Distance measuring device

Abstract

This ranging device comprises an object interferometer (2) and a signal processing device (5). The object interferometer (2) outputs a first interference signal resulting from light reflected by an object (100) being measured. The signal processing device (5) extracts a second interference signal resulting from light reflected by a reflecting surface in the object interferometer (2), from the first interference signal by means of a BPF (51), and calculates the distance to the object (100) being measured on the basis of the frequency of the peak of a signal obtained by resampling the first interference signal in synchronization with a resampling time, which is calculated on the basis of the phase change curve of the second interference signal, and the frequency of the peak of a signal obtained by resampling the second interference signal in synchronization with the resampling time.

Description

Technical Field

[0001] The present invention relates to a distance measuring device using the FMCW method.

Background Art

[0002] Non-Patent Documents 1, 2, and 3 disclose a distance measuring device using the FMCW LiDAR (Frequency Modulated Continuous Wave Light Detection and Ranging / Laser Imaging Detection and Ranging) method or the SS-OCT (Swept Source Optical Coherence Tomography) method equivalent to FMCW LiDAR.

[0003] In the methods disclosed in Non-Patent Documents 1 and 2, a reference interferometer for obtaining a reference interference signal used to calculate a resampling timing signal used for resizing processing is provided separately from the target interferometer whose reflection surface of the measurement object is included in the arm (optical path). Then, the reference interference signal and the target interference signal obtained from the target interferometer are simultaneously obtained by separate AD (Analog-to-digital) converters, and the target interference signal is resampled using the resampling timing data calculated from the reference interference signal, that is, resizing processing is performed.

[0004] Further, Non-Patent Document 2 describes resizing the target interference signal obtained using the same AD converter using the resampling timing data calculated using the reference interference signal obtained in advance using the target interferometer and the AD converter.

[0005] In the methods disclosed in Non-Patent Documents 1 and 2, due to the fluctuations (jitter) in each sampling timing of the AD converter when acquiring the reference interference signal and the fluctuations in each sampling timing of the AD converter when acquiring the target interference signal being different for each sampling timing, and the overall time shift (skew) between channels, there was a problem that the resampling timing of the target interference signal deviated from the original timing.

[0006] Due to this problem, temporal frequency fluctuations occur in the target interference signal after the rescaling process. Therefore, the peak position of the point spread function (PSF) obtained by performing a Fourier transform on the target interference signal after the rescaling process (corresponding to the distance from the LiDAR or OCT to the light reflection position of the measurement target) fluctuates for each measurement, and the measured distance fluctuates. Thus, in the conventional method, there was a problem that the accuracy (repeatability) of the measured distance was poor (the standard deviation value of the measured value was large).

Prior Art Documents

Non-Patent Documents

[0007]

Non-Patent Document 1

Non-Patent Document 2

[0008] The present invention has been made to solve the above problems, and an object thereof is to provide a distance measuring device capable of realizing highly accurate distance measurement that is not affected by the jitter of an AD converter. [Means for Solving the Problems]

[0009] The distance measuring device of the present invention includes a first interferometer configured to convert first interference light, which is obtained by interfering continuous light output from a light source with its wavelength swept in time and reflected light obtained by reflecting the light output from the light source by a measurement target, into an electrical signal and output a first interference signal, and a signal processing device configured to calculate the distance to the measurement target in the first interferometer. The first interferometer includes a reflecting surface in the optical path of the reflected light. The signal processing device extracts, from a second interference signal obtained from second interference light, which is obtained by interfering the light output from the light source with the reflected light obtained by reflecting the light output from the light source by the reflecting surface, from the first interference signal by a frequency filter, resamples the first interference signal in synchronization with a resampling time calculated based on a phase change curve of the second interference signal, and then performs discrete Fourier transform. Based on the frequency of the peak of the first signal and the frequency of the peak of the second signal obtained by resampling the second interference signal in synchronization with the resampling time and then performing discrete Fourier transform, the distance to the measurement target in the first interferometer is calculated.

[0010] In addition, one configuration example of the distance measuring device of the present invention further includes a second interferometer configured to convert third interference light, which is obtained by interfering the light output from the light source with light having a predetermined optical path length difference from the light output from the light source, into an electrical signal and output a third interference signal. The signal processing device includes the frequency filter and a time-frequency characteristic calculation unit configured to calculate an estimated curve showing the time transition of the frequency of the second interference signal from the third interference signal. The frequency filter is characterized in that a high cut-off frequency and a low cut-off frequency change in proportion to the estimated curve calculated by the time-frequency characteristic calculation unit. Also, in one configuration example of the distance measuring device of the present invention, the signal processing device includes the frequency filter, and a time-frequency characteristic calculation unit configured to calculate an estimation curve showing the time transition of the frequency of the second interference signal from the interference signal output from the first interferometer in a state where the light incident from the light source to the measurement target is blocked before distance measurement. The frequency filter is characterized in that the high cut-off frequency and the low cut-off frequency change in proportion to the estimation curve calculated by the time-frequency characteristic calculation unit.

[0011] Also, in one configuration example of the distance measuring device of the present invention, the signal processing device includes a coefficient calculation unit configured to calculate a coefficient determined by the measurable distance range. The high cut-off frequency and the low cut-off frequency of the frequency filter are determined by the estimation curve and the coefficient. Also, in one configuration example of the distance measuring device of the present invention, the signal processing device includes a phase change curve calculation unit configured to calculate a phase change curve showing the time transition of the phase of the second interference signal, a resampling time calculation unit configured to calculate the resampling time based on the phase change curve and a predetermined phase interval, a first resampling unit configured to resample the first interference signal in synchronization with the resampling time, a second resampling unit configured to resample the second interference signal in synchronization with the resampling time, a first peak position frequency calculation unit configured to detect the frequency of the peak of the first signal obtained by performing discrete Fourier transform on the resampled first interference signal, a second peak position frequency calculation unit configured to detect the frequency of the peak of the second signal obtained by performing discrete Fourier transform on the resampled second interference signal, and a distance calculation unit configured to calculate the distance to the measurement target in the first interferometer based on the frequency of the peak of the discrete Fourier-transformed first signal and the frequency of the peak of the discrete Fourier-transformed second signal.

[0012] Also, in one configuration example of the distance measuring device of the present invention, the first interferometer includes a first coupler configured to divide the light from the light source into two, and one of the lights branched by the first coupler and the other light branched by the first coupler and reflected by the measurement target are combined. A second coