Sponsor Message:
Aviation Technical / Operations Forum
My Starred Topics | Profile | New Topic | Forum Index | Help | Search 
FMS Nav Accuracy  
User currently offlinephunc From United Kingdom, joined Mar 2011, 140 posts, RR: 0
Posted (1 year 6 months 1 week 6 days ago) and read 3514 times:

Few questions about FMS nav systems;

1. Generally speaking, the crew will input an RNP value into the FMS. E.g. RNP5. This will be displayed on the FMS MCDU along side an "actual" readout. Based on this, the A/C is trying to maintain RNP5 but the actual maybe 0.2NM. How does the A/C know it is maintaining RNP0.2NM? Is it based on INS inputs? Or a mix of inputs from the INS / DME/DME or VOR data?

2. Following on from the question above, if the aeroplane is oceanic with no radio nav in the area, the only backup method of navigation is INS. Is that correct?

3. Following on again, when the aeroplane is back in VOR / DME/DME range will it begin to recalculate and backup it's position based on these now available navaids?

4. In the event of an INS failure while oceanic, GPS is the only method of nav. What would generally happen in this case?

Many thanks!

29 replies: All unread, showing first 25:
 
User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 1, posted (1 year 6 months 1 week 5 days 19 hours ago) and read 3445 times:

Quoting phunc (Thread starter):
2. Following on from the question above, if the aeroplane is oceanic with no radio nav in the area, the only backup method of navigation is INS. Is that correct?

Goodness no. There are lots of backups.

- GPS (if so equipped). This may be wired into the FMS as an input.
- Dead reckoning with magnetic compass and stopwatch. Magnetometer input if you have one.
- Pilotage if there are any landmasses around and you can see them. Not the most accurate in most cases but better than nothing.
- In the olden days, celestial navigation with a sextant. Heck, you can still very roughly use the sun without a sextant if you are really really lost. But at that point someone has smashed everything including the compass so methinks you have bigger problems.
- It is pretty likely there is someone on board with a smartphone or tablet that has GPS.
- If you are close enough to shore, you might be able to get ATC to triangulate on your radio signal.

Quoting phunc (Thread starter):
4. In the event of an INS failure while oceanic, GPS is the only method of nav. What would generally happen in this case?

Depends how it is wired. If the GPS is an input to the FMS, you're good. If not then I guess the FMS has no valid inputs. Break out your stopwatch.



"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlineFabo From Slovakia, joined Aug 2005, 1219 posts, RR: 1
Reply 2, posted (1 year 6 months 1 week 5 days 17 hours ago) and read 3418 times:

I am assuming you are talking about a modern airliner.

An airliner has several sources of position. Amongst them typically 2-3 GPS recievers (except older frames that never were updated) 2-3 Inertial Reference Units, a DME/DME / VOR/DME / VOR/VOR input mixer.

There is then a computer (FMC, FMGS, whatever) that takes those signals, and calculates the most probable real position (using weighed averages and whatnot math). This computer can keep track of things like differences between sensors, how is IRS position diverging and much more.


Consider this: There is an airplane witout a GPS, and with three IRUs and a DME/DME updater.
This airplane will keep track of its position, and a position of each of three IRUs. When it goes out of DME range, it will not only average out the IRU position, it will work in the last known IRU deviation for each IRU and mix in deviation from IRU average from DME/DME updated average and trends.

As you can see, these systems were quite precise even before GPS.

Quoting phunc (Thread starter):
How does the A/C know it is maintaining RNP0.2NM? Is it based on INS inputs? Or a mix of inputs from the INS / DME/DME or VOR data?

So to sum up, airplane will do this with taking several different inputs into accord, and can also keep track of inherent precision of each method and of precision error of the calculation itself, and will show the error (and compare with required precision). Naturally, the most precise method is GNSS mixing with GBAS. Then you have GNSS without GBAS, then radio scans etc.

Quoting phunc (Thread starter):
2. Following on from the question above, if the aeroplane is oceanic with no radio nav in the area, the only backup method of navigation is INS. Is that correct?

Not exactly INS, that is something different, but if the airplane does not have GNSS receivers, then yes, it is only relying on its IRS. There used to be different methods of global positioning (LORAN, Omega), but these are not used now, and hardly anyone would use them anyway, since precision is really low for todays GNSS standarts.

Quoting phunc (Thread starter):
3. Following on again, when the aeroplane is back in VOR / DME/DME range will it begin to recalculate and backup it's position based on these now available navaids?

Yes, automatically.

