 |
Description  |
|
|
FIELD OF THE INVENTION
The present invention relates to an optical fiber gyroscope which detects
an angular velocity based on an interference of light transferred through
an optical fiber loop, more particularly, the present invention relates to
a phase modulation type optical fiber gyroscope which is suitable for a
navigation system of automobiles.
FIELD OF THE INVENTION
An optical fiber gyroscope has been considered to be useful for a
navigation system of vehicles such as automobiles, since it is possible to
detect an angular velocity without a moving portion. As a kind of optical
fiber gyroscope, there is a phase modulation type fiber gyroscope.
A prior phase modulation type optical fiber gyroscope is introduced in
ELECTRONICS LETTERS 10th Nov. 1983 Vol. 19 No. 23 from pages 997 to 999 in
the title of "Direct Rotation-Rate Detection with a fiber-optic Gyroscope
by using digital processing". According to the prior art, the phase
modulation type optical fiber gyroscope realizes high performance by
converting an analog signal output from the optical fiber system into a
numerical signal, analyzing a wave thereof by a large scale numerical
calculator, and processing the calculated result into signals.
In a navigation system of an automobile, although an angular velocity has
to be detected in high accuracy, the navigation system of the prior art is
not considered on the scale of the numerical calculation portion. Then,
when the prior navigation system is demanded to be high accurate and to
obtain a quick response velocity, the numerical portion has to be
enlarged.
SUMMARY OF THE INVENTION
An object of the present invention is to provide a phase modulation type
optical fiber gyroscope of a small size and high performance without using
a large scale numerical calculating portion.
The object of the present invention is attained by classifying an output,
or an interference, signal of the optical fiber system into one group of
odd harmonic components and another group of even harmonic components of
phase modulation frequency of a phase modulator and calculating an output
angular velocity based on the odd harmonic components and even harmonic
components.
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 is a block diagram of one embodiment of a phase modulation type
optical fiber gyroscope of the present invention;
FIG. 2 is a block diagram of other embodiment of a phase modulation type
optical fiber gyroscope of the present invention;
FIG. 3 is a block diagram of another embodiment of a phase modulation type
optical fiber gyroscope of the present invention;
FIG. 4 is a block diagram of further another embodiment of a phase
modulation type optical fiber gyroscope of the present invention;
FIG. 5 is a time chart for explaining FIGS. 3 and 4;
FIG. 6A and FIG. 6B are waveform diagrams of synchronous pulse voltages and
spectra diagrams applied to synchronous detectors, respectively;
FIG. 7 is a block diagram of one embodiment of a synchronous detector; and
FIG. 8 is a block diagram of other embodiment of a synchronous detector.
DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
Referring to FIG. 1, an optical fiber system comprises a coherence light
source 1, couplers 2a, 2b, a polarizer 3, an optical fiber loop, and a
phase modulator 5. For instance, the coherence light source 1 is an
ordinary laser diode or a super luminescence diode. The couplers 2a, 2b
are beam splitter of an optical fiber having evanescent effect. The
polarizer 3 is a special wound optical fiber in a coil shape having
polarizing characteristic. The optical fiber loop 4 is an optical fiber
having a number of hundreds whole length in a coil shape. The phase
modulator 5 is formed by winding an optical fiber to an electrostriction
element in cylinder shape, and changing light transmitted length by
electric signal. The connection between parts and parts is formed by
fusion of each optical fiber. The optical fiber is, for instance,
polarization maintaining fiber.
The phase modulating type optical fiber gyroscope explained above shows a
most representative standard construction, and fitted construction for
practicing the present invention. Other portion shown in FIG. 1 is a
signal processing portion, and main portion of the present invention.
A photoelectric transfer portion 6 is used for converting light power to
voltage, and comprises mainly a photodiode and a current voltage
converter. Synchronous detector 7a or 7b comprises a non-inverting
amplifier 15, an inverting amplifier 16, a changeover switch 17, and a
low-pass filter 18. As apparent from explanation mentioned later, the
changeover switch 17, which is enclosed in each synchronous detector 7a,
or 7b having a signal gate for inputting the interference signal E.sub.11
from the photoelectric transfer portion 6, is changed over corresponding
to each synchronous pulse voltage, which is applied to the reference input
port of each synchronous detector 7a, or 7b, for obtaining the first
sensed, or added, value (hereunder, odd harmonic components) comprising
odd harmonic components including a fundamental component of the
modulation frequency from the synchronous detector 7a, and the second
sensed, or added, value (hereunder, even harmonic components) of even
harmonic components excluding a direct current component from the
synchronous detector 7b. Each odd harmonic component and even harmonic
component is smoothed by a low-pass filter 18 to obtain detecting output
E.sub.12 or E.sub.13.
