Note : Les descriptions sont présentées dans la langue officielle dans laquelle elles ont été soumises.
CA 02416166 2003-01-13
Method and device for compensating firing errors and system computer for
weapon system
The present invention relates to a method and a device for compensating firing
errors of a gun, having a weapon barrel, of a weapon system, which are caused
by static gun geometry errors, according to the preambles of Claims 1 and 12,
respectively, and a system computer for a weapon system according to the
preamble of Claim 17.
In principle, the present invention relates to all possible static gun
geometry
errors and their compensation.
Guns comprise numerous individual parts which are connected rigidly or
movably to one another. The individual parts can never be produced with
precise dimensional accuracy, but rather only with certain manufacturing
tolerances and/or deviations from the theoretically determined dimensions, and
CA 02416166 2003-01-13
2
deviations, within the fixed assembly tolerances, from the intended mutual
positions also result during assembiy. The totality of the deviations has the
consequence that every gun has deviations from its ideal geometry, which are
referred to as gun geometry errors. Such guri geometry errors are composed of
numerous types of errors. For example, gun geometry errors are manifested in
that azimuth a of the weapon barrel in the zero position, as it is indicated
by an
azimuth display of the gun, is not equal to 0 in actuality, but deviates from
0 by
a slight angle Da. Correspondingly, elevation A of the weapon barrel in its
zero
position may not have the value 0 indicated by the elevation display of the
gun,
but rather may deviate by a siight angle AA from 0 . In certain cases, Aa and
DA may be equal to zero, but only if differerit gun geometry errors compensate
one another.
The manufacturing tolerances may be equal or approximately equal for identical
individual parts of a series of guns, if such individual parts are always
produced
on the same machines, using non-wearing or precisely adjustable tools and in
identical external conditions, such as temperature conditions. However, after
assembly the gun geometry errors will be different from gun to gun.
The problem is worsened in that the gun geometry errors, particularly the
angular errors, are not constant, but rather change for different reasons. For
movable individual parts, such changes are primarily the consequence of wear;
therefore, they increase over the course of time. The change of the errors is,
however, also connected to the existing environmental conditions, such as the
air and gun temperatures; they may therefore alternately increase and
decrease.
A further complication arises in that the gun geometry errors are also
influenced
by the respective positions of the individual parts, since the mechanical
loads
and therefore the deformations of the iridividual parts are partially
dependent on
position.
CA 02416166 2003-01-13
3
Finally, the gun geonietry errors which manifest iri a specific position of
the
weapon barKel and at a specific tirrie may also be a function of the
rotational
direction in which the weapon barrel is brought into this specific position.
The gun geometry errors characterize the individual guns and therefore
represent actual gun parameters. Firing errors andlor a reduction of the
accuracy performance of the gun result as a consequence of the gun geometry
errors, particularly as a consequence of the angular errors. Due to the large
distances between the muzzle of the weapon barrel and the targets which are to
be hit by the projectiles fired from the weapon barrel, even slight angular
deviations of the weapon barrel cause significant deviations of the
projectiles
from the targets to be combated.
If the gun geometry errors and/or guri parameters are known, then the firing
errors which they cause may be compensated, in that the gun parameters may
be taken into consideration in addition to other data by the software of a
computer assigned to the gun during the determination of the aiming values.
The concept of a computer assigned to the gun is to be understood to mean a
gun computer and/or a computer of a fire control device. Other data which is
to
be taken into consideration by the computer particularly iricludes target
data,
which describes the location and the movernent of the target, meteorology
data,
which describes the respective meteorological conditions, vo data, which
relates
to the deviation of the actual muzzie velocity from a theoretically determined
muscle velocity, and possibly shell data, which characterizes the respective
shells fired.
The determination of the gun geometry errors and/or gun parameters, their
evaluation to obtain correction furictions, and the implementation of the
correction functions in the software of the computer rnust be performed before
the gun is put into operation, and must be done individually for each gun.
The previously known methods for measuririg the gun parameters have
numerous disadvantages. Not all types of gun geometry errors may be
CA 02416166 2009-09-28
4
measured. The nieasurenients canriot be performed in an automated way and
therefore require a large arnourit of tirne; as a consequence, only a few
measurements are niade per rneasureiYierit position of the weapon barrel,
which has the consequence that random measurernent errors cannot be
eliniinated. The measurements not only require a large amount of time, but
also require a relatively large number of personnel, so that they are very
costly.
In addition, some of the measurernent personnel are subjected to relatively
great danger, since they must be located in the region of the weapon barrel
muzzle to perform the measurements; for larger elevations and long weapon
barrels, this means that measurement personnel must be lifted into the region
of
the weapon barrel muzzle using a lift device or rneasurements must be
performed on a ladder.
The object of the present invention is therefore,
- to indicate a method for compensating firing errors of the type initially
described which allows complete detection of the gun geometry errors
and may be performed precisely, rapidly, using few personnel, and
preferably automated;
to suggest a device for performing this method; and
- to suggest a fire control computer and/or system computer for a weapon
system, to which the novel device may be coupled.
According to the present invention, there is provided a method for
compensating
firing errors of a gun having a weapon barrel, caused by static gun geometry
errors,
which influence the position of the weapon barrel during aiming of the weapon
barrel
at aiming values,
- the weapon barrel is brought into measurement positions in steps by rotation
around an axis,
- at each measurement position
CA 02416166 2009-09-28
4a
- an intended value, which describes the intended position of the weapon
barrel, and
an actual value, which describes the actual position of the weapon
barrel, are detected,
- a difference between the actual value and the intended value, defined
as an error value, is calculated,
correction values are established from multiple error values and
- the correction values are taken into consideration during later aiming of
the
weapon barrel.