Quoting phunc (Thread starter):
4. In the event of an INS failure while oceanic, GPS is the only method of nav. What would generally happen in this case?

You could also try Glonass... :P No, really. A triple IRU failure is quite unlikely. The aircraft would still however attempt to calculate its position using last known variables, such as wind and so, mixed with variables it can still get (airspeed, altitude, TAT, thus also TAS etc). It will not be very precise, but it will in all probability drive you into radio recievable range.



The light at the end of tunnel turn out to be a lighted sing saying NO EXIT
User currently offlineMrBuzzcut From United States of America, joined Jan 2008, 63 posts, RR: 0
Reply 3, posted (1 year 6 months 1 week 5 days 17 hours ago) and read 3408 times:

Thanks for the replies to the question that wasn't even mine, but one I had wondered about. I never was quite sure how, especially on an oceanic track one knew exactly where they were. Just for an idiot like me that doesn't have much knowledge of avionics but loves aircraft, and knows a bit about flight routing, how would a trip from, say, HNL to AWK work from a navigation standpoint in the days before GPS, or on a non-GPS equipped aircraft.

Reason I wonder is I was always amazed when I worked at AWK that the old DC-8 that is still flying the route every other week hit that runway in the middle of the ocean, especially for my first year and a half there when the VORTAC was OTS due to Typhoon Ioke.

I know the route was KATHS A450 AWK, but what actually navigates the track, other than the heading published on the sectional, and how do you find something like a 3000M runway in the middle of the ocean if the weather is overcast and there is no VORTAC or other navaids on the field?


User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 4, posted (1 year 6 months 1 week 5 days 16 hours ago) and read 3404 times:

Quoting MrBuzzcut (Reply 3):
how would a trip from, say, HNL to AWK work from a navigation standpoint in the days before GPS, or on a non-GPS equipped aircraft.

Dead reckoning with compass navigation and celestial navigation are valid methods and the first is still taught today. Certainly they are nowhere near as accurate as LORAN, GPS, INS etc but they do work well enough to get you in the vicinity of where you need to go. The biggest unknown is winds aloft. Forecasts over the ocean may be less than accurate and that can throw you quite a bit off course.

Your DC-8 would have inertial navigation of some sort right? Unsure about this.

Radio navaids were crucially important. If the plane was in the general vicinity of AWK, it could tune an NDB (I assume there is/was one) and home in with ADF. And if that (or your VORTAC) was OTS commercial radio stations work too. Once you know where you are in relation to the radio navaid the rest is just geometry. And if you have DME or two navaids to triangulate on, Bob's your uncle.

Fixes would also be taken on other radio navaids enroute assuming there were any.

If the weather is overcast that throws an interesting wrinkle into things. However the instrument approach(es) would have published minimums. The one big wrinkle is the possibility of diversion. Lots of fuel to get to the nearest alternate. And of course you'd have the weather forecast. But if the overcast is anything over a few thousand feet you can duck under and look. And that's what the old timers did a lot if you go back to the pre-jet days.

I highly recommend the book "Fate is the Hunter". The descriptions of navigation "issues" before and during the war are hair raising. The risks taken would have sent today's FAA examiners scurrying for cover.

[Edited 2013-01-18 19:09:43]

[Edited 2013-01-18 19:10:32]

[Edited 2013-01-18 19:11:08]


"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlineMrBuzzcut From United States of America, joined Jan 2008, 63 posts, RR: 0
Reply 5, posted (1 year 6 months 1 week 5 days 16 hours ago) and read 3386 times:

Quoting Starlionblue (Reply 4):
And of course you'd have the weather forecast.

That was always fun at AWK. The forecast was pretty general due to a lack of radar coverage over the ocean. Overcast would usually happen with a pop up storm that developed an hour or two before something was due to arrive.

Quoting Starlionblue (Reply 4):
But if the overcast is anything over a few thousand feet you can duck under and look. And that's what the old timers did a lot if you go back to the pre-jet days.

I once saw a C-17 that was GPS equipped, but the crew not certified to fly the GPS/RNAV approach take that option. The controller had to talk them into it by letting them know that the ceiling was 2200', and there were no obstacles above 50' between us and GUM to the west (1500 NM) and HNL to the east (2400 NM). They went ahead and broke through the cloud layer and there they were, about 10 miles out and right on course for the centerline of RWY 10.

I'm sure the DC-8 had INS, probably GPS as well, I was just curious as to how they did it in the old days. Well than and when the NAVAIDS are OTS and there is no local radio station that broadcasts for more than 5 miles.