The changeover switch 8 outputs a number of input signals E.sub.12 and
E.sub.13, which are obtained in the synchronous detectors 7a, and 7b by
weighting of the reference signals E.sub.7, and E.sub.8, selectively
corresponding to an instruction signal E.sub.4 from a numeral calculator
10. A numerical transfer portion 9 samples an analog value from the
numerical transfer portion 8 corresponding to an instruction signal
E.sub.5 from the numerical calculator 10. The numerical calculator 10
comprises, for instance, a microcomputer and an input and output
interface, outputs an instruction signal to each calculating or other
portion, and outputs a numeral output angular velocity .OMEGA..sub.out. An
oscillator 15 comprises a crystal oscillator. Dividers 12a, and 12b output
predetermined synchronous pulse voltages E.sub.7, and E.sub.8 by dividing
an output pulse E.sub.6 of the oscillator 15. The oscillator 15, the
divider 12a, and the oscillator 15, the divider 12b constitute reference
signal generators to the synchronous detectors 7a, and 7b for outputting
the reference inputs to the synchronous detectors 7a, and 7b,
respectively. A low-pass filter 13 removes all high harmonic components of
the synchronous pulse voltage E.sub.7 and produces an alternating voltage
E.sub.9 of a, sinusoidal wave comprising only a fundamental wave component
thereof. A voltage gain amplifier of controls an amplitude 14 an
alternating voltage E.sub.9 from the low-pass filter 13 according to a
control signal E.sub.14 from the numerical calculator 10, and outputs a
suitable driving voltage E.sub.10 to the phase modulator 5.
In the above explained constitution of the phase moduration type fiber
optic gyroscope, when the frequency and amplitude of the modulation signal
E.sub.10 are settled to arbitrary values based on the whole length of the
optical fiber loop 4, the interference signal, or the distorted
alternating output E.sub.11 out of the photoelectric transfer portion 6
includes every kind of frequency components as follows:
##EQU1##
Wherein, A.sub.1 to A.sub.6 are amplitudes of each frequency components,
K.sub.p constant relating to light power and photoelectric transfer
efficiency, K.sub.S constant of optical system sensitivity using the
Sagnac-phase shift, .OMEGA..sub.in input angular velocity, Km constant of
phase modulation factor, J.sub.1 (Km) to J.sub.6 (Km) are constant
obtained by Bessel function, .omega. frequency of the modulation signal
E.sub.10 or angular velocity of the modulation frequency, t time, and
.tau. necessary time for passing the light wave to the optical fiber loop
4.
Hereunder, we will explain about formulae (1) to (6). The formula (1)
denotes the first harmonic component of the modulation frequency, the
formula (2) the second harmonic component of the modulation frequency. The
formulae (3), (4), (5), and (6) denote the 3rd, 4th, 5th, and 6th
modulation frequency components, respectively. In fact, the interference
signal E.sub.11 from the photoelectric transfer portion 6 can be denoted
by infinite number of formulae. Suppose that the value of Km is 3. The
absolute value of J.sub.n (Km) obtained by Bessel function becomes small
when the degree of the modulation frequency becomes higher. For instance,
the 10th harmonic component of the modulation signal becomes one-twenty or
thirty thousandth of the first harmonic component of the modulation
signal. However, the high harmonic components of the modulation signal are
useful information source in the phase modulation type optical fiber
gyroscope which is demanded to be high accurate.
Referring to these formulae explained above, the input angular velocity
.OMEGA..sub.in is included in all the formulae. All the frequency
components of the interference signal E.sub.11 are information source.
When all the frequency components are used for calculating the input
angular velocity, the most accurate input angular velocity will be
obtained. Referring to the formulae in detail, it is understand that each
amplitude of the odd harmonic components including the fundamental
component of the modulation signal in formula (1), (3), or (5) is
proportional to sin K.sub.S .OMEGA..sub.in, each amplitude of the even
harmonic components of the modulation frequency applied to the phase
modulator 5 in formular (2), (4), or (6) is proportional to cos K.sub.S
.OMEGA..sub.in, and each phase in the formulae (1) to (6) is fixed to
(.omega.t-.tau./2).