According to the present invention, there is also provided a device for
compensating
firing errors of a gun having a weapon barrel, these firing errors being
caused by
static gun geometry errors, which influence the position of the weapon barrel
during
aiming of the weapon barrel at calculated aiming values, this device having a
measurement facility for establishing actual values, which describes the
position of
the weapon barrel, the measurement facility having an optical-electronic
gyroscopic
measurement system on the weapon barrel, having a first measurement unit, in
order
to detect azimuth synchronization error and possibly perpendicular offset
error.
According to the present invention, there is also provided a system computer
of a
weapon system for calculating aiming values for aiming a weapon barrel of a
gun of
the weapon system, characterized in that:
- the system computer has a data input for data representing correction
values,
- the system computer calculates aiming values, based on said data
representing correction values, compensating aiming errors caused by static
gun geometry error which influence the position of the weapon barrel.
CA 02416166 2003-01-13
- AII angular errors which are caused by static gun geometry errors may
be detected and therefore conipensated.
- Static gun geometry errors, which until now could only be determined
imprecisely and at great cost, may now be measured precisely and may
correspondingly be efficieritly compensated.
- The use of a gyroscopic measurement system allows angular
measurements to be performed without previously leveling the horizontal
of the weapon.
- The use of an optical-electronic gyroscope, particularly a fiber-optic
gyroscope, allows angular measurements to be performed whose
precision, reliability, and reproducibility greatly exceeds the previously
performable measurements and whicti provide significantly more detailed
measurement results than those which could previously be achieved; in
this way, much more precise compensations of the firing errors caused
by the gun geometry are made possible.
- The measurements may bc, performed rapidly and automatically; the
outlay in time and personnel for measuring a gun is low, which results in
significant cost savings.
- The danger of accidents for persons taking part in the measurements
may be greatly reduced.
Before the invention is described in more detail in the following, several
basic
concepts will be explained.
Although only azimuth synchronization error, elevation synchronization error,
perpendicular offset error, wobble error, and squint error, as well as their
compensation, are described in more detail in the following, the basic idea of
the present invention is applicable to all gun geometry errors which occur.
The weapon barrel, whose position is influenced by the gun geometry errors,
may be brought into various positions through back-and-forth pivoting or
complete rotation, each position being defined by the corresponding azimuth,
i.e., the corresponding lateral angle, and by the corresponding elevation,
i.e.,
CA 02416166 2003-01-13
6
the corresponding vertical angle. A rotation around the vertical axis changes
the azimuth and a rotation around the lateral axis changes the elevation. The
vertical axis and the lateral axis are two axes of a spatial, preferably
orthogonal,
axis system, whose axes are defined in Table 1. In the framework of the
present description, the azimuth is understood to be not the deviation from
north, as in firing operatiori, but from a zero position.
Table I
definition of the axes
L axis lateral axis (theoretical) horizontal axis, around which the
weapon barrel is pivotable; elevation A is set in this
way;
A axis vertical axis (theoretical) vertical axis, around which the weapon
barrel is
pivotable; azimuth a is set in this way.
R axis longitudinal axis (theoretical) horizontal axis of the weapon barrel in
the
tied down position, with azimuth a = 0 and elevation
A =0;
Firing errors occur because the actual position of the weapon barrel is not
equal
to its intended position. The intended position is defined by, among other
things, the values for azimuth and elevation established by the fire control
computer and/or system computer, but is not assumed due to static gun
geometry errors. The angular error of the position of the weapon barrel which
occurs, the gun geometry errors which cause it, and the primary causes of the
gun geometry errors may be seen in Table 2. The angular errors, which
manifest as azimuth errors and elevation errors, comprise the following five
3() types of errors, which, however, are not independent of one another:
(1) azimuth synctironization error Aa1
(2) wobble error Ar
CA 02416166 2003-01-13
7
(3) elevation synchronization error AA
(4) perpendicular offset error Aa2
(5) squint error Aci
Table 2
Angular errors of the position of the weapon barrel, gun geometry errors, and
their causes
CA 02416166 2003-01-13
8
Angular Gun geometry Causes
errors errors
- - -------- - ------~..- - Azimuth Aa1 azimuth 1. Eccentricity of the lateral
pivot bearing
errors (lateral synchronization 2. Out-of-round of the lateral pivot bearing
errors) error 3. Variable tooth intervals in the crown
gear of the lateral pivot unit
4. Coder error
Aa2 5. Tilting of the elevation axis toward the
perpendicular offset horizontal
error 6. Non-orthogonality of barrel axis and
elevation axis
Acr squint error 7. Non-parallelism of barrel axis and line
of sight
elevation AA elevation 8. Eccentricity of the vertical pivot bearing
errors synchronization 9. Out-of-round of the vertical pivot
(vertical error bearing
errors) 10. Variable tooth intervals in the crown
gear of the vertical pivot unit
11. Coder error
12. Backwards skipping of the gun with
increasing elevation
Ar 13. Elasticity of the structure
AQ squint error 7. Non-parallelism of barrel axis and line
of sight
In order to determine these partial errors, multiple measurements are
performed. For an efficient procedure it is advantageous to perform the
measurements in the course of three rrieasurement procedures, since in each
position of the weapon barrel, rneasurements are performed which relate to
more than one type of error. Three measurement procedures, the partial errors,
and the respective measurement devices used are shown in Table 3.
