Re: [Flightgear-devel] Typical NAV system errors

[Thorsten R. <tho...@sc...>]

Thanks for all the additional information.

I think for my particular modeling case I'm primarily interested in the  
intrinsic accuracy of the radio signal received with near-optimal  
equipment.

* The Shuttle doesn't use TACAN directly as guidance signal, but as one of  
several sources to constrain the state vector updates with Kalman  
filtering merging the various sources of information

* There are three receivers aboard with a veto/averaging procedure run to  
deal with deviations among them

* I *suspect* NASA didn't run the Shuttle on cheap hardware or skip on  
maintenance, so we may assume all receivers have been calibrated and  
tested a few weeks at most before usage, and also that NASA wouldn't rely  
just on map information so that any correction of the real alignment have  
been coded into the GPC

* The altitude (and hence the slant angle) is generally fairly high which  
should suppress ground effects - according to SCOM TACAN can first be  
received about 400 miles out from 160.000 ft, and at TAEM interface when  
it's most important to have, we're at 85.000 ft with 60 miles to the  
runway. The last altitude at which TACAN matters is ~ 12.000 ft by which  
time MLS signal should be well incorporated (or TACAN is dismissed and the  
rest runs on inertial guidance).


I'm not (yet?) at the point where I want to model deviations of the  
individual receivers and the averaging-out procedure (this might be  
slightly over the top...), but I want to get the general picture of the  
'volume of uncertainty' right when approaching with pure inertial guidance  
or add TACAN, barometric altitude, MLS or GPS.

Thanks again, the links have been very instructive!

* Thorsten