GPS leading your phone astray? We can just fix that in code, startup claims A startup that's just emerged from stealth claims it's solved smartphone GPS positioning problems by adding software to triangulate a handset with "sub metre absolute accuracy." Colorado-based Zephr has decloaked to tell the world about its "networked GPS" idea, which it claims in field trials showed sub-60cm GPS accuracy …
They say this is for cell phones and then say that the use case is for autonomous vehicles, etc. That sort of thing should already be using WAAS, which gives pretty good accuracy. But you can't rely on GPS for avoiding the environment, especially other people in it. So what are they really up to? Is this going to let them sell more adverts?
You can already use differential GPS if you need accuracy, which is the sort of thing you might do if doing some sort of survey-by-drone arrangement. But I guess this idea could allow you to get improved - but not as good - accuracy, whilst needing to be slightly less organized about how you achieve it: :-)
WAAS may be adequate for general use (free) and Differential for survey grade 1cm accuracy at the cost of specific equipment ££. The RTK equipment sits between those as a 10cm accuracy at more modest cost. So very suitable for vehicle guidance e.g. Combine Harvesters or Tractors.
This innovation has potential for RTK level results with only software rather than RTK equipment. Effectively the other handsets are that equipment.
What would be given to enable that and is it clearly described in the privacy agreement?
RTK GNSS has no need for network and data sharing on the internet.
Exactly, GPS I've seen quoted to 10cm accuracy 99.99% of the time.
So the other 0.01% of the time, it propels me/you into the other lane?
On a 10 mile drive, that's once per journey, probably by someone I'm passing.
There is still a long way to go. We need 6*9's availability.
The accuracy of your car sat nav is nowhere near 10cm, 3m if you are lucky. It only appears more accurate because it can snap you to not just the road, but the correct side of the road depending on the direction of travel - most of the time.
I had an early sat nav where you could turn off snap to road, it was interesting driving along and seeing its apparent position wandering off over fields.
"the use case is for autonomous vehicles"
Not really - a disaster for an autonomous vehicle just relying on GPS and its maps - need to react to changes in roads that will not be on its map - be that a new road, or (if we assume fantasy world of fast, frequent & accurate map updates) things like contraflows due to roadworks where you can easily end up on the "wrong" road e.g. on motorway work, in some worst case scenarios* may have all of Northbound motorway closed and traffic heading "North" forced to use a (coned off) lane of the "Southbound" motorway (& traffic heading South forced to have less lanes).
Other fun situations are where roads are at different heights, so in our nice 3D world can get stretches where road A is above road B for a while (in some cases a few 10s of metres), but can be a few km sometimes. These scenarios have caused confusion in my SatNav where it gets confused on what speed limit should be - will it decide on the 70 of the A road dual carriageway I am on or the 30 of the B road below.
Any autonomous vehicle should be fine without GPS (just like a human, people managed OK without SatNavs for a long time & any driver with any sense treats them as a useful tool when travelling somewhere unfamiliar, but definitely not as an infallible source of road truth) - and be able to cope with just "looking" at the road layout and signs,
* 2nd worst case, worst ones are when whole Motorway closed in both directions... But worst case as far as autonomous cars are concerned.
Personally I think a self driving car should use GPS only the same way as I use it. It should be able to drive safely along any road without any GPS like we did for many years. Or if an inaccurate GPS or map data tells it “you are 12 meters left of the road” then it should look at its cameras and say “no, I’m right on the road” just like I would.
The GPS head unit I use on my bike yields far more accurate results (to within low single digit metres and quite repeatable) than the output of <social network masquerading as a fitness tracking app> (which is low double digit metres at best).
Why is this? Are the GPS receivers in phones that bad, or is it simply the app being terrible?
The GPS on my phone typically struggles if I am in or near buildings. I guess if you like this sort of thing, it being able to query a nearby somephone that gets a better view of the sky, to help out with positioning, might be considered useful by some.
This post has been deleted by its author
We were staying in a hotel recently where we were parked in a room at the end of the corridor that was also at the end of a corridor. Not a big deal (apart from the hike for breakfast) except it caught my wife out at least twice because her phone not only said the hotel was on a different street but also turned 90 degrees from where it really was. The GPS was accurate, of course -- it was showing us the coordinates and orientation of our room -- but that's not a great consolation when you're looking for the entrance to the hotel on the wrong street.
An old-fashioned paper map would have just shown the hotel and left internal navigation to the floor plan.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