CA 02416166 2003-01-13
9
Table 3
Angular errors, measurement procedures, and measurement devices
Measurement relates to partial Measurement
procedure error device
1 Azimuth Da1 Gyroscopic
synchronization Ar measurement
error device
Wobble error spirit level
2 Elevation Olk Gyroscopic
synchronization Ocx2 measurement
error device
Perpendicular offset Gyroscopic
error measurement
device
3 Squint error Aa Optical device
(target telescope)
For compensating the firing errors, which are based on static gun geometry
err-ors of a gun, in principle the procedure is: as follows: an angular error
which
arises during movement of the weapon barrel around one of the rotational axes
is determined. The weapon barrel is brougtit into a final position, which is
also
a measurement position, from a zero position in steps, by rotation in one
rotational direction around the rotational axes described, via sequential
measurement positions. The rotation is controlled by a computer. Using a
suitable measurement unit of a measurement facility, after each step the
actual
angle around which the weapon barrel has rotated is determined; this angle is
referred to as the actual value. Sirnultaneously, after each step the
theoretical
angle around which the weapon barrel is to have rotated, for example in
accordance with information on a scale on the gun or on the assigned fire
control computer and/or system computer, is determined; this angle is referred
to as the intended value. The angular difference between the intended value
and the actual value is then calculated for each measurement position; this
CA 02416166 2009-09-28
difference is referred to as the error value. A correctiori value is
established
from the error value, which is implemented in the software of the fire control
computer and/or systerri computer and is subsequently taken into consideration
in the deterrrlination of the aiming values, i.e., the values for azimuth and
elevation. The aiming values are primarily calculated using target data, i.e.,
ciata which describes positions and possible movements of a target to be
combated, and using ballistics data. This primary calculation is corrected
with
the aid of the method according to the present invention.
In particular, the actual values may be represented as a function of the
intended
values to establish the correction values and may be prepared in such a way
that the correction values may be determined therefrom. Such a preparation, in
which correction values result from the rneasured angular errors, may be
performed numerically and/or with tabular aids or mathematically or
numerically/mathematicaliy combined.
Preferably, for the numeric method, value pairs are stored in a table, a first
value
being the intended value and a second value being the actual value or the
difference
between the actual value and intended value in each value pair. The value
pairs may
also be considered as an empirical error curve. The table and/or the empirical
error
curve is then available during the calculation of aiming values in such a way
that the
calculation of each aiming value is performed in a corrected way, taking into
consideration the corresponding values of the table and/or the empirical error
curve.
Preferably, for the mathematical method, the error values are first
represented in
tabular form as a function of the intended angle and/or as an empirical error
curve
and then approximated by at least one mathematical function; i.e., the
empirical error
curve is either approximated over its entire course by one single mathematical
error
function or in each section by a mathematical partial error function, and thus
as a
whole by multiple mathematical partial error functions. The mathematical error
CA 02416166 2009-09-28
11
function is then made available to the computer, which determines a correction
function therefrom, which it takes into consideration during the calculation
of the
aiming values for the weapon barrel, i.e., the azimuth and the elevation.
The numerical method may be designed in such a way that the necessary
precision
for the compensation of the firing errors is ensured. However, as described
below,
the mathematical methods have the advantage that mathematical error functions
may be analyzed simply, specifically using known mathematical methods; not
only
may the values for the compensation of the firing errors be obtained
therefrom, but
also insights into the influence of individual constructive conditions on the
error
functions; constructive improvements resulting therefrom serve, in the final
analysis,
to combat the firing errors caused by gun geometry at the root, in that the
gun
geometry errors are eliminated. The concept of constructive relates to both
conceptual conditions, and to conditions relating to production and assembly.
To remove random measurement errors, it is advantageous to repeat the
measurement procedure described above once or multiple times and to average
the
values obtained in tabular form. Alternatively, an average empirical error
curve may
be formed from all identically performed measurement procedures or a
mathematical
error function may be formed from each empirical error curve and an average
mathematical error function may be formed from these functions or a correction
function may be formed from each empirical error curve and an average
correction
function may be formed from all correction functions.
Preferably, for the measurements described above, the rotation of the weapon
barrel
is always in the same rotational direction; the error values obtained in this
way are
mono-directionally determined error values, which may be numerically or
mathematically prepared. In particular, the empirical error curve and/or
mathematical
error function is a mono-directionally determined and/or mono-directional
error curve
and/or error function. The error values are, however, as described above,
generally a
function of, among other things, the rotational direction in which this
rotation was
performed. It is therefore advantageous to
CA 02416166 2003-01-13
12
perform two measurements. For this purpose, ttie weapon barrel is rotated,
around the- same rotational axis, in one rotational direction for the first
measurement and in the opposite rotational direction for the second
measurement. The measurernent positions of the first-directional rotation and
the measuremerit positions of the second-directional rotation may correspond,
but do not have to. During these rotations, first-directional and second-
directional error values are establistied. If the deviations between the first-
directional and the second-directiorial error values are small, then a
direction-
free error value may be established and prepared and/or analyzed further. In
particular, an average direction-free empirical error curve may be
established,
from the first-directional empirical error curve and the second-directional
empirical error curve, from wtiich an average directiori-free mathematical
error
function and, from this, an average direction-free correction function may be
established, the correction function being taken into consideration in the
calculation of the aiming values. Since, however, the influence of the
rotational
direction results in a systematic error component of the overall error values,
both the first-directional error values and the second-directional error
values are
preferably prepared and/or analyzed separately.
Depending on the errors to be detected, as already described, various
measurement devices are used. In particular, spirit levels, preferably
electronic
spirit levels, and gyroscopic nieasurement systems, preferably optical-
electronic
gyroscopic measurement systems, these being understood to include, for
example, ring laser gyroscopes and fiber-optic gyroscopes, are used. The
measurement devices must generally be calibrated after being mounted on the
gun and/or on the weapon barrel, before beginning a measurement procedure.
For the use of gyroscopic measurerrient systems, the continuously changing
gyroscopic drift must also generally be detected and the values measured must
be corrected in accordance with the gyroscopic drift. An example of the
detection and consideration of the gyroscopic drift is described in European
Patent Application 00126917.4.
CA 02416166 2009-09-28
13
The description above relates to establishirig a correction function, which is
based on detecting error values arising during the rotation of the weapon
barrel
around one of the axes. However, the weapon barrel is not rotated around only
one axis, but around two non-coincident, generally orthogonal axes. The first
axis is preferably vertical axis A and the second axis is preferably iateral
axis L,
azimuth a being set by rotation around vertical axis A and elevation A being
set
by rotation around lateral axis L.
In the course of a first measurement procedure, azimuth synchronization error
Aa1 and wobble error AT may be establishecl.
Preferably, to detect azimuth synchronization error Aa1, azimuth a of the
weapon
barrel is changed in steps at an elevation of 0 . For the mathematical
methods, the
azimuth errors established in this way provide an azimuth error curve which is
generally constituted so that it may be approximated by a sine function, a
rotation of
the weapon barrel by 360 corresponding to one or more periods of the sine
function.
A first measurement unit of the gyroscopic measurement system is used as a
measurement device.
Preferably, wobble error AT is also detected within the first measurement
procedure.
For this purpose, the rotations of the weapon barrel performed to detect
azimuth
synchronization error Aa1 may be repeated. However, the actual azimuth and the
intended azimuth and/or their difference are not detected and/or established.
The
actual angle of inclination of the weapon barrel axis to the horizontal is
detected; this
angle of inclination is referred to as the actual wobble angle and/or actual
value. The
theoretical angle of inclination, which is referred to as the intended wobble
angle
and/or intended value, is always zero in this case, since the measurement
procedure
is performed at an elevation of 0 . Therefore, the wobble movement during a
rotation
around vertical axis A is detected. However, it would also be possible to
perform the
measurement procedure at a constant angle of elevation which is not 0 ; in
such a
CA 02416166 2009-09-28
14
case, the intended wobble angle would correspond to this constant, theoretical
angle
of elevation, and the actual wobble angle would correspond to the deviation of
the
actual angle of elevation from the theoretical angle of elevation. A spirit
level,
preferably an electronic spirit level, is used as a measurement system.
Elevation synchronization error Ak and perpendicular offset error Aa2 may be
determined in the course of the second measurement procedure.
Preferably, elevation synchronization error A. comprises two components, which
may only be determined jointly.
Preferably, a first component of elevation synchronization error 0k is based -
analogously to the azimuth synchronization error - on the fact that the
respective
actual angle of the weapon barrel does not correspond to the intended angle. A
partial error curve and/or partial error function describing this component of
elevation
synchronization error Aa, has the nature of a sine function, possibly having
multiple
angular frequencies.
Preferably, a further component of elevation synchronization error 0k is based
on the
fact that the torque applied to the gun carriage by the weight of the weapon
barrel
becomes lower with increasing elevation; this torque has the tendency to
rotate the
weapon barrel downward; in a tied down position, for example with azimuth 0
and
low elevation, the gun would tend to tip forward. Due to the reduction of the
torque
with increasing elevation, the weapon barrel is pulled downward less, with the
consequence that the gun tips forward less and/or, in comparison to the tied
down
position, tips backward. The partial error curve and/or partial error function
which
describes this component of the elevation synchronization error has the nature
of a
cosine curve subtracted from 1 with a single angular frequency.
CA 02416166 2009-09-28
Preferably, the measurements of the second measurement procedure, using which
the elevation synchronization error is determined, run analogously to the
measurement procedure using which the azimuth synchronization error is
detected.
For the mathematical method, they provide an error function like a sine
function
corresponding to the first component of elevation synchronization error,
however, this
sinusoidal function does not oscillate around a horizontal, but around the
continuously rising curve of the cosine curve subtracted from 1, corresponding
to the
second component of the elevation synchronization error. The two partial error
functions may be separated mathematically. Such a separation does not have to
be
10 performed to calculate the corresponding correction function, since only
the result,
specifically the correction of the overall elevation synchronization error, is
significant.
The partial error functions may, however, possibly be of interest, because
they more
clearly display errors of the gun construction, the temperature dependence of
individual assemblies, the wear, and other things. A second measurement unit
of the
gyroscopic measurement system is used for the measurement.
Preferably, perpendicular offset error Aa2, which may also be established
within the
second measurement procedure, is based on the fact that elevation axis L and
azimuth axis A are not, as desired, orthogonal to one another, and the weapon
barrel
axis is not, as desired, orthogonal to elevation axis L. Even with the gun
leveled to
the horizon, a change of elevation a, results in an error of azimuth a.
Perpendicular
offset error Aa2 may in principle be described and/or appropriately corrected
using a
function which is essentially proportional to the sum of a tangent function of
~' and an
inverse cosine function of 7,, specifically Aa2 = a tg a+ b/cos ~, - b. At an
elevation of
90 or nearly 90 , correction may obviously not be performed on the basis of
this
function, since cos a, is infinite here. Perpendicular offset error Aa2 is
measured
using the first measurement unit of the gyroscopic measurement system.
CA 02416166 2009-09-28
15a
Preferably and finally, squint error D6 is detected in a third measurement
procedure.
This error represents the non-parallelism of the weapon barrel axis and the
line of
sight. Squint error Da is established and prepared in the method according to
the
present invention in a typical way, which is therefore not described in more
detail.
Further characteristics and advantages of the present invention are described
in the
following with reference to examples and in relation to the drawing.
CA 02416166 2003-01-13
16
Fig. IA shows a weapon system having a device according to the present
invention in a schematic illustration;
Fig. 1B shows a gun of the weapon system in Fig. 1A in a simplified
illustration, with three axes of an orthogonal axis system;
Fig. 2A shows a schematic illustration to explain the azimuth
synchronization error;
Fig. 2B shows empirical error curves of the azimuth synchronization error;
Fig. 3A shows empirical error curves of the wobble error,
Fig. 3B shows an empirical error curve of the wobble error; only the error
component caused by the lower gun carriage is illustrated;
Fig. 3C shows an empirical error curve of the wobble error; only the error
component caused by the leg support is illustrated;
Fig. 4A shows an empirical error curve of the elevation synchronization
error for a constant azimuth;
Fig. 4B shows elevation synchronization errors as a function of azimuth
with various elevations as a parameter; and
Fig. 5 shows an empirical error curve and a mathematical error function
of the perpendicular offset error.
Since detectable errors are known to be small in comparison to the absolute
values, for example of azimuth or elevation, diagrams which represent error
curves and error functions are not to scale, so that the course of the
functions is
clearly visible.
CA 02416166 2003-01-13
17
Fig. 1 A schematically shows a weapon system 10. A weapon system 10 has a
gun 10.1 having a weapon barrel 10.2, a fire control device 10.3, and a fire
control computer and/or system computer 10.4. A weapon system 10 also has
an intended value sensor 10.5, using which the intended position of weapon
barrel 10.2 is detected.
Furthermore, Fig. 1 A shows a device 20 for performing the method according to
the present invention. Device 20 has a measurement facility 20.1 for detecting
the actual values, which describe the actual positions of weapon barrel 10.2
after aiming, and a computer unit 20.2. Intended value sensor 10.5 is
typically a
component of weapon system 10, but its functions may also be included in
device 20.
Fig. 1 B shows gun 10.1 of weapon system 10, having a lower gun carriage 12,
an upper gun carriage 14, and weapon barrel 10.2. Lower gun carriage 12 is
supported via three legs 12.1, 12.2, and 12..3 on a horizontal support surface
1.
In Fig. 1, the orthogonal axis system of the three axes is also shown, the
vertical axis being indicated with A, the lateral axis with L, and the
longitudinal
axis with R. Weapon barrel 10.2 may be rotated around vertical axis A to
change the lateral angle and/or azimuth a and may be rotated around lateral
axis L to change the vertical angle and/or elevation A.
An optical-electronic gyroscopic measurement system 22, which forms a
component of measurement facility 20.1, is positioned on weapon barrel 10.2 in
the muzzle region. A gyroscopic measurement system 22 includes a first
measurement unit and/or a-measurement unit and a second measurement unit
and/or A-measurement unit, using which angle changes resulting from changed
azimuth a and/or changed elevation A of weapon barrel 10.2 may be detected.
In the following, the procedure for cornpensating an azimuth synchronization
error Aa1 and for compensating a wobble error dT, which are detectable within
CA 02416166 2003-01-13
.18
a first measurement procedure, but in separate partial procedures, are
described. .
Figs. 2A to 2C relate to the partial procedure concerning azimuth
synchronization error Da1. In Fig. 2A, gun 10.1 is illustrated greatly
simplified in
a top view. Weapon barrel 10.2, illustrated in simplified form as a weapon
barrel axis, is indicated with solid lines in its zero position and with
dashed lines
in one of the measurement positions, which, with the zero position, encloses
an
angle of, for example, 20 . Starting from the zero position, weapon barrel
10.2
t0 is rotated a total of 180 into a final positiori in steps of, for example,
50 in the
direction of arrow Dl. The rotation of weapon barrel 10.2 is controlled by
fire
control computer 10.4. Each measurement position is determined by the
associated lateral angle and/or associated azimuth a. After each step, weapon
barrel 10.2 is theoretically in an intended position, which is defined by the
associated intended value and/or an associated intended azimuth al(theor),
which is displayed, for example, on gun 10.1. In reality, weapon barrel 10.2
is,
however, in an actual position, which is indicated by an actual value and/or
an
actual azimuth al(eff) detected by the a-measurement unit of gyroscopic
measurement system 22 of measurement facility 20.1. Computer unit 20.2
20 computes the error value and/or the error angle in each case, i.e., the
deviation
of actual value a1(eff) from intended value a1(theor). The error values are
then
illustrated, as a function of a1(theor), as first-directional empirical
azimuth error
curve fa1(D1),. The method steps described up to this point are repeated
multiple times in order to remove random errors in the detection of actual
azimuth and intended azimuth as much as possible. In this way, further first-
directional empirical azimuth error curves fa1(D1)2, fa1(D1)3, fa1(D1)i are
established. As shown in Fig. 2B, an average first-directional azimuth error
curve fa1(D1) finally results from all of the first-directional azimuth error
curves.
