576 CHINESE OPTICS LETTERS / Vol. 4, No. 10 / October 10, 2006
Investigation on the temperature compensating model for
ring laser gyroscope
Chuang Guo (
)
1
, Yo ng jun Xu (
þþþ
)
1
, and Xiaoning Zhao (
ÛÛÛ
)
2
1
Engineering Institute, The Air Force University of Engineering, Xi’an 710038
2
Flight Automatic Control Research Institute, Xi’an 710065
Received February 27, 2006
Thermal effects impose the greatest limit on the precision of a ring laser gyro (RLG). Selections of temper-
ature sensing points were comparatively discussed based on large numbers of experimental data to impro ve
its precision, and the optimum combination was selected to establish a practical compensating model. The
model is applied to n ew experimental data under the given and varied temp eratures. Results show that
the bias trend changing with the temperature is basically eliminated and the bias stability is enhanced
significantly.
OCIS codes: 140.3370, 060.2800, 120.6810, 120.6780, 070.6020.
Ring laser gyroscope (RLG) is a kind of optical gyroscope
by means of the Sagnac effect, and it is one of the most
ideal devices for the Strapdown Inertial Navigation Sys-
tem (SINS). But the temperature characteristic of RLG
brings lots of inverse influence and confines the further
increase of its precision. Therefore, necessary temper-
ature control or temperature compensating measures
must be put into practice in order to improve gyroscope
precision when it is requested to work in high accuracy
situation. Generally, we always adopt the soft technique
such as temperature compensating.
The temperature field becomes complex when the RLG
is working, the compensating model is difficult to realize
high-precision, many scholars have made in-depth study,
their works
[2−5]
mostly concentrate on the mechanism
analysis of RLG and test data processing. Little work
is done in the selection of the sensor position. With
those model to compensate some RLG, the compensat-
ing effect is not well. In the course of building the right
model, selection of the sensor position (temperature sens-
ing point) is an important factor.
On the basis of lots of environmental temperature ex-
periments, we analyze the choice of several temperature
sensing points of some RLG, and select the best temper-
ature sensing point to set up model, at the same time,
the model is validated experimentally.
A typical quadrangular high-precision single-axis RLG
is shown in Fig. 1. There is a quadrangular beam of light
Fig. 1. High-precision single-axis RLG.
channel in the cavity, and the reflectors are installed in
four big holes at the top angles. There are two anodic
subassemblies and a public cathode in the RLG.
This structure has some more typical applications than
other ones, internationally which mainly include LG8028
and LG9028 (American company Litton), GLS16 and
GlS32 (French company SAGEM), and SINGLE33 and
Pixyz22 (French company SEXTENT), etc..
Temperature is the main cause to the change of bias,
as depicted in Refs. [1—4], which is the primary error
source to the RLG, the temperature influences on the
RLG mainly behave in the following several aspects.
Firstly, in view of the hot source, when RLG is working,
it will have a fever, the course to achieve balance will al-
ways spend several hours, furthermore, the temperature
field of RLG will become very complex when the sur-
rounding temperature is changing, and it is difficult to
keep balance.
Secondly, in view of the physics characteristic, gaseous
refractive index, heat transmits coefficient of material,
and optical characteristic of optical devices etc will
change with the variety of temperature.
Thirdly, in view of the geometrical characteristic, ther-
mal expansion, shrink, and flexural deformation will all
lead to the change of optical loop. It will result in the
increase of the resonance system’s spoilage.
When the resonant cavity length changes with the tem-
perature, the relative error can be expressed as
η =
ΔL
L
=
ΔT · L · α
L
= α · ΔT, (1)
where L is the resonant cavity length, ΔL variance of
length, ΔT variance of temperature, and α thermal ex-
pansion coefficient.
At present time, Zerodur crystallite glass materials
mostly produced in Germany are usually used to man-
ufacture the high-precision RLG. Its thermal expansion
coefficient is 5 × 10
−8 ◦
C
−1
, when the variance of tem-
perature is 100
◦
C, the relative error is 5 ppm, from
the formula of scale factor K ≈ L/4λ,wecanseethat
its effect which will cause the variety of bias cannot be
ignored.
1671-7694/2006/100576-04 http://www.col.org.cn