During the Apollo program, the IMU did drift and it was standard procedure to periodically correct it. This was done by having the astronauts locate a pair of known stars using a sextant, and then entering their positions using computer program P52 "IMU Realign":
In order to check the alignment of the guidance platform, reference had to be made to the stars. Only two were required and Collins used the sextant, a 28-power optical device built into the spacecraft’s hull, to take marks once the instrument had been precisely aimed at each one.
Mike would enter the star code and ask the computer to drive the optics to aim at that star. By looking through the sextant eyepiece, he could see how far the star was from its expected position by virtue of the inertial platform having drifted. He then carefully moved the sextant’s aim to centre the star and pressed a button upon perfect alignment, thereby informing the computer where the star really was.
Once Mike had done this for two stars, the computer told him the star angle difference, a comparison of the known angle between the stars and the measured angle between Mike’s marks. '00000' for the top cluster was known as 'all balls' and made Mike a happy man for it meant that his aim at the two stars had been accurate to less than one hundredth of a degree. The star angle difference for the lower cluster shows '00001', one hundredth of a degree difference - still considered very good.
The point of the exercise was to realign the slowly drifting inertial platform. For this, the computer calculated the error in its orientation. This yielded three Euler angles by which the platform’s orientation had to be rotated or ‘torqued’ in order to restore perfect alignment with the stars, in thousandths of a degree. These are the next three items in the clusters.
The alignment of this platform was crucial to some of the most important operations of the spacecraft; navigation and engine burns. Engineers were therefore keen to keep an eye on its rate of drift. Mike would bring all three angles onto the three register displays of the DSKY by asking it to display Noun 93. Since the flight controllers were able to see the contents of the DSKY’s display on their consoles, this gave them an opportunity to take a note of them.
Realignments done during Apollo 11 are shown in the following table:
Occasionally CMP Michael Collins would jot down the results of the corrections onto the walls of the spacecraft. The two corrections highlighted in yellow in the table above can still be seen today as graffiti on the walls of the Columbia in the National Air and Space Museum: