IRS Alignment
 

Why the IRS cant be realigned in flight? I know the aircraft need not to be in motion during alignment. But if the IRS loses it’s position, why it stays like that for the reminder of light. Why cant we just provide it with actual position from the GPS so it can realign?

IRS is detecting earth's rotation during alignment, and it would be difficult to do so when doing 500 kts in turbulent air.

There are some military-grade IRUs that are able in-flight alignment though.

Older military aircraft often had INS/IRSs that could be aligned inflight. An example is the USAF C-130Hs with self-contained navigation system (SCNS). Inflight INS alignment requires valid GPS or Doppler velocity sensor data. The alignment process requires that the airplane fly straight and level for a minimum of 1 minute after commanding the alignment, although 5 minutes is recommended. After 5 minutes in alignment, S turns are recommended. The KC-135 fleet was equipped with Carousel IV INSes that also be aligned inflight using similar procedures.

787 IRS may be aligned in flight as well...

this might be dumb question but at ground why the IRS would need to sense the earth rotation when it already have position from the GPS? Or is it just need it for a comparison?

The IRS determines the direction of True North by measuring the direction of apparent wander of its “giros” in 3 directions, due to Earth’s rotation. The rate of apparent wander enables it to determine the Latitude. When we insert present position, it compares the value to its calculated Latitude, but it needs Longitude because that can’t be determined.

That makes sense! So the IRS just compares both latitudes, and if both matches, it will then take/add GPS longitude to it. That explains it, thank you!

Actually, that is incorrect. The GPS provides alignment to grid north, not mag north.

There are many systems, in fact most commercial ac systems, which update realtime the IRU through the GPS, most not more than a 5 minute update.(due to latency)
The kalman filters compensate between the updates. If there is not an update within 15 minutes, most newer IRS system will compensate and keep the headings.... depending on the system...but after their respective time frames, the IRS drift comes into play.
Other factors such as the maneuvers during the update timeframes...if the ac keeps a steady heading and speed, it will compensate longer...make a turn or change altitude, and the accuracy diminishes rapidly...

This is a common misunderstanding. The IRS position is never updated.

The data from all the navigation sources (IRS, GPS, VOR, LOC, DME) goes into the FMC through Kallman filter, where FMC weighs reliability of each data (satellite coverage with GPS, geometry with VOR/DME, etc.) as well as compare outputs of data sources to each other (an IRS which shows completely different position would have lower weight) and calculate what it thinks is the position of the aircraft - so called FMC position, which is then displayed on the ND.

But the IRS itself never gets updated. On the B737, it is not unusual for IRS to be away a mile or two from GPS position after a long flight and Boeing recommends full alignment prior flight whenever possible.

what do you mean by IRS will compensate and keep the headings?

totally agree.. thats what I always thought!

Salute!

At all above! PLZ read up a bit about how inertial systems work and what is needed to "align" them. Several misconceptions and misunderstandings repeated here and incomplete knowledge of how the damned systems work.

First, a few clues to guide you....

- The "system" may be composed of several sensors, not just one magic box.
- The system may not be meant to provide extremely accurate body rates and such for display ( e.g. inertial vertical velocity versus barometric perssure changes) , but basic navigation coordinates with a specified accuracy and update timeframe. You know, where am I?
- an inertial system does not need to be mechanically postioned in "inertial" space to work like they did in the really old days. See "strapdown" systems and RLG's. The tiny one in an AMRAAM is ready to go within a second, and does real well.
- Since the 60's, at least one USAF attack plane could do airborne alignments for navigation and then bombing calculations. A rough initial position was good enuf, then takeoff after 2 minutes or less. The navigation computer then used Doppler info and crude body rates from the inertial platform, which was only good for AHRS at that stage. Kinda like 99% of everything flying. Using doppler and great software, the system could be "aligned" to a "reasonable" state. A radar or visual or flyover position update refined the whole deal and then the thing got better as it then knew lat long, which provided earth rotation rates and better calculation of the wander angle, local level, etc. etc.. I leave it to the reader to research that system and plane.

Gums sends...

I don’t see how an airplane can do a full alignement in the air like some guys are mentioning here. You can have basic mode like attitude and heading but how the IRS will sense the earth rotation during flight???

It gets the information from the GPS.

I if the aircraft maintains a constant heading in level flight for some minimum time and the IRS system is fed ground velocity (Doppler or GPS info) then it can recalculate, from the expected apparent wander, and determine the direction of True North again.

