Gyro drift
The slow drift of the axis of rotation of a gyroscope instrument from its original position is called gyro drift . A top is (among other things) an integral part of the artificial horizon in the aircraft.
Basics
Gyro drift is caused by small mechanical imperfections and leads to errors in direction and position measurements, especially in flight navigation over long distances, which - if not corrected - can lead to significant course deviations.
The drift of a gyro depends on the mechanical quality of the rotor and its bearing and can be kept below about 0.01 ° per hour with today's gyro platforms . It has systematic and random components, of which the former predominate and increase linearly to quadratically with time. Their effect on dead reckoning can therefore partly increase with the 3rd power of time. B. the inertial navigation (INS) is limited to accuracies of about 1 nautical mile per hour.
A fully cardanically suspended top with a horizontal axis of rotation would apparently drift off at the geographic North Pole in 24 hours by 360 degrees or 15 degrees per hour, due to the rotation of the earth, since the top is stable in space and the earth rotates. At the equator this would not be the case, unless a gyro is used whose axis of rotation is vertical, in this case one speaks of tilting rather than drifting.
Causes and compensation calculation
The most important cause of a gyro drift is a small mechanical imbalance in the rotating gyro body, which is inevitable due to manufacturing tolerances. It causes an individual drift component which, after lengthy analyzes, can be largely compensated for mathematically using mathematical-physical models and filter methods . In addition, the systematic precession , nutation, acts .
Further influences are changes in temperature , internal temperature gradients , influences of bearing friction , interactions with accelerometers ( accelerometers ), as well as various external disturbances (mechanical shocks , electrostatic effects, vibrations , etc.) that cannot be avoided in traffic engineering . Magnetic and many electrical influences, on the other hand, can largely be avoided by placing the most delicate sensors as far away from the cockpit as possible .
In addition to the above-mentioned modeling, which is particularly important for inertial navigation , the effect of gyro drifts can also be recorded or reduced using other measurement methods. That includes
- the Doppler radar , which enables the flight speed to be continuously measured and thus prevents the accumulation of small errors in accelerometers as they arise,
- various methods of radio navigation - but only indirectly via location determination
- Support by magnetic field probes (see also Gyrosyn ),
- Control loops such as the Schuler period in inertial navigation,
- Position and speed comparison with satellite navigation systems .
- For about two decades, various of these methods have been combined to form largely optimal computer models under the name of integrated navigation .
Comparative calculation
For the mechanical compensation of gyro drifts, which today is increasingly being replaced by post-computational procedures, very small manipulated variables are required in order not to additionally disturb the mechanically sensitive gyro systems. With current systems, in spite of modern modeling methods, the sufficiently accurate detection of the drift errors cannot always be guaranteed. The development of the laser gyro - which is not a rotating gyro in the strict sense of the word - has not increased its accuracy significantly.
Therefore, z. B. a really reliable inertial navigation the combination of at least two independent systems. If the location of these two INS now diverges beyond the tolerable amount, a third measuring system - e.g. from radio navigation - is required to be able to decide between the correct display and the falsified one.
See also
literature
- Wolf von Fabeck : Gyro systems . The different types of devices and their technical applications, errors due to the principle and technical solutions, physical principles. Vogel, Würzburg 1980, ISBN 3-8023-0612-0 , chap. 3, 5 and 13f.