Subsequently, the method steps described above are performed again, weapon
30 barrel 10.2 being rotated in the opposite direction, i.e., in the direction
of arrow
D2. Multiple second-directional azimuth error curves fa1(D2)1, fa1(D2)2,
fal(D2)3 and an average second-directional empirical azimuth error curve
fal(D2) result from this, also shown in Fig. 2B. Next, an average direction-
free
CA 02416166 2003-01-13
19
empirical azimuth error curve fa1(D0) , which is also shown in Fig. 2B, is
calculated fr~om average first-directional empirical azimuth error curve
fa1(D1)
and average second-directional empirical azimuth error curve fal(D2). As
shown in Fig. 2B, average direction-free azimuth error curve fa1(D0), which
describes azimuth synchronization error Aa1, runs approximately in the shape
of a sine curve having a double angular frequency. This indicates that there
is a
slight ovality in the lateral pivot bearing.
In the nurriericai methods, average direction-free empirical azimuth error
curve
fa1(D0) and/or the value pairs which define this curve are made available to
the
fire control computer and/or system computer, in order to make them available
during further calculations of aiming values. The numerical methods may be
performed analogously for all measurement procedures.
In the mathematical methods, average direction-free empirical azimuth error
curve fa1(D0) is approximated by a mathematical azimuth error function Fa1.
The approximation is performed either by a mathematical partial error function
for each section, the totality of the partial error functions being referred
to as the
mathematical error function, or as a whole by one single mathematical error
function. Mathematical error function Fa1 is used to produce a correction
function, which is taken into consideration during calculation of the aiming
values, together with other available data. To check, after the implementation
of the correction function in the software of system computer 10.4, the method
steps described up to this point may be performed again; corrected azimuth
error curve fa1(DO)korr established in this way runs significantly flatter
than non-
corrected error curve fa1(D0); the original observable azimuth synchronization
error therefore may be reduced to a very small residual error and/or may be
compensated almost completely.
i0 The method steps described above may be performed partially in other
3
sequences, which influences the results insignificantly or not at all. In
particular,
it is time-saving to perform the measurerrients to establish the first-
directional
error function and the second-directional error function alternately.
CA 02416166 2003-01-13
In order to achieve more precise results, establishing direction-free azimuth
error curve fa1(D0) may be dispensed with; in place of this mathematical
azimuth error functions Fa1(D1) and Fa1(D2') are determined for first-
directional
empirical azimuth error curve fa1(D1) and second-directional empirical azimuth
error curve fal(D2), respectively, and the corresponding correction functions
are determined therefrom.
Figs. 3A to 3C relate to wobble error Or. Weapon barrel 10.2 is theoretically
to
10 be directed horizontally at an elevation of 0 , i.e., the intended
elevation must
be 0 . In reality, weapon barrel 10.2 will always have a slight inclination to
the
horizontal, i.e., the actual elevation is not 0 , but differs from 0 by Ar.
Angle AT
is a function of azimuth a. During a rotation through 360 around vertical
axis
A, weapon barrel 10.2 therefore performs a wobble motion, which is described
by a wobble error function. To detect wobble error Ar, weapon barrel 10.2 is
moved without elevation A in the same steps as for establishing azimuth
synchronization error Aa1. However, the effective inclination and/or wobble
angle of weapon barrel 10.2 is detected after each measurement step, which is
referred to as weapon barrel wobble angle T(eff). The theoretical inclination
20 and/or wobble angle, which is referred to as intended value and/or intended
wobble angle r(theor), is zero. Actual value and/or actual wobble angle T(eff)
may be represented in a function of azimuth a(theor). Now, an average first
directional and an average second directional empirical wobble error curve
fT(D1) and fT(D2), respectively, are determined analogously to the
establishment
of average empirical azimuth error curve fa(D1) and fa(D2). Finally, a
direction-
free empirical wobble error curve fT(D0) results therefrom, which is
approximated by a mathematical wobble error function FT. In Fig, 3A, the two
extreme wobble error curves of multiple established empirical wobble error
curves are illustrated, between which all other wobble error curves lie; the
measurements appear to be quite precise, since the curves only deviate
slightly
from one another; the wobble movement is a sinusoidal movement. An analysis
of the measurement data for the wobble niovement provides results which are
illustrated in Figs. 3B and 3C. As a consequence, the wobble error has two
CA 02416166 2003-01-13
21
causes: firstly, the azimuth-dependent rigidity of the lower gun carriage; the
cornponent of the wobble error resulting therefrom is illustrated in Fig. 3B;
secondly, the stiffening effect due to the legs, also azimuth-dependent, this
component of the wobble error being illustrated in Fig. 3C. In Figs. 3B and
3C,
the positive values of the wobble error are illustrated using solid lines and
the
negative values of the wobble error are illustrated using dashed lines.