For some mind boggling feats of navigation, see https://web.mit.edu/digitalapollo/Do...progreport.pdf
Apparently one crew managed to input “on launch pad at Cape Canaveral” in mid flight. They realigned the IMU (Inertial measurement unit) using star shots.

As I understand it, Garmin and Aspen GA EFIS systems align anytime they need to. A/C state and position seems not to matter.

AHRS is totally different to INS/IRS.

Attitude Heading Reference Systems provide attitude and heading, strangely enough.

INS / IRS also provides position. Hence the required alignment time.

Actually, and counterintuitively, exactly the opposite is true. The below is from the Challenger 605 FCOM. Note 1 is interesting.

That’s odd. I was so intrigued by what you quoted, I found this reference:
http://www.smartcockpit.com/docs/CL6...ON_SYSTEMS.pdf

”In-Flight Alignment
If a power interruption or a transient system fault occurs in-flight, and a GPS signal is valid, the IRS system will have the capability for re-alignment. During in-flight re-alignment, the IRS ALIGNING message appears on the PFDs.
The FMS will prompt the crew with the SET IRS HDG amber message on the FMS CDU scratchpad (required for FMS navigation – IRS will align without this action). If the crew enters a heading, the IRS will work as a DG until it transitions to the ATT mode, at which point it starts calculating its own HDG.
Typically, the time required for a full alignment in navigation mode is between 10 to 20 minutes. At the beginning of the alignment process, it is recommended that the aircraft maintain a straight and level flight attitude (see Note 1).
NOTE
1. Aircraft maneuvers involving changes in heading increase the alignment time.
2. The alignment time increases as a function of latitude, with the minimum time occurring at the equator, and the maximum time occurring at the poles.

Er, so it can work out your position.
To work out where you are, it does a mathematical calculation, based on your speed over the ground, and heading.
Ignoring ALL other factors , in simples terms your speed over the ground is going to be different if flying east to west or west to east, if flying at a constant air speed.

Ttfn

Salute!

We may be closing in on how the doggone nav system works depending on your ride. Before waiting for questions and enjoying Easter, I put on my instructor hat and try to help. You can't "teach" anything to anybody. All you can do is help them to learn. Bear with me for 5 or 6 paragraphs and then class is dismissed.

Goldenrivett and Check Airman are homing on to the ability of modern systems to do airborne alignments of the inertial unit. Your overall nav system may only use the inertial unit for certain things, and have other subsytems provide inputs to the overall nav and guidance solution. First time I saw a change was in late 60's, and the nav system called the inertial unit "inertial measurement unit" verus inertial navigation system. In other words, our nav system used the inertial unit and other inputs to provide us a super positional and velocity output for display and other purposes. Wasn't a self-contained system as the Double Ugly had.

The biggie way back was digital computers the size of shoeboxes versus your refrigerator. And then the folks at Draper Lab and other places implemented Kalman filters or their equivalent to the nav system. Apollo and sub-launched missiles were the impetus, and I personally benefited along with many others in my profession. Our nav gear and computed weapon delivery capability skipped a generation of traditional systems development. So by 1971, I flew a neat jet with a nav system which combined various subsystems in a central digital computer to provide me position, velocity, wind data, steering to a selected point in space, etc. One unit was the inertial measurement unit.

PLZ take a look at the links I include later to help you "Fathom" what I try to explain. BTW, sir, I completely unnerstan what an AHRS system does and how it works or I wouldn't be on the platform now.

So the short answer concerning airborne alignments is your system uses one box to sense rotation and acceleration - six degrees of freedom. Another box to integrate the data to get a velocity vector( you know, v=a*t). Given a starting position, assumed attitude ( like st and level) and a clock, it's easy to calculate where you are, how fast you are going, etc. Even wind! Using heading and groundspeed, the "wind triangle" is a no brainer. Washington Center used to routinely ask us what the wind was at FL250 as we came home from a range up north of The Beach. You don't need a "platform" stabilized in space as we did in the 60's and 70's. Given an initial attitude, the accelerometers and rate sensors can measure body rates and new attitude referenced to the initial values. The nav computer can use the data with initial inertial position in space to compute and display all kindsa good stuff. Your iPhone can "navigate" real well between position updates from its GPS. I truly believe we could have gone to the moon using one, with the IBM PI box as a backup, heh heh. The Draper folks used predicted visible posiiton of certain stars at predicted points in space and a great, small "telescope" to correct the nav solution, and they didn't have GPS back then. See how Kalman filters and their cousins work, and the airborne alignment solution is obvious, especially with frequent positional data from GPS or some external area nav system.