This means that by classifying the frequency components of the interference
signal E.sub.11 into the first component group which is proportional to
sin K.sub.S .OMEGA..sub.in and the second component group which is
proportional to cos K.sub.S .OMEGA..sub.in, and detecting synchronously
the first and second component groups, respectively; the first component
group comprising even harmonic components including the fundamental
component of the modulation frequency and the second component group
comprising even harmonic components are obtained as the mixture signals,
respectively. This is the gist of the present invention.
Hereunder, we will explain the means for detecting each mixture signal of
the odd harmonic components and the even harmonic components of the
modulation frequency. The synchronous detector 7a is a portion for
detecting the first mixture signal of the even harmonic components
including the fundamental component of the modulation frequency. The
interference signal E.sub.11 is detected synchronously, smoothed, and
obtained the detected output signal E.sub.12 as the mixture signal of the
even harmonic components corresponding to the synchronous pulse voltage
E.sub.7 comprising even harmonic components including the fundamental
component as shown in FIG. 6A, by forming the waveform of the synchronous
pulse voltage E.sub.7 into a rectangular wave having the relation of
P.sub.1 =T.sub.1 /2. Wherein, P.sub.1 is a time width of a low level or a
high level of the synchronous pulse voltage E.sub.7, T.sub.1 a period of
the synchronous pulse voltage E.sub.7.
On the other hand, the synchronous detector 7b is a portion for detecting
the second mixture signal of the odd harmonic components. The waveform of
the synchronous pulse voltage E.sub.8 is formed to the rectangular wave
having the relation of P.sub.2 =0.2-0.25T.sub.2, and T.sub.2 =2 as shown
in FIG. 6B. By forming the waveform of the synchronous pulse voltage
E.sub.8 as explained above, each frequency component included in the
synchronous voltage E.sub.8 becomes even harmonic components of the
modulation frequency as shown in the parentheses of FIG. 6B, so that the
detected output signal E.sub.13 becomes the mixture signal of the odd
harmonic components.
Incidentally, the gain of the synchronous detection at each frequency does
not become to 1, but becomes corresponding value of the amplitude of each
frequency component of the synchronous pulse voltages E.sub.7 and E.sub.8.
For instance, the amplitude of each frequency component of the synchronous
pulse voltage E.sub.7 becomes as follows: When the amplitude of the first
harmonic component is one, the amplitude of the Nth harmonic component
becomes 1/N.
Accordingly, the gain of the synchronous detection at each frequency
becomes 1/N, when the frequency is the Nth harmonic. The interference
signal E.sub.11 is detected by these gain, and the detected output signal
E.sub.12 is obtained as the mixture value of the odd harmonic component
including the fundamental component. On the other hand, the amplitude of
each frequency component of the synchronous pulse voltage E.sub.8 changes
as well as that of the synchronous pulse voltage E.sub.7 in such a manner
that the amplitude becomes smaller when the frequency becomes higher,
although the magnitude of the amplifier is not expressed simply since the
amplitude is changed according to the pulse width shown in FIG. 6B.
Although the gain of the synchronous detection is changed according to the
magnitude of the frequency, there is no problem since the gain is
stabilized.
Hereunder, we will explain the operation of the synchronous detectors 7a,
and 7b. The output signal E.sub.15 of the noninverting amplifier 20 and
the output signal E.sub.16 of the inverting amplifier 16 are changed over
by the changeover switch 17 in synchronous with the synchronous voltages
E.sub.7, and E.sub.8 of the rectangular waves as shown in FIG. 7. The
operation of obtaining the detected outputs E.sub.12, and E.sub.13 from
the input voltage E.sub.11, is same to the operation of multiplying, or
weighting by, the input voltage E.sub.11 and the rectangular wave pulse
voltage having plus and minus voltages which are removed a direct current
component from the synchronous pulse voltages E.sub.7 and E.sub.8, and
obtaining the multiplied voltage using analog multipliers. The frequency
spectrum obtained by Fourier series expansion of the rectangular wave
pulse voltage having plus and minus values and excluding a direct current
component of the synchronous voltage, or the reference input, E.sub.7 or
E.sub.8, is shown in right-hand of FIG. 6A, or 6B in which zero-order is
removed. The multiplication by the analog multiplier explained above is
carried out by multiplying two frequency components excluding a direct
current component of frequency components included in the interference
signal E.sub.11 and the synchronous pulse voltages E.sub.7, and E.sub.8.
According to trigonometric function, the multiplication result is able to
show as follows:
##EQU2##
Wherein, K.sub.1, and K.sub.2 are amplitudes, .omega..sub.1, and
.omega..sub.2 angular velocities, t time, and .phi..sub.1, and .phi..sub.2
phases.
Namely, when one of frequency components included in the interference
signal coincides with one of frequency components included in the
synchronous pulse voltage E.sub.7, or E.sub.8, the relation between
.omega..sub.1 and .omega..sub.2 becomes as follows:
.omega..sub.1 -.omega..sub.2 =0 --- (8)
.omega..sub.1 +.omega..sub.2 =2.omega..sub.1 or 2.omega..sub.2 --- (9)
When .phi..sub.1 is equal to .phi..sub.2, or .phi..sub.1 and .phi..sub.2
are same phase, a direct current component whose amplitude is (K.sub.1
K.sub.2 /2, and an alternating component in which the amplitude is K.sub.1
K.sub.2 /2 and the angular velocity is 2.omega..sub.1 or 2.omega..sub.2,
are obtained as the detecting output signal E.sub.17.
When there are many coincident frequency components in the multiplication
by the analog multiplier, each detecting output signal E.sub.17 concerning
each frequency component is same to the case in which an unitary frequency
component included in the interference signal E.sub.11 and the synchronous
pulse voltage E.sub.7, or E.sub.8 is same to each other. The mixture value
of the direct current components, in the case that there are many
coincident frequency components in the multiplication explained above,
becomes to the summed value of each direct current component in each
frequency component. However, when .phi..sub.1 is different from
.phi..sub.2 and the relation between .phi..sub.1 and .phi..sub.2 varies in
each frequency component, the amplitude in each frequency component is not
the term of (K.sub.1 K.sub.2 /2) since the term of .phi..sub.1
-.phi..sub.2 is not always equal to zero.
Above explanation is in the case that the analog multiplier is used for the
synchronous detector. The operation of the synchronous detectors shown in
FIG. 7 is same to that of the analog multiplier. The detecting outputs
E.sub.12, and E.sub.13 which are similar to direct currents, are obtained
by smoothing the detected output signal E.sub.17 by the low-pass filter 18
and removing all alternative components.
The detected outputs E.sub.12, and E.sub.13 are input to the numerical
calculator 10 as the output signal E.sub.20 through the changeover switch
8 and the numerical transfer portion 9. The output angular velocity
.OMEGA..sub.out corresponding to the input angular velocity .OMEGA..sub.in
is calculated in high accuracy by calculating according to the following
formula (10) at the numerical calculator 10.
##EQU3##
Wherein, K.sub.x, K.sub.y are compensation coefficients.
According to the embodiment explained above, almost all the signal
procedures are taken place at the synchronous detectors 7a, 7b, so that
the numerical calculator 10 is able to be in a small scale. Since a
rectangular wave or rectangular waves are used for the synchronous signals
or the reference inputs, high performance filters are not necessary for
rectifying the waveform of the synchronous signals. However, the low-pass
filter 13 has to be in high performance for obtaining the modulation
signal E.sub.10 to the phase modulator 5. Since no multiplier is used,
there is no zero point fluctuation and no gain fluctuation. All the signal
processing can be carried out by ordinal integrated circuits, so that the
cost of the fiber gyroscope of the present invention can be decreased, and
the response speed of the signal processing can be fast.
Although the embodiment shown in FIG. 7 constitutes the synchronous
detector 7a, or 7b by an noninverting amplifier 15, an inverting amplifier
16, and a changeover switch 17 for obtaining full wave rectification, the
synchronous detector 7a, or 7b is not limited to the embodiment. For
instance, as shown in FIG. 8, the synchronous detector 7a, or 7b can be
formed by a half wave rectifier type synchronous detector in such a manner
that the interference signal E.sub.11 is applied directly to one input
terminal of the changeover switch 17 and other input terminal thereof is
earthed. The embodiment shown in FIG. 8 is formed by a simple circuit.
The synchronous detectors 7a, and 7b shown in FIG. 1 can be replaced by
multipliers. However, the effect of the present invention is not lost by
such a embodiment.
In the embodiment shown in FIG. 1, the interference signal E.sub.11 is
applied directly to the synchronous detectors 7a, and 7b. However,
unnecessary frequency components, for example a direct current component
or high harmonic components, can be removed by inserting a proper filter
for obtaining high accuracy.
