IRS don't have to be calibrated but aligned...
they are gyros systems, and as such before being the accurate navigation system they are the gyros have to speed up to a certain speed ,where they become "rigid". which mean that there spining axis is always in the same position in respect of stars, ie whatever movement did the plane do, and even the earth, the spin axis is always pointing to the same direction.
when the A/C move calculations take place and process the actual A/C position + many other parameters. it just needs to be fed with the actual A/C position while aligning.
for that reason A/C must not be moved during the alignement process.

INS are based on the same principle but use accelerometers fixed to the frame of the aeroplane. every single movement is caused by an acceleration in different axis. this acceleration is measured and integrated one time to get a speed, and a second time to get a distance. again they need to be aligned, and you need to tell the "thing" what's its actual position during alignment. then you have an initial position, an acceleration, a speed and a distance, which is far enough to compute the actual position.

add to this few Air Data Computer which will process, from static pressure and pitot pressure, many parameters like speed, altitude, vertical speed...

feed all the informations computed by the IRS/INS, ADC to the FMC and then you have an extremely accurate and powerfull tool to navigate.

to understand and use these systems need A LOT of reading and training, and one may not feel the need to use them when others may will to master them. both are respectable.

myself I prefer flying manually ...

If PMDG can do it, MS should be able to do it too... then it would probably be a great added value to FSX. for virtual airline, for example....

my two cents.....


vince

Thebrelon, IRS does indeed have accelerometers aswell, since gyros alone cant give any measurement  of the aircraft's movement through space.

Another valuable note: the IRS today aren'y really used as a navigation platform.  The FMC on modern airplanes are certified to navigate accurately with  VOR/DME and GPS. The FMC always assumes that radio position is more accurate than IRS, always.

What the Inertial Reference does provide is to sense and compute linear accelerations and angular turning rates about every axis. This data is used for pitch and roll displays.

You need to not move the aircraft during alignment because the IRS/INS automaticly assumes acceleration during alignment is due to the Earth's gravity, and the only motion during alignment is because of the Earth's rotation.

This phenomena explains why it takes so much longer to align the IRS/INS close to the poles, where rotation is much more slower, compared to the equator.

No commercial aircrafts have inertial systems that can be aligned during motion afaik. However the Military has had it for years (in submarines, for example). That's how it used to be, may very well have changed since my books were printed. 

The laser gyros are indeed a huge improvement compared to their mechanical counterparts.
The ring laser gyros consume less power, generate less heat and also decrease alignment time. That combined with  accelerometer improvements has resulted in significant decrease in weight, size and cost and of course, increase in reliability.

Simply put; inertial accuracy is a direct function of technology, The more sensitive accelerometers, the more stable the gyros, and the more sophisticated the torquing software and hardware...the more accurate the system

Also had a healthy debate about ring laser here
http://www.simviation.com/cgi-bin/yabb/YaBB.cgi?board=paywaregen;action=display;...


Think we're moving slightly off topic 

Laser Gyros do indeed drift, watch after a flight the IRS position, and compare it to the FMC position, they will always disagree (not by much though!) And even during flight, the individual IRS position will disagree, sometimes more, and sometimes less.

However depending on the hardware and software set-up you can eliminate drifting down to <1 nm during a flight.

Effects that are compensated for are schuler-tuning, coreolis effect, coning, sculling etc.
Stuff that really doesn't belong here