User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 6, posted (1 year 6 months 1 week 5 days 15 hours ago) and read 3375 times:

Quoting MrBuzzcut (Reply 5):
I was just curious as to how they did it in the old days.

In very scary ways sometimes.  As I said read "Fate is the Hunter".



"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlinezeke From Hong Kong, joined Dec 2006, 8872 posts, RR: 75
Reply 7, posted (1 year 6 months 1 week 5 days 15 hours ago) and read 3368 times:

Quoting phunc (Thread starter):

I do not think you should be inserting the RNP value into the FM, it should be coded into the approach. During the approach, if it coded correctly, it should display the required and actual RNP value, and produce a nav accuracy error if th required RNP is not being achieved.



We are addicted to our thoughts. We cannot change anything if we cannot change our thinking – Santosh Kalwar
User currently offlinemandala499 From Indonesia, joined Aug 2001, 6768 posts, RR: 76
Reply 8, posted (1 year 6 months 1 week 5 days 12 hours ago) and read 3333 times:

Quoting MrBuzzcut (Reply 3):
how do you find something like a 3000M runway in the middle of the ocean if the weather is overcast and there is no VORTAC or other navaids on the field?

Back in the old days, celestial navigation was one method. DC-8s have the pesicopic sextant port on top of the flight deck. A good navigator can be very very accurate...
Back in the old old old days, use "Chichester's Theory of Deliberate Error", Pan Am used this as a standby procedure for finding... guess what? Wake Island.

Quoting MrBuzzcut (Reply 5):
Well than and when the NAVAIDS are OTS and there is no local radio station that broadcasts for more than 5 miles.

Anyone got a HF transmitter there? (Non-Navaid grade, all you need is an ADF that can get the frequency).

Mandala499



When losing situational awareness, pray Cumulus Granitus isn't nearby !
User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 9, posted (1 year 6 months 1 week 5 days 11 hours ago) and read 3318 times:

Quoting mandala499 (Reply 8):
Back in the old days, celestial navigation was one method. DC-8s have the pesicopic sextant port on top of the flight deck. A good navigator can be very very accurate...

And all the while looking awesome.
http://www.vc10.net/Technical/Images/vc10_periscope.jpg



"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlinephunc From United Kingdom, joined Mar 2011, 140 posts, RR: 0
Reply 10, posted (1 year 6 months 1 week 5 days 6 hours ago) and read 3279 times:

Thanks for the replies, guys.

Quoting Fabo (Reply 2):
I am assuming you are talking about a modern airliner.

Yes indeed.

Quoting Fabo (Reply 2):
Not exactly INS, that is something different, but if the airplane does not have GNSS receivers, then yes, it is only relying on its IRS.

Could you expand on this? I assume that there might be a subtle difference between INS and IRS.

Quoting Fabo (Reply 2):
You could also try Glonass... :P No, really. A triple IRU failure is quite unlikely. The aircraft would still however attempt to calculate its position using last known variables, such as wind and so, mixed with variables it can still get (airspeed, altitude, TAT, thus also TAS etc). It will not be very precise, but it will in all probability drive you into radio recievable range.

A triple failure of IRU plus GPS/GNSS (which ever you're using) would be a bad day! I've read reports of aircraft suffering 'degraded nav accuracy' while enroute or losing nav accuracy. I've always wondered exactly what it going on here.

Quoting zeke (Reply 7):
I do not think you should be inserting the RNP value into the FM, it should be coded into the approach. During the approach, if it coded correctly, it should display the required and actual RNP value, and produce a nav accuracy error if th required RNP is not being achieved.

I didn't think RNAV/RNP value was part of ARINC 424 NavData coding and thus if crew were in MNPS airspace, for example, then they would insert RNP10. Again, flying an RNAV approach would require 0.3NM and an RNAV RNP AR would require 0.1NM. I'm just baffled to understand how there aeroplane can monitor its accuracy. If I had a GPS in my left hand and an IRU in my right hand, and walked in a straight line...which do I believe is correct? I would surely need a third type of device to back up either of the devices in my left or right hand.


User currently offlineFabo From Slovakia, joined Aug 2005, 1219 posts, RR: 1
Reply 11, posted (1 year 6 months 1 week 5 days 2 hours ago) and read 3240 times:

Quoting phunc (Reply 10):
Could you expand on this? I assume that there might be a subtle difference between INS and IRS.