Some decent learning material, especially about Kalman:
https://en.wikipedia.org/wiki/Kalman_filter

https://www.kalmanfilter.net/default.aspx

https://en.wikipedia.org/wiki/Inerti...igation_system

Gums sends...

My question here was not how an IRS do alignment. My question was that why not an IRS just take all the position from the GPS rather than taking time sensing and calculating.

Align is necessary to determine the direction of True North.

As pointed out, it's so the box knows where true north is, and at what latitude it is currently parked.

The whole point of an inertial system is it can work without a gps input, they have been around since long before GPS was invented.

The answer was, it can't figure out the earth's rotation from gps alone. Gps satellites are mainly geo stationary , they stay in same place relative to earth's surface, and therefore GPS signal cant be used to sense earth's rotation. What the GPS is good for, is an initial latitude and longitude position.

Ttfn

Salute!

Nope, Ivor.

Firstly, GPS doesn't give a rats about where you are on the Earth or at a fairly high altitude.

And secondly, The nav computer will use the lat/long from the GPS and then start looking at sensed accelerations and rates. [EDIT: Using a Kalman filter or other algorithm, the nav system "corrects" the inertial until its output is what the theoretical values are for your position in space. The really old systems used a technique called gyrocompassing, and those usually "pointed" north once aligned. Most also used accelerometers fixed on a gyro-stabilzed platform. Computer and sensor technology got much better, and nowadays strapdown systems literally bolted on the airframe are as good as or better than the best systems we had in the 70's, 80's and early 90's. ] You can speed things up a lot by providing rough attitude relative to the Earth, but even without that a good system will get your "platform" aligned. And BTW, since the late 70's, inertials didn't have to "point" to true north all t he time. See "wander angle" and such. Hell, they don't even have to be "stabilized" in space either.

Gums sends...

GPS satellites are not geostationary. But it doesn't matter, the receiver knows where they are and calculates its position accurately.

Goldenrivet,

My quote is from the current version of the FCOM. Interesting the smartcockpit version, I have no idea where they got it from as there is no version information on the page. Maybe it reflects advances in the software over time.

Hi BizJetJock, thanks for your reply. The reference I found was for “The Honeywell Laseref V micro inertial reference system (IRS)”. What system do you have? I’d like to do some more reading to learn what advances they have made.

G: the greatest advances are probably still ahead https://news.umich.edu/small-precise...g-without-gps/

https://2020.ieee-inertial.org/inert...rtual-platform

Thanks very much for links FlightDetent.
They have come a long way since my gyroscope on a model of the Eiffel Tower.

The current manual is also for the Laseref V. If there had been a change of spec during the aircraft production life, the FCOM would cover both variants, since (unlike airliners) the FCOM is not tailored and so must cover all the options.

Salute!

Very good, Detent. We are talking about orders of magnitude accuracy compared to what several of us flew in the 60's/70's. And they were "state of the art" that got us to the moon, etc.

As Golden commented, we have gone a long.long way from a gyro-stabilized "platform" with mechanical accelerometers "glued" on it, along with small gyros to detect rotation. And we have done it faster than anyone would have imagined.

Gums sends...

The key thing is establishing the axes. Giving it position won't help with that. Need to maintain a reference such that g doesn't leak into the horizontal components. It does a bit but that's were Schuler tuning came in. All much better now with kalman filtering to estimate the error sources and gps truth. Mind you, it's been 30 odd years.......

Salute!

No doubt that starting the process with a good idea of "which way is up" helps to reduce "alignment" time, Mr Optimistic. And you are spot on about 30 years ago as you said.

OTOH, if I know where I am starting the "alignment" process, I can look up what my velocity in inertial space is supposed to be real easy. In milliseconds. So I tell my "platform" what all the velocities and rates should be for its frame of reference and go from there. Remember, the new systems are not using clumsy space-stabilzed" gyro platforms and are bolted to your plane's frame. Back in those golden days 40 years ago, we typed in our coordinates and the system compared sensed values with predicted values for that geo-referenced position in space. Airborne we did the same thing and tried to stay st-and-level for a bit. Airborne alignment wasn't as good as in the late 90's using GPS to provide actual position for the Kalman doofers to do their trick, but way better than 1970.

For those wondering where my education about these things advanced beyond what I taught in the 80's, it was during the JDAM operational concept demonstrations about 10 years later. I got to see all the goodies that various companies had for sale and had their engineers explain stuff so we could provide USAF with a good basis for further development or actual production. The JDAM was "aligned" in about one or two seconds from the carrier platform. The biggie then and now is present position, and rough body rates can be transmitted in a few milliseconds. The thing will get its own position from the satellites on its own very quickly after you drop it. If it knows where it was, then the calculations are able to get it squared away in seconds versus a half a minute. Over 20 years ago the B-2 could drop 16 of the suckers from it bays in a very short time, and the things could not get a GPS signal until a few seconds after release because they were in giant Faraday cage. Ditto for the Bone's bays and now the F-35.

Gums sends...

Also the 737MAX

As an aside, and this may come as a shock to the GPS fans, there are actually certain airports around the world these days where you will see evidence of GPS jamming on the ramp......In which case you revert (carefully) to the more traditional methods of telling the Inertial it's starting position - the sort of methods that I'm sure gums and others of a certain vintage are familiar with.:E

It's nice to have GPS available as a source for the present position at the start of the align process, if for no other reason than to reduce the risk of finger trouble during data entry leading to embarrassment, but there are reasons why it's not essential.

Sorry Gums, I couldn't resist...

The missile knows where it is at all times.....

It knows this because it know where it isn't. By subtracting where it is from where it isn't, or where it isn't from where it is, whichever is greater, it obtains a difference, or deviation. The guidance subsystem uses deviations to generate corrective commands to drive the missile from a position where it is to a position where it isn't and, arriving at a position where it wasn't, it now is. Consequently, the position where it is, is now the position that it wasn't, and it follows that the position that it was is now the position that it isn't. In the event that the position that it is in is not the position that it wasn't, the system has acquired a variation; the variation being the difference between where the missile is and where it isn't. If variation is considered to be a significant factor, it, too, may be corrected by the GEA. However, the missile must also know where it was. The missile guidance computer scenario works as follows: because a variation has modified some of the information the missile has obtained, it is not sure just where it is, however it is sure where it isn't, within reason, and it knows where it was. It now subtracts where it should be from where it wasn't, or vice versa. And by differentiating this from the algebraic sum of where it shouldn't be and where it was, it is able to obtain the deviation and its variation, which is called error.

Or in other words, Black Voodoo Magic...

Taipei (RCTP) was one of these airports with a well known GPS distortion presumably to confuse possible mainland chinese missiles.

As a mere avionics engineer I'm baffled. If the Inertial reference System loses its position in flight it is seriously unserviceable, so why would you want to realign it?

So what makes you think that a "strap down" system doesn't need to know its starting orientation in inertial space? What do you think it does with the coordinates that you type in? :confused:

Salute!

Good question, Blacksheep, and I thot I had answered some of it without going into minutiae.

Your system uses present position for the Kalman or other algorithm to compare predicted body rates and accelerations with what the inertial sensors are actually detecting, whether they are gyro-stabilzed like the old days or bolted on your airframe. The old ones assumed you were stationary on the ramp, huh? If you had an alternate means of providing movement over the earth like a doppler, things got easier. If you had postion data like from a GPS, even without velocity, things got easier. So for our aviation use navigating across the earth, initial position for alignment is a biggie. For systems lile the AMRAAM and other weapons, not so much. JDAM was provided persent position, aircraft velocity thru space and rough attitude, as they could be loaded on various store stations. That allowed it to get its own posiiton quickly and begin the hunt.
- Good first question, and I mentioned that providing attitude will speed up the alignment process, especially for Earth coordinate referenced systems concerned with true north. But for some of the strapdown systems we don't have to be at a specified attitude like st-and-lvl or wait for the "platform" to "settle" like we used to. Even with a rough attitude to start with, the doggone things nowadays eventually figure it out. When I say "eventually", that could range from seconds to several minutes depending on your set. I'll have to get with the sfwe troops, but my hunch is that orientation of the fixed sensors WRT the airframe is the primary requirement for your navigation not orientation WRT the Earth-referenced coordinate system. In other words, the overall nav system applies a bias to the sensed data, Gotta hit the books.

Gums sends..