is "no". It would appear that gps devices that we use on the roads are not accurate enough to determine who pulled out on who or other such fine details. I suppose it makes sense not to let us potential terrorists have access to gps that is accurate to a few cms. But then again I would have thought that with a bomb you wouldn't have to be centimetre accurate. I digress. "No" then. Thanks for the input folks.

 

[Canrider]

It's also worth actually looking at a log of a GPS. Only the most guilty of nerd would even think about editing enough of a file like this to make yourself appear slower or in a different place.



Nearly 261,726 lines to edit over a 70mile ride.

That would be extremely easy to edit quickly. Pull out everything between the <speed> tags (for example), remove X from each value, put them back in, and you never were speeding, officer.

 

[T.M.H.N.E.T]

That would be extremely easy to edit quickly. Pull out everything between the <speed> tags (for example), remove X from each value, put them back in, and you never were speeding, officer.

After an accident. Yes perfectly possible

 

[Canrider]

Going by what has been said then I'd say the answer is "no". It would appear that gps devices that we use on the roads are not accurate enough to determine who pulled out on who or other such fine details. I suppose it makes sense not to let us potential terrorists have access to gps that is accurate to a few cms. But then again I would have thought that with a bomb you wouldn't have to be centimetre accurate. I digress. "No" then. Thanks for the input folks.

If you look at the data dump TMHNET put up, you can see it takes a reading every second, so you don't have enough data there to guarantee you could work out who hit who from which direction when. You could work it back from a crash scene, but if you've got a crash scene, the investigator can probably *see* who hit who from where.

 

[Cyclopathic]

Perhaps if one carried say 20 gps devices then you might get an accurate positioning by averaging them all out at any given time. If the person or persons you have an accident with are also taking similar precautions then it just may be possible to build up a completely accurate timeline of all the events that took place in said accident. Easy.

 

[Canrider]

After an accident. Yes perfectly possible

Not while you're strapped to a gurney in the ambulance, but your argument was that the data file is too big to be easily editable. This is simply not correct.

(Liking your avatar!)

 

[T.M.H.N.E.T]

Not while you're strapped to a gurney in the ambulance, but your argument was that the data file is too big to be easily editable. This is simply not correct.

(Liking your avatar!)

And how would you account for, acceleration and deceleration at every single point where the data is recorded? You didn't start your journey at x speed and continue that way for all of it.

So. Too big to edit immediately,and easily prior to an accident.(and that's just one data field) I think so.

 

[HLaB]

I think that is the case the .gpx file is too long and complicated to edit for most folk but the possibility remains and some 'clever' lawyer would use that, so even if it was 100% accurate I don't think it could be used