The INS stands for Inertial Navigation System. Perhaps the best known is Delco Carousel IV-A. A variant of this was used on Concorde for example (supersonic variant, forgot the designation, sorry).
This unit would typically have mechanical gyroscopes, and in terms of user interface, would look kinda like a calculator.
You would only use "GPS" coordinates with this units, not waypoint names, although it can be made to read a flightplan snippet from a punchcard. You would have 9 waypoints to put in, and a waypoint "0" that is the present position. There was triple mixing if you had three units, for better precision, and you could do manual DME update.

IRS stands for Inertial Reference System. This is what is found in modern airliners.
The system would be based on laser gyros, as opposed to mechanical gyros, and does not have any "navigation" capability. Mostly it just sends its outputs to a different computer(s) for further processing.

Quoting phunc (Reply 10):
plus GPS/GNSS (which ever you're using)

GPS is a GNSS. There is also GLONASS, Galileo and the chinese Baido (I think it is called).

Quoting phunc (Reply 10):
suffering 'degraded nav accuracy' while enroute or losing nav accuracy.

Degraded accuracy usually comes to a failure of one or more inputs. For example, if a radio failed in an airliner without GPS, it could not use DME/DME for position update anymore, and would have to resort to VOR/DME update or just IRU mixing. Losing nav capability would usually not be due to a failure of IRS or whatnot, but due to a failure of the computer that actually makes these inputs usable - the FMC. This happens sometimes, you would still be left with position indicators on IRS panel (at least for Boeing airplanes) and radio navigation.

Quoting phunc (Reply 10):
I didn't think RNAV/RNP value was part of ARINC 424 NavData coding and thus if crew were in MNPS airspace, for example, then they would insert RNP10. Again, flying an RNAV approach would require 0.3NM and an RNAV RNP AR would require 0.1NM. I'm just baffled to understand how there aeroplane can monitor its accuracy. If I had a GPS in my left hand and an IRU in my right hand, and walked in a straight line...which do I believe is correct? I would surely need a third type of device to back up either of the devices in my left or right hand.

Either RNP value is a part of 424 coding or there is a reference table to invoke RNP based on procedure type (enroute, star, transition, ILS approach, RNP approach, whatever.
In any case, the FMC knows the RNP.

If you had a GPS in your left hand, and an IRU in your right hand, you would know the precision and rate of degradation of each of them, and you would also use the knowledge of your previous position and you would know how to tell something is amiss (i.e. there is a tolerance of position difference, growing with time, but if suddenly the GPS has you 50 miles from where you were 5 minutes ago, something is not right)



The light at the end of tunnel turn out to be a lighted sing saying NO EXIT
User currently offlinemusang From United Kingdom, joined Apr 2001, 861 posts, RR: 7
Reply 12, posted (1 year 6 months 1 week 4 days 8 hours ago) and read 3138 times:

As an aside, with reference to the RNP (required Navigational Performance) and the ANP (Actual...), typically displayed side by side on a Control/Display Unit on the Legs page, I'm sure I was once told by someone higher up the food chain that even though the numbers are suffixed "NM", its not really giving you a reading in nautical miles, but rather an indication of relative accuracy in proportionate terms.

I'm not in a position to seek the proof, but I recall our manual doesn't explain this, as Boeing Manuals are mercifully written on a "need to know" basis!

Anyone have any more on this notion?

Regards - musang


User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 13, posted (1 year 6 months 1 week 4 days 8 hours ago) and read 3137 times:

Quoting musang (Reply 12):

As an aside, with reference to the RNP (required Navigational Performance) and the ANP (Actual...), typically displayed side by side on a Control/Display Unit on the Legs page, I'm sure I was once told by someone higher up the food chain that even though the numbers are suffixed "NM", its not really giving you a reading in nautical miles, but rather an indication of relative accuracy in proportionate terms.

If it was in proportionate terms it would just say so right? Certainly on the CDI it is in relative terms (number of ticks the D-bar is off representing a proportion of the specified RNP). But if it says XTK 0.5 NM I don't see how the figure could mean anything else.

[Edited 2013-01-20 02:57:45]


"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlinemusang From United Kingdom, joined Apr 2001, 861 posts, RR: 7
Reply 14, posted (1 year 6 months 1 week 4 days 5 hours ago) and read 3106 times:

Thats what I would have thought. I'll research it next time I'm in the office.

Regards - musang


User currently offlineFabo From Slovakia, joined Aug 2005, 1219 posts, RR: 1
Reply 15, posted (1 year 6 months 1 week 4 days 4 hours ago) and read 3086 times:

Quoting Starlionblue (Reply 13):
But if it says XTK 0.5 NM I don't see how the figure could mean anything else.

That would be something different. XTK ERROR says how far your computed position is from the desired position, not how precise the computed position is.



The light at the end of tunnel turn out to be a lighted sing saying NO EXIT
User currently offlinezeke From Hong Kong, joined Dec 2006, 8872 posts, RR: 75
Reply 16, posted (1 year 6 months 1 week 4 days 3 hours ago) and read 3079 times:

Quoting phunc (Reply 10):
I assume that there might be a subtle difference between INS and IRS.

Yes, one provides a navigation solution, the other does not. The INS knows the position, and can output position information to the aircraft, it only provided position information (some provided autopilot steering commands). The IRS these days is part of combined system (ADIRU) that provides attitude, airspeed, and position information to the aircraft, so it provides the attitude, position, and airspeed. The position that is supplied to the FM from the ADIRU is a GPS/IRS mix position.

Quoting phunc (Reply 10):

I didn't think RNAV/RNP value was part of ARINC 424 NavData coding and thus if crew were in MNPS airspace, for example, then they would insert RNP10. Again, flying an RNAV approach would require 0.3NM and an RNAV RNP AR would require 0.1NM. I'm just baffled to understand how there aeroplane can monitor its accuracy. If I had a GPS in my left hand and an IRU in my right hand, and walked in a straight line...which do I believe is correct? I would surely need a third type of device to back up either of the devices in my left or right hand.

The RNP is part if ARINC 424. The ADIRU provides a mix of IR/GPS position to the aircraft, they monitor the mix position.

Quoting musang (Reply 12):

Anyone have any more on this notion?

The Actual Accuracy defined as Total System Error (TSE), which is combination of Path Definition Error (PDE) – Accuracy of Fix definition (usually zero), Navigation System Error (NSE) – Accuracy of Navigation System and aids used - GNSS/DME etc, Flight Technical Error (FTE) – Ability of a/c systems to follow accurate path. Allowable error 0.5 x RNP – usually ≤0.01nm with use of FDs and AP.



We are addicted to our thoughts. We cannot change anything if we cannot change our thinking – Santosh Kalwar
User currently offlinePihero From France, joined Jan 2005, 4392 posts, RR: 76
Reply 17, posted (1 year 6 months 1 week 4 days 3 hours ago) and read 3073 times:
Support Airliners.net - become a First Class Member!

Quoting phunc (Thread starter):
1. Generally speaking, the crew will input an RNP value into the FMS. E.g. RNP5. This will be displayed on the FMS MCDU along side an "actual" readout. Based on this, the A/C is trying to maintain RNP5 but the actual maybe 0.2NM. How does the A/C know it is maintaining RNP0.2NM? Is it based on INS inputs? Or a mix of inputs from the INS / DME/DME or VOR data?

We do not in normal OPS input a manual RNP value. The system does it for you... You'll just have to check the accuracy prior to an approach. Otherwie, there are warning messages in the cases of degraded accuracy of your system.

Quoting phunc (Thread starter):
2. Following on from the question above, if the aeroplane is oceanic with no radio nav in the area, the only backup method of navigation is INS. Is that correct?

Your question is rather strange as you refer to your title as " FMS Nav accuracy". We should therefore concentrate on this :
- On a 'Bus - and I'm quite sure on all modern aircraft - the FMGS uses these modes in the order of decreasing accuiracy :
- Mixed GPS / IRS
- Radio Position : DME/DME ; VOR / DME ; LOC ; DME/DME/LOC....
- Mixed IRS
Moreover, GPS and/or radionavs are used to update the "raw" mixed IRS position.
This means that IF you lose GPS data, the system goes back to the - updated / amended...- mixed IRS position we call FM position.

Quoting phunc (Thread starter):
3. Following on again, when the aeroplane is back in VOR / DME/DME range will it begin to recalculate and backup it's position based on these now available navaids?

Yes

Quoting phunc (Thread starter):
4. In the event of an INS failure while oceanic, GPS is the only method of nav. What would generally happen in this case?

Losing all three IRSs is so remote as to be close to utter impossibility, but you'll have quite a few problems to cope with, in terms of AP / ATHR / Flight controls and use of stby instruments... You'll have to manually set heading, using the MCDUs raw data.

Quoting MrBuzzcut (Reply 5):
I'm sure the DC-8 had INS, probably GPS as well, I was just curious as to how they did it in the old days.

That DC-8 was triple-INS equipped and the accuracy was quite good... In the last 2 to 300 Nm, a radar in MAP mode would refine the last stage of the flight... especially with the "old" sets.

