Attitude and Heading Reference Systems Questions and Answers

Answer: A

Correct answer: Statement I is correct but Statement II is incorrect

Statement I is correct. A Kalman filter can include the gyro biases as states in its model, so that each time the accelerometer or magnetometer reference disagrees with the propagated attitude, the filter attributes part of the discrepancy to bias and updates its estimate. The biases are then subtracted from the raw rates continuously.

Statement II is incorrect. A fixed complementary filter has no states beyond the blended output and no covariance to reason with; it can only suppress the effect of a bias through its crossover, never estimate the bias itself. That capability is one of the main reasons the Kalman formulation displaced it.

Answer: B

Correct answer: Both Statement I and Statement II are correct

Both statements are correct and the second explains the first. Coning arises when two axes oscillate simultaneously, typically at airframe vibration frequencies, and because finite rotations do not commute the resulting attitude drift is invisible to an integration loop too slow to resolve the oscillation.

The compensation term must therefore be computed at a rate well above those frequencies - commonly a few hundred hertz or more - even though the outer filter that blends in the reference sensors may update far more slowly. Splitting the computation this way keeps the fast, error-critical part cheap while the slower blending runs at a comfortable rate.

Answer: C

Everything afterwards is computed as a change from this initial state, so an error here persists for the whole flight. Movement during alignment corrupts the gravity reference, which is why the aircraft must be held still.

Answer: D

Prediction from the gyros comes first and correction from the reference sensors second, which is the basic structure of every such filter. The bias estimates improve on each cycle, which is what bounds the long term drift.

Answer: A

Correct answer: 1, 2, 3, 4

The sequence follows the technology. Gimballed spinning mass vertical gyros (1) are the earliest, dating from the interwar period and providing attitude alone. Gimballed inertial platforms (2) followed, adding accelerometers on a stabilised platform to give full navigation from the 1950s.

Strapdown systems with ring laser gyros (3) came once processors could perform the coordinate transformation in software, from the 1970s and 1980s. Strapdown systems with micro-electromechanical sensors (4) are the most recent, bringing the architecture within reach of general aviation and unmanned aircraft from the 1990s onward.

Answer: B

Correct answer: 1, 2, 3, 4, 5

The chain runs from cause to correction. A small constant bias exists on one gyro (1); the attitude algorithm integrates the rate, so the bias accumulates into an attitude error (2), which grows linearly with elapsed time (3) because the offset is constant.

Eventually the error is large enough that the accelerometer-derived vertical disagrees with the computed attitude (4), and the filter uses that residual both to correct the attitude and to revise its estimate of the bias (5). The last step is what distinguishes a Kalman filter from a fixed blend: it does not merely mask the symptom, it learns the cause.

Answer: C

Correct answer: 1, 2, 3, 4, 5

The steps run in the order the logic requires. Each of the three units first computes an independent solution (1); the solutions are then compared pairwise (2), which is what makes disagreement visible.

The comparison shows that one unit disagrees with both of the others (3), identifying it as the outlier - something two units could never establish. Median voting then excludes it (4), and the system continues on the two that agree (5). Independence at step 1 is essential: units sharing a sensor, a power supply or a common calculation could fail together and defeat the vote.

Answer: D

Correct answer: Only three

Three pairs are right. The rate gyro triad measures angular rate about three axes; the accelerometer triad measures specific force along three axes; and the magnetometer measures the direction of the earth's magnetic field.

The fourth is wrong. An accelerometer triad does not measure angular displacement, directly or otherwise - it senses specific force, and attitude is inferred from it only by resolving the gravity vector, and only while the aircraft is unaccelerated. Angular displacement is obtained by integrating the gyros, never by reading an accelerometer.

Answer: A

Correct answer: Only three

Three pairs are right. Gyro bias gives attitude error growing linearly with time; angle random walk gives error growing as the square root of time; and an accelerometer bias in a navigation solution gives position error growing with the square of time, since it is integrated twice.

The fourth is wrong. A scale factor error is an error in the sensor's gain, so its effect is proportional to the motion being measured: it vanishes when the aircraft is still and is largest in a fast roll. Independence from the motion describes bias, not scale factor.

Answer: B

Correct answer: Only three

Three pairs are right. Levelling on the accelerometers establishes initial pitch and roll from the sensed gravity vector; gyro compassing establishes true heading from the horizontal component of earth rate; and a magnetometer reading gives magnetic heading.

The fourth is wrong. Coning compensation is a correction applied during attitude integration to account for the non-commutativity of rotations under two-axis oscillation. It establishes nothing about position - an attitude and heading reference system does not compute position at all.