Referring to FIG. 3, a synchronous detector 7C is added for sensing even
harmonic components larger than the third harmonic or sensing odd harmonic
components larger than the 4th harmonic from the interference signal
E.sub.11. For obtaining even harmonic components larger than the third
harmonic, the divider 12C outputs E.sub.18 to the reference input port of
the synchronous detector 7C. In this case, the divider 12C outputs twice
frequency of the frequency of the divider 12a.
When the synchronous detector 7C is used for sensing odd harmonic
components larger than the 4th harmonic from the interference signal
E.sub.11, the divider 12C outputs E.sub.18 to the reference input port of
the synchronous detector 7C. In this case, the divider 12C outputs twice
frequency of the frequency of the divider 12b.
The output signal E.sub.19 of the synchronous detector 7C is applied to the
numerical calculator 10 through the changeover switch 8 and the numerical
transfer portion 9.
In the embodiment shown in FIG. 3, the synchronous detector 7C is used for
sensing even harmonic components larger than the third harmonic or sensing
odd harmonic components larger than the 4th harmonic component of the
modulation frequency of the phase modulator 5. However, the present
invention is not limited to the embodiment shown in FIG. 3 in which one
synchronous detector is disclosed. Namely, more than two synchronous
detectors can be used instead of the single synchronous detector 7C for
obtaining the synchronous detectors which are able to sense even harmonic
components larger than the third harmonic and sense odd harmonic
components larger than the 4th harmonic component from the interference
signal E.sub.11.
According to the present invention, the interference signal E.sub.11 is
classified into odd and even groups as already explained above. A ratio
signal E.sub.14 among the output signals from the synchronous detectors in
odd group or among the output signals from the synchronous detectors in
the even group is calculated at the numerical calculator 10. The ratio
signal E.sub.14 is used for controlling the modulation signal of the phase
modulator 5. By controlling the ratio signal of E.sub.19 /E.sub.13 to be
constant by the voltage gain amplifier 14, the modulation signal E.sub.10
is controlled so that the variation of the modulation operation by the
phase modulator 5 is controlled to be small and finally the fluctuation of
the output angular velocity .OMEGA..sub.out can be small and in high
accuracy.
In the embodiment shown in FIG. 3, phase shifters 30a, 30b, and 30c are
connected between the dividers 12a, 12b, 12c, and the synchronous
detectors 7a, 7 b, 7c, respectively.
Although the first, second, and 4th harmonics are shown in sinusoidal
waves, and the synchronous pulse voltages, E.sub.7, E.sub.8, and E.sub.18
are shown in rectangular waves; these phase shifters are used for
adjusting the mutual phases among the modulation signal and the reference
signals in such a manner, for instance, f.sub.E18-1, f.sub.E18-2,
f.sub.E18-3 are coincided to f.sub.4-1, f.sub.4-2, f.sub.4-3,
respectively.
Referring to FIG. 4, the fiber optic gyroscope of FIG. 4 is constituted to
be similar to that of FIG. 3. However, in the embodiment of FIG. 4, the
phase shifter 30a' is connected between the divider 12a and the lowpass
filter 13 in place of 30a shown in FIG. 3. The phase shifter 30a' is used
for a variable phase shifter adjusting the mutual phase between the
modulation signal and one of the reference signals. The phase shifters
30b, and 30c are used for adjusting the mutual phases among the modulation
signal and the reference signals.
According to FIG. 4, the phase adjustment is easily carried out by only
adjusting the phase of the phase shifter 30a', since the phase shifters
30b, and 30c are adjusted their phases beforehand.
Referring to FIG. 2, a synchronous detector 7 is used instead of the
synchronous detectors 7a, and 7b shown in FIG. 1. The variable divider 11
outputs the synchronous pulse voltage E.sub.1 of the rectangular waves
shown in FIGS. 6A and 6B by turns and sequentially by receiving changeover
instruction signals E.sub.21 of the rectangular waves from the numerical
calculator 10. The synchronous detector 7 outputs the signal E.sub.3 to
the numerical calculator 10 through the numerical transfer portion 9.
The interference signal classified into odd and even groups in time sharing
by the instruction of the reference signal E.sub.1 is stored into memories
(not shown) as data storage elements of the numerical calculator 10. The
memories store the signals of the odd and even groups, respectively. The
analog velocity .OMEGA..sub.out is calculated based on the signals of the
odd and even groups.
According to the embodiment shown in FIG. 2, the fiber optic gyroscope is
formed in more small size compared with other embodiments explained above.
* * * * *
|
|
 |
|
 |
Description |
|