Quoting phunc (Reply 10):
. If I had a GPS in my left hand and an IRU in my right hand, and walked in a straight line...which do I believe is correct?

that's a strange statement : The GPS is only dependent on satellite availability. It doesn't drift, contrarily to an INS set.



Contrail designer
User currently offlineStarlionblue From Greenland, joined Feb 2004, 17001 posts, RR: 67
Reply 18, posted (1 year 6 months 1 week 3 days 20 hours ago) and read 3008 times:

Quoting Fabo (Reply 15):
Quoting Starlionblue (Reply 13):
But if it says XTK 0.5 NM I don't see how the figure could mean anything else.

That would be something different. XTK ERROR says how far your computed position is from the desired position, not how precise the computed position is.

You are correct of course. But wouldn't this still be expressed in NM and not as a ratio/proportion?



"There are no stupid questions, but there are a lot of inquisitive idiots."
User currently offlineairmagnac From Germany, joined Apr 2012, 303 posts, RR: 44
Reply 19, posted (1 year 6 months 1 week 3 days 19 hours ago) and read 2999 times:

Quoting phunc (Reply 10):
I'm just baffled to understand how there aeroplane can monitor its accuracy

The "monitoring of the accuracy" is also called integrity, and it is quite a baffling concept
As a simple definition, integrity can be seen as the capacity of a system to detect an error and take appropriate action (warn the user and/or automatically correct the error), all this happening sufficently rapidly to avoid bad problems. For example, you can allow a bit more time for a nav system to detect an error in cruise than during a CATIII landing.

Quoting phunc (Reply 10):
If I had a GPS in my left hand and an IRU in my right hand, and walked in a straight line...which do I believe is correct? I would surely need a third type of device to back up either of the devices in my left or right hand.

Actually no, you would not, and that's the beauty of hybrid INS/GPS.
The error sources are completly different : inertial systems drift over time ; GPS has lots of sources for small errors, but the error does not evolve as a function of time.
GPS data on its own also has little integrity - if it's bad, you won't be warned. If the satellite constellation positioning information is bad, GPS will happily triangulate your position with wrong references, and provide you with a wrong fix without warning.

But if you use both together, they complement each other. You can use GPS to reset your INS over time, and if the GPS data was to go bad, it would show as a difference from the INS position. So the hybrid GPS+INS is not only accurate, but also has rather high integrity, much more than taking each system seperatly.

Actually GPS was developed for exactly this purpose of complementing inertial systems (of ICBMs), which is why the raw GPS has no integrity and has to be complemented with RAIM and/or augmented with ground- or satellite-based augmentation systems (GBAS or SBAS) to recover some integrity.
GALILEO should have built-in integrity features



One "oh shit" can erase a thousand "attaboys".
User currently offlinetdscanuck From Canada, joined Jan 2006, 12709 posts, RR: 80
Reply 20, posted (1 year 6 months 1 week 3 days 14 hours ago) and read 2965 times:

Quoting musang (Reply 12):
As an aside, with reference to the RNP (required Navigational Performance) and the ANP (Actual...), typically displayed side by side on a Control/Display Unit on the Legs page, I'm sure I was once told by someone higher up the food chain that even though the numbers are suffixed "NM", its not really giving you a reading in nautical miles, but rather an indication of relative accuracy in proportionate terms.

The numbers are in NM (unless you have a very funny FMC). The displays (if equipped) are proportional.

Tom.


User currently offlinemusang From United Kingdom, joined Apr 2001, 861 posts, RR: 7
Reply 21, posted (1 year 6 months 1 week 3 days 5 hours ago) and read 2904 times:

Quoting musang (Reply 12):
even though the numbers are suffixed "NM", its not really giving you a reading in nautical miles, but rather an indication of relative accuracy in proportionate terms.

Here's what gave me the notion. Its from a Flight Crew Operational Memo from a few years ago.

[Slightly paraphrased for brevity, but meaning is preserved]

RNP ops is sensor-independent. Operational criteria specifies the need for crew to be able to monitor RNP capability during flight. Because mixed sensors can be used, this monitoring would be cumbersome, so the industry solution was to provide the pilot with an indication of FMC position accuracy (it's actually a measure of position uncertainty based on nav sensor error models in the FMC). This measure of FMC position uncertainty is called ANP on Boeing (Estimated Position Error on Airbus).

If estimated error approaches the RNP value, warnings are generated - "UNABLE RNP" (Boeing) or "NAV ACCUR DOWNGRADE" (Airbus).

