Forgive me for intervening, for I am not a world expert on INS.
It seems to me that we are talking here of two different classes of errors linked to the Schuler characteristics of the system, those caused by internal system faults and those caused by external problems.
As Oxford Blue says, any INS system has to determine during set-up and maintain during flight a value of earth gravity in order to establish first the vertical and then the horizontal navigational axes, N/S and E/W. To maintain these datums once set up and free to navigate it has to respond to movement about the earth as though it were a Schuler pendulum, rotating the horizontal axes by one degree for every 60nm moved. This means that for any error in the earth gravity loop, that is alignment or function of the accelerometers or of the first stage integrators, there will result bounded errors of velocity and position oscillating at the Schuler frequency. Errors in the gyros or in the second stage integrators will produce steadily increasing errors, sometime with a Schuler oscillation superimposed.
However, a basic INS may assume it is working over a uniform spherical earth, and the earth is neither uniform nor spherical. This will mean that the calculated position of the vertical and horizontal axes will diverge from the actual earth axes with movement around the earth, and that this will introduce Schuler errors
even in an INS that is internally error free. There is also the problem of errors due to vertical aircraft acceleration and the navigation errors that result from flying at different heights above the earth’s surface
An INS that contains a model of the earth that is a geometrically correct oblate spheroid will be more accurate, and one that contains a map of the earth’s gravitational fields even more accurate. This data is available, but only the latest INS systems have the capacity to use it.
It is comforting to know that errors due to external factors are small, much smaller than those that routinely appear from internal INS defects.
I thought the Marconi summary was very clear and good.
Dick W