Next, the compensation of elevation synchronization error L1A, which is
detected
in a second measurement procedure, will be described. Elevation
synchronization error AA comprises two error components. Both error
components are detectable using a second measurement unit and/or A-
measurement unit of gyroscopic measurement system 22 of measurement
facility 20.1, and only as their sum. Therefore, A refers to and/or indexes
data
and/or functions which relate to total elevation synchronization error AX. In
this
case, elevation A is understood to be the angle of inclination of weapon
barrel
10.2 to the horizontal assumed by weapon barrel 10.2 while keeping azimuth a
constant. Elevation A, starting from a horizontal position, i.e., from an
elevation
of 0 and also a perpendicular deviation of 0 , is changed in steps of, for
example, 5 up to a final position of, for example, 85 . The movement of
weapon barrel 10.2 is controlled by computer. After each step, weapon barrel
10.2 is in a measurement positiori. Iri this case, its elevation is
theoretically a
value which is referred to as an intended value and/or intended elevation
,&(theor) and which is indicated by intended value sensor 10.5. However,
weapon barrel 10.2 is in another position, which is described by actual value
and/or actual elevation J\(eff). As described above in regard to the azimuth
synchronization error, the difference between the A(theor) and A(eff) is
represented in a function of A(theor). The movement of weapon barrel 10.2 is
repeated multiple times in both rotational directions. An average first-
directional
empirical elevation error curve fA(D1) and an average second-directional
empirical elevation curve fA(D2) is obtained from the measurement results
recorded in this case. A direction-free elevation error curve fA(D0) results
therefrom, which is shown in Fig. 4A witti a solid line. It may be seen in
Fig. 4A
that with increasing elevation A, i.e., with continuously steeper positioning
of
CA 02416166 2003-01-13
22
weapon barrel 10.2, elevation error curve fA(DO) rises. Empirical elevation
error
curve fA(D0} is then approximated by a mathematical elevation error function
FA, and a correction function is determined which is taken into consideration
in
the calculation of the aiming values. If the measurements are repeated, but
taking the correction functions into consideration, the corrected elevation
error
function runs much more flatly than the uncorrected one.
The error components of elevation synchronization error AA which may not be
individually detected during the measurement may be established using a
mathematical analysis of mathematical elevation error function FA.
The first error component of the elevation synchronization error would, by
itself,
resuit in an error function which essentially corresponds to a sine function
having multiple angular frequencies.
By itself, the second error component of the elevation synchronization error
would result in an error function fA(D0)2, which essentially follows a cosine
function subtracted from 1, illustrated in Fig. 4A using a dashed line. This
corresponds to the fact that, with increasing elevation, the torque exercised
by
the weight of weapon barrel 10.2 on the gun carriage is reduced, because the
distance of the line of application of the weight of weapon barrel 10.2 from
lateral axis L is reduced; this torque tends to tilt gun 10.1 and therefore
weapon
barrel 10.2 forward; a reduction of this torque consequently has the effect
that
gun 10.1 having weapon barrel 10.2 tilts forward less and/or relatively tilts
backward.
The sum of the error components corresponds to elevation error curve fA(D0),
resulting from the measurements performed. This is represented as an
oscillation corresponding to the first error component around a rising curve
corresponding to the second component.
The measurements described above of elevation synchronization error AA of
the second measurement are performed with azimuth a kept constant. Multiple
CA 02416166 2003-01-13
23
further measurement series are then performed for further azimuths, with the
azimuth kept constant in each measurement series, the angular intervals
between the fixed azimuths able to be, for example, 5 . In this case as well,
two
measurement series are preferably performed for azirnuth, rotation being in a
first rotational direction for the first measurement series and in the
opposite
rotational direction for the second measurement series. Fig. 4B shows a
spatial
parameter illustration of elevation synchronization error AA as a function of
azimuth a, using various elevations A as a parameter, the bottom curve
corresponding to the smallest elevation.
The further steps for compensating the elevation synchronization error are
performed analogously to the compensation of the azimuth synchronization
error described above.
It is also to be noted that, as described above in regard to the compensation
of
the azimuth synchronization error, the individual measurement and analysis
procedures may be performed at least partially in different sequences, without
influencing the results.
Perpendicular offset error Aa2 is also established within the second
measurement procedure. For this purpose, in each of the measurement
positions in which, with the aid of the A measurement unit, elevation
synchronization error AA is determined, perpendicular offset error Aa2 is
determined with the aid of the a-measurement unit. Fig. 5 shows the
perpendicular offset error as a function of elevation A. Empirical
perpendicular
offset error curve fa2, shown with a dashed line, may be approximated by a
mathematical perpendicular offset error function Fa2, shown with a solid line,
for
example by a second order polynornial.
The detection and the compensation of perpendicular offset error Aa2 is
performed analogously to the compensation of azirriuth synchronization error
Aa1 described above.
CA 02416166 2003-01-13
24
Finally, a third measurement procedure is performed, with the aid of which a
compensation of squint error AQ is performed. Squint error OQ arises because
the directions of the weapon barrel axis and the line of vision of the gun are
not
coincidental, but rather enclose a squint angle. To establish the squint
error,
the extensions of the weapon barrel axis and the line of vision are displayed
at
a certain distance to the muzzle of the weapon barrel, for example using a
projection, the weapon barrel axis and the line of vision appearing as points.
The deviation of the two points is a measure of the squint error, the distance
between the weapon barrel and the projection surface also having to be
considered to establish this error. This method of establishing the squint
error
is not novel and is only described here for supplementary purposes, since
complete compensation of firing errors which are caused by static gun geometry
errors must also take squint error into consicleration.
While the description above predominantly relates to the method according to
the present invention, in the following, the device used to perform this
method is
described in more detail.
It is noted here again that the novel method is performed using the novel
device
on a weapon system 10 as shown in Fig. 1 A. Weapori system 10 has gun 10.1
having at least one weapon barrel 10.2, whose movements are controlled in a
typical way using gun servomotors. Furthermore, weapon system 10 has fire
control device 10.3. Weapon system 10 also has system computer and/or fire
control cornputer 10.4, which is positioned on fire control device 10.3 or, at
least
partially, on gun 10.1. Weapon system '10 typically also has an intended value
sensor 10.5, which indicates the intended values, particularly of azimuth a
and
elevation A, which describe the intended position of aimed weapon barrel 10.2
determined by system computer 10.4.