ANP (/EPE) consists of a calculated Nav System Error and an assumed model of the Flight Technical Error and is not a direct measure of the actual FMC position error.

The last ten words are what had stuck in my mind.

Regards - musang


User currently offlinetdscanuck From Canada, joined Jan 2006, 12709 posts, RR: 80
Reply 22, posted (1 year 6 months 1 week 3 days 4 hours ago) and read 2887 times:

Quoting musang (Reply 21):
ANP (/EPE) consists of a calculated Nav System Error and an assumed model of the Flight Technical Error and is not a direct measure of the actual FMC position error.

The last ten words are what had stuck in my mind.

What they mean there is that if your ANP is 0.5, that doesn't mean you're 0.5NM off target, it means you're within 0.5NM of target (but you don't know where within that 1NM wide band). It's describing the error band, not where you are in the band.

Tom.


User currently offlineFabo From Slovakia, joined Aug 2005, 1219 posts, RR: 1
Reply 23, posted (1 year 6 months 1 week 3 days 4 hours ago) and read 2886 times:

Uh huh.

What I gather from this wording is, the ANP is indeed in NM, but this sentence reminds you that ANP of (say) 0.15 miles does not mean you are at exactly 0.15 miles from computed position, but that you are anyway from 0 to 0.15.
And/or that this is calculated, not measured - that does make sense, as if you could measure your position error, you could also work it out into correction and get better PPOS.

Quoting Starlionblue (Reply 18):
You are correct of course. But wouldn't this still be expressed in NM and not as a ratio/proportion?

Yes, but you are mixing two indications. XTK is the difference between PPOS and desired track, while ANP is accuracy index.

FWIW I also believe that ANP in nm is indeed in NM, but did not discard the ratio theory, as it had some merit.



The light at the end of tunnel turn out to be a lighted sing saying NO EXIT
User currently offlinePihero From France, joined Jan 2005, 4392 posts, RR: 76
Reply 24, posted (1 year 6 months 1 week 3 days 3 hours ago) and read 2876 times:
Support Airliners.net - become a First Class Member!

Quoting tdscanuck (Reply 22):
What they mean there is that if your ANP is 0.5, that doesn't mean you're 0.5NM off target, it means you're within 0.5NM of target (but you don't know where within that 1NM wide band).

Shouldn't it be : "... you don't know where within that .5Nm-radius circle or 1 Nm-diameter circle " ?



Contrail designer
User currently offlinePihero From France, joined Jan 2005, 4392 posts, RR: 76
Reply 25, posted (1 year 6 months 1 week 3 days 3 hours ago) and read 2892 times:
Support Airliners.net - become a First Class Member!

Quoting airmagnac (Reply 19):
GPS data on its own also has little integrity - if it's bad, you won't be warned. If the satellite constellation positioning information is bad, GPS will happily triangulate your position with wrong references, and provide you with a wrong fix without warning.

Not quite : The GPS system - I believe it's a feature of the receiver - constantly computes the inherent iaccuracies of the environment : Satelite position, wave propagation, time-base errors, etc... and generates a " Figure of merit " one could assimilate as the system precision. The max allowed FOM is .28 Nm. It is then used to generate the GPS/IRS - alias GPIRS - EPE or estimated error position of the mixed IRS/GPS.
For info, an FOM above .28 Nm will cause the rejection of the GPS position.



Contrail designer
User currently offlineairmagnac From Germany, joined Apr 2012, 303 posts, RR: 44
Reply 26, posted (1 year 6 months 1 week 3 days ago) and read 2861 times:

Quoting Pihero (Reply 25):
" Figure of merit "


The FOM is a gross indication, reflecting an estimation of cumulative receiver error. But you have no way (with a basic GPS receiver) of knowing if there is an error in the satellite segment or ground-control segment. Granted, those are strictly controlled and fail rather rarely, but it's not impossible. So as you say, it is useful for gross monitoring of the GPS position, with a necessary condition (if receiver error is bigger than 0.28nm, then don't use the GPS - sounds reasonable !). But it is not sufficient to validate the position data.

Which is why I wrote it as "GPS data on its own also has little integrity" and not "no integrity"  

Ref here : http://www.navcen.uscg.gov/pubs/gps/gpsuser/gpsuser.pdf
§6.3 on page 93 of the pdf and chapter 12, especially around page 158

Also I doubt many systems use basic GPS on its own these days, especially on an airliner. The GPS will usually be wrapped inside a more complete nav system, with RAIM and/or associated to augmentation services or assistance services (through mobile networks) which can provide information about the entire system status. Which is why I was referring to "basic" GPS.

