Confidence in GPS

Cornishman

Cornishman

Well-Known Member
Joined
29 Jul 2002
Messages
6,402
Location
Cornwall
Visit site
For those who are not members of the Royal Institute of Navigation, there is an alarming report on the web-site and in "Fairway", the Small Craft Group newsletter, of a serious malfunction of GPS on the 1st of January 2004.
GPS relies on atomic frequency standard (AFS) for accuracy and an anomaly occurred when this failed for about 3 hours on one satellite. "AFS can significantly degrade navigational accuracy" and results in the transmission of "Hazardously Misleading Information"
On this occasion it must have been immediately obvious to navigators, the position error from the affected satellite being around 300km by the time it was set to "unhealthy".
It is not infallible!

<hr width=100% size=1>
 
Worrying - I've started using some electronic charting software on an old laptop hooked up to the GPS. I find it so good that if I'm not careful I have a tendency to forget to use other methods of determining exactly where I am as I was taught!

Just out of interest, would differential GPS either terrestrial or satellite have corrected for this rogue position?


<hr width=100% size=1>Gavin
 
Terrestial DGPS would not pick this up unless it had RAIM (integrity monitoring) but WAAS/EGNOS would.

<hr width=100% size=1>
 
On a crossing I always plot direction and speed and do a rough dead reckoning every hour before I check the GPS. This way a major anomally will be self evident. If the results are close I use the GPS position.

John

<hr width=100% size=1>
 
Tome, can you explain this a bit further for me. As I understand it, the terrestrial DGPS station knows it's position on the earth very accurately and therefore can calculate errors in individual satellites position data and transmit this information to a DGPS set. Whats the reason for it not applying a correction to this errant satellite or doesn't it work quite like this?

Just interested in how all this stuff works.

Thanks

<hr width=100% size=1>Gavin
 
Tome, can you explain this a bit further for me. As I understand it, the terrestrial DGPS station knows it's position on the earth very accurately and therefore can calculate errors in individual satellites position data and transmit this information to a DGPS set. Whats the reason for it not applying a correction to this errant satellite or doesn't it work quite like this?

Just interested in how all this stuff works.

Thanks

<hr width=100% size=1>Gavin
 
Yes I think I would do something similar but I haven't done any real passages with it yet. Poodling around the Southampton area I tend to be more concerned with traffic and a quick glance at the screen to see where you are is quite nice.

Cheers

<hr width=100% size=1>Gavin
 
It does attempt to apply a correction, but as this error affected the timing of the satellite and wasn't marked as 'unhealthy' it would continue to attempt to calculate corrections until the size of the corrections became excessive. It would also degrade the corrections it calculated for all other sats as the satellite was still being used as part of the fix solution.

A system with integrity checking would have detected the error and have flagged the solution as bad.

<hr width=100% size=1>
 
Hi John

Also as in the old days with Decca, wild errors show in the course to steer, COG & SOG figures. Remenber the days when a thunderstorm, not necessarily even local to you but to the transmitter, got speeds >100kts backwards? Small errors are much less important offshore but could be very much so in a narrow channel in fog, though radar would be the confirming/questioning aid here. We plot on paper as well as electronic charts, using a Yeoman so there is a visual comparison versus the expected course and anticipated cross track error from the passage planning. We have a cockpit GPS repeater AND a separate C-Map plotter outside so routinely scan DTW/BTW/COG/SOG and would probably pick up any meaningful error I hope.

Robin



<hr width=100% size=1>
 
I think the old advice still holds - never trust just a single source of information. Apart from keeping an EP going, it makes sense to take every chance to cross-check with any other position line available (handbearing compass, radar range, depth contour, sun sight or whatever).

<hr width=100% size=1>
 
Yes of course though in reality this need not be done continuously IMO. We carry all the things you mention including a sextant (with a sight program on the laptop AND on a Casio, plus the tables..) Checks by other means would realistically be brought into play if there was a reason to suspect the GPS data (strange or erratic COG/SOG/DTW/BTW or CTE). We actually have 3 fixed GPS sets operating, 2 of these are WAAS/EGNOS ( see Tome's post, though we currently have EGNOS disabled until it is fully up and running, it avoids the regular EGNOS warnings that it is or isn't receiving).

Bearing in mind yacht speeds even on our greyhound are pretty slow in relation to aircraft which pretty well rely 100% on GPS these days I understand.

Finally of course there is the mark 1 eyeball, if it looks like Libenter Buoy though the GPS says it is St Paul's Cathedral, smell a rat!

<hr width=100% size=1>
 
By the number of replies it looks as if it was worth repeating.
Spring edition of Fairway arrived today.

Best wishes
Mike

<hr width=100% size=1>
 
You must log in or register to reply here.

Share this page

Top