[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [gpsd-users] Coding Question

From: Gary E. Miller
Subject: Re: [gpsd-users] Coding Question
Date: Tue, 5 Feb 2019 12:15:26 -0800

Yo Roger!

On Tue, 5 Feb 2019 08:15:02 +0000
Roger Oberholtzer <address@hidden> wrote:

> >> I need RTCM because I am working on Intelligent Transportation
> >> Systems and I need to have individual lane accuracy.  
> > RTCM would only get you lane accuracy if the base station is
> > no more than a few kilometers from the rover.  
> We have found that one also needs a system with an IMU, like
> something from Oxford Technical Solutions or Applanix.

Or the cheap u-blox/IMU combos sold for drones.  The problem is they
drift off center very fast.  Ok, for mementary underpasses, not much

I have seen $70k IMU, and even they drift a bit.

> A GPS-only
> solution simply does not provide the needed accuracy.

Unless you have a very close base station, or do PPP.

> At least not in the real world.

Best I can do, L1 only, is around 1 meter.  But with a local
base station I can get to 1 centimeter.

> I consider GPS salesman to be the equivalent of used
> car salesmen. They will claim anything. The truth is in the fine
> print.

Well, salesman are another thing.  I'm talking about my real results.

> We measure vehicle location at driving speed and require sub-meter
> accuracy. We get it when we get it. And we don't when we don't.

Makes me feel real safe.  Not.

Some of the u-blox have wheel sensors to improve the fix.

> Oxford Technical Solutions do a lot with autonomous vehicles.

The death rate per mile for autonomous vehicles is 20x for manned
vehicles.  Not Ready For Prime Time.

Gary E. Miller Rellim 109 NW Wilmington Ave., Suite E, Bend, OR 97703
        address@hidden  Tel:+1 541 382 8588

            Veritas liberabit vos. -- Quid est veritas?
    "If you can’t measure it, you can’t improve it." - Lord Kelvin

Attachment: pgpMcd5XPOU0T.pgp
Description: OpenPGP digital signature

reply via email to

[Prev in Thread] Current Thread [Next in Thread]