Multiple components are necessary to perform the novel method, which are
described in more detail in the following:
CA 02416166 2003-01-13
A first component is formed by intended value sensor 10.5, which is used for
the purpose= of indicating the intended values, which describe the intended
and/or supposed position of weapon barrel 10.2. The intended value sensor
present on weapon system 10 in any case is used as the intended value
sensor.
A second component of the novel device is formed by measurement facility
20.1, for detecting the actual values, which describe the actual position of
weapon barrel 10.2. Measurement facility 20.1 includes at least optical-
10 electronic gyroscopic measurement system 22.1, for example a fiber-optic
measurement system. Gyroscopic measurement system 22.1 has at least a
first and/or a-measurement unit for detecting changes of angle, preferably of
azimuth a, of weapon barrel 10.2. Preferably, gyroscopic measurement system
22.1 also has a second and/or X-rneasurernent unit for detecting changes of
elevation A of weapon barrel 10.2.
In the framework of the present invention, optical-electronic gyroscopic
measurement systems are to be understood to include not only fiber-optic
measurement systems, but also other measurement systems, for example ring
20 laser gyroscopic measurement systems. Gyroscopic measurement systems
generally have the advantage that they operate autonomously; therefore, no
reference points external to the system have to be used. Guns do not have to
be brought into a separate measurement station. However, because there is no
reference external to the system, the system generally drifts over time. The
gyroscopic drift manifesting in this case must be determined and taken into
consideration during the analysis of the measurement results. A laser
positioning system may be used in connection with this.
In order to detect the static gun geometry errors more completely and
therefore
to perform a more precise compensation of firing errors caused by them, the
second component of the novel device, i.e. measurement facility 20.1,
preferably also has measurement systems for detecting further errors,
particularly wobble error L1T and squint erro-- AQ.
CA 02416166 2003-01-13
26
In order to =detect wobble error Ar, in addition to gyroscopic measurement
system 22.1, a further measurement system 21.2 in the form of a typicai,
preferably electronic, spirit level is used. This level measures angles in
relation
to the horizontal, in the present exemplary embodiment, the respective angle
of
the weapon barrel axis to the horizontal. An electronic spirit level is
understood
as a sensor which measures the horizontal angle, i.e., the angle to a
horizontal,
and outputs an electric signal correlated to this angle. The measurement uses
effects of gravitation, which define the vertical and therefore also the
horizontal.
In this case, it is unimportant how the sensor uses gravitation.
It is also noted here that the tilt of gun 10.1 may be determined with the aid
of
an electronic spirit level. Tilt is understooci as the following: if weapon
barrel
10.2 is only moved in the azimuth, theri the movement of the muzzle of the
weapon barrel may be approximately considered as a circular line which defines
a plane. The angular deviation of this plane in relation to the horizontal
plane is
referred to as tilt; in other words, without tilt, this plane would be a
horizontal
plane. Generally, in new guns, the tilt is automatically compensated and/or
the
gun is automatically leveled to the horizontal. The leveling to the horizontal
of
the gun is, however, not necessary for performing the novel method.
In order to detect squint error Or, in addition to gyroscopic measurement
system
22.1 and electronic spirit level 22.2, a further measurement system 22.3 in
the
form of a typical, preferably optical, device is used. This device measures
the
angular difference between the weapori barrel axis and the line of vision of
gun
10.1.
A computer is required as a third component for performing the novel method.
The computer is implemented, as shown in Fig. 1A, as a separate computer unit
20.2, which is used exclusively to perform the novel method or also for other
purposes and is only coupled to the weapon system 10 for this purpose.
However, fire control computer and/or system computer 10.4 of weapon system
10 may also possibly be used as the computer.
CA 02416166 2003-01-13
27
The third component of the novel device, in the present case computer unit
20.2, has a data input and/or a data interface, via which it is supplied at
least
data which represent the detected interided values and actual values. The data
may be made available to computer unit 20.2 in any desired suitable way, for
example with the aid of a data carrier such as a diskette, or via a data
circuit,
which may be material or immaterial.
If fire control computer and/or system computer 10.4 is used as the computer,
then it already knows the intended values and the actual values are made
available to it via a data input and/or a data interface 24.
The third component of the novel device, in the present case computer unit
20.2, further has software implemented in order to determine the correction
values from the intended values and the actual values. The steps to be
performed in this case are described in rriore detail above in relation to the
method according to the present invention.
If fire control computer and/or system computer 10.4 is used as the computer,
the correction values established may be implemented directly in the fire
control
software.
If the fire control computer and/or systE:m computer is not used as the
computer, but rather separate computer unit 20.2, the established correction
values must be made available to fire control computer and/or system computer
10.4 via data input and/or data interface 24 and implemented in the fire
control
software on the computer.
The third component, i.e., the computer, preferably has an input unit 20.3,
such
as a keyboard, via which further data may be made available, particularly if
it is
formed by separate computer unit 20.2. This may include, for example, data
which controls the progress of the novel method, in ttiat it, among other
things,
controls the step-by-step rotation of the weapon barrel into the measurement
CA 02416166 2003-01-13
28
positions by the servomotors and the coupiing of the respective measurement
systems and/or measurement uriits to be used.