So while my answer applies to the simple scenario proposed by the OP, to help explain the concept of hybrid INS/GPS, it probably can't be used as-is for airliner ops.



One "oh shit" can erase a thousand "attaboys".
User currently offlinetdscanuck From Canada, joined Jan 2006, 12709 posts, RR: 80
Reply 27, posted (1 year 6 months 1 week 2 days 20 hours ago) and read 2835 times:

Quoting Pihero (Reply 24):
Shouldn't it be : "... you don't know where within that .5Nm-radius circle or 1 Nm-diameter circle " ?

Nope. As far as I know, RNP is just lateral so it's not a circle. The system doesn't care about fore/aft accuracy along the track.

Physically, it's probably an ellipse with the wide axis lateral, but the RNP procedures are just about distance off track.

Tom.


User currently offlinePihero From France, joined Jan 2005, 4392 posts, RR: 76
Reply 28, posted (1 year 6 months 1 week 2 days 10 hours ago) and read 2786 times:
Support Airliners.net - become a First Class Member!

Quoting tdscanuck (Reply 27):
Nope. As far as I know, RNP is just lateral so it's not a circle. The system doesn't care about fore/aft accuracy along the track.

I beg to differ, Tom.
If so, I would never use RNav procedures into a mountanous airfield. I'd revert to NavAidfs let-downs as my system could not give / insure an exact position to start my descent.
Problem is that most manuals illustrate RNP with a X-track diagram, containment limits and all that jazz, which is fine to understand the basic concept in the lateral dimension, but is quite wrong in terms of global concept understanding... On top of that, comes the all important necessity of lateral separation and then we complete the circle for concept misunderstanding.

This is an Airbus text, highlights are my own :
"RNP Type
This is a designator, established according to navigational performance accuracy in the
horizontal plane, namely lateral and longitudinal position fixing. This designator is
indicated by an accuracy value given in nautical miles.
RNP-X
A designator is used to indicate the minimum navigation system requirements needed to
operate in an area, on a route, or on a procedure (e.g., RNP-1, RNP-4). The designator
invokes all of the navigation system requirements, specified for the considered RNP
RNAV type, and is indicated by the value of X (in NM).



Contrail designer
User currently offlinemusang From United Kingdom, joined Apr 2001, 861 posts, RR: 7
Reply 29, posted (1 year 6 months 1 week 2 days 10 hours ago) and read 2786 times:

Quoting tdscanuck (Reply 27):
As far as I know, RNP is just lateral so it's not a circle. The system doesn't care about fore/aft accuracy along the track.

Agreed Tom, but its interesting that most of the diagrams in our literature show aircraft in the centre of concentric circles!

Regards - musang


Top Of Page
Forum Index

Reply To This Topic FMS Nav Accuracy
Username:
No username? Sign up now!
Password: 


Forgot Password? Be reminded.
Remember me on this computer (uses cookies)
  • Tech/Ops related posts only!
  • Not Tech/Ops related? Use the other forums
  • No adverts of any kind. This includes web pages.
  • No hostile language or criticizing of others.
  • Do not post copyright protected material.
  • Use relevant and describing topics.
  • Check if your post already been discussed.
  • Check your spelling!
  • DETAILED RULES
Add Images Add SmiliesPosting Help

Please check your spelling (press "Check Spelling" above)


Similar topics:More similar topics...
737-300 FMS posted Tue Sep 25 2012 16:43:18 by Novice
Delco Carousel: The First Real FMS? posted Mon Jun 18 2012 06:01:16 by faro
Question FMS Fokker F100 posted Wed Nov 2 2011 01:09:59 by Oli
CRJ FMS Database Problem? posted Thu Oct 20 2011 03:21:26 by bluewhale18210
MD-11 FMS SID Question posted Mon Oct 4 2010 16:49:05 by bob34
Icao Fuel Burn Statistics: Accuracy? posted Sat Aug 14 2010 06:48:58 by warreng24
Aircraft Dimensions Accuracy? posted Sun Jun 27 2010 15:01:47 by ariis
Future Air Nav System ( Fans ) : When And How? posted Thu May 20 2010 15:12:08 by Gonzalo
Reading 'Air Nav' Radar Screen posted Wed Mar 17 2010 06:41:48 by Aeolus
ILS With Same NAV Freq.? Almost Made Me Crash posted Wed Jan 6 2010 09:52:39 by Wardialer

Sponsor Message:
Printer friendly format