Inertial Navigation Systems — Page 235, Lesson 279

Let's start with the alignment sequence, because that's the heart of getting an Inertial Navigation System ready to work. The stable element — that's the platform inside the INS that holds the accelerometers — must be accurately aligned in both azimuth and attitude. Azimuth is your heading direction, and attitude is your pitch and roll. The whole point is to let the accelerometers measure acceleration along their chosen axes, and if the platform isn't aligned, they'll measure the wrong components. The first stage is the warm up period. The INS contains fluid-filled components, and they need to be brought to the correct operating temperature. This phase normally takes between 3 to 4 minutes. You can't skip it, because the fluid's behaviour changes with temperature. Next comes coarse alignment. The platform is roughly levelled and aligned in azimuth. The purpose here is to remove gyro alignment errors and cut the time to a minimum. Within coarse alignment, we have two sub-steps. First, coarse levelling. Pitch and roll are driven until they are at 90° to each other. The platform is then roughly levelled using either the aircraft frame as reference, or using the outputs from gravity switches or the horizontal accelerometers. So you have a choice of reference sources. Second, coarse azimuth alignment. This is achieved by turning the platform until the heading output agrees with the aircraft's best known True Heading. Note that word "True" — this is heading relative to true north, not magnetic. The result of coarse alignment is that it levels and aligns the platform within 1° to 2° in a few seconds. So it's fast, but not precise. Then we move to fine levelling. With zero output from the accelerometers, fine levelling is achieved. The process takes anything up to 1 to 1½ minutes, and it levels the platform to within 6 seconds of arc. That's a very tight tolerance — 6 seconds of arc is a tiny fraction of a degree. Now, gyro compassing. This is how the platform gets aligned in azimuth precisely. The gyro normally used to stabilize the platform about an east-west axis is connected to the azimuth gimbal motor. Here's the principle: with the platform correctly aligned in azimuth, the east gyro should not be subject to rotation of its input axis due to earth rotation. When the platform is out of alignment, the east gyro will detect a component of earth rotation. The resultant output signal can then be used to torque the azimuth gyro until the table is aligned. So the earth's rotation itself becomes the reference. If the platform is aligned to true north, the east gyro sees no earth rotation component. If it's off, it sees one, and that error signal drives the platform back into alignment. Two more conditions must be met. The accelerometers must be levelled, meaning velocity is set to zero. And the platform must be orientated to true north — that's the gyro compassing step, with position verified. Now let's talk about the Schuler Period, because this is the concept that makes the whole system stable. Schuler postulated an earth pendulum — imagine a pendulum whose length equals the radius of the earth. Its bob is at the earth's centre, and its point of suspension is at the earth's surface. If the suspension point were accelerated around the earth, the bob would remain vertically below the suspension point, because it sits at the earth's centre of gravity. Now, a platform mounted on the suspension point, tangential to the earth's surface — that is, horizontal — would therefore remain horizontal irrespective of the acceleration experienced. The vertical defined by the normal to the platform is therefore unaffected by acceleration. That's the key insight: the platform's vertical reference doesn't get disturbed by the aircraft's acceleration. But here's the catch. If, for any reason, the bob on the earth pendulum became displaced from the earth's centre, the pendulum would start to oscillate. And the oscillation period would be 84.4 minutes. That's the Schuler Period — a natural oscillation of about 84.4 minutes. So how does the INS use this? The stable element is maintained normal to the local vertical by feeding back the aircraft's radial velocity as levelling gyro signals. In this way, the north and east accelerometers are prevented from detecting components of the gravity acceleration. The system is tuned so that any error oscillates with that 84.4-minute period rather than growing unbounded. Let me tie this together. The alignment sequence gets the platform precisely oriented. The Schuler tuning keeps it stable and horizontal during flight, so the accelerometers only measure true horizontal acceleration, not gravity components. That's the foundation of how an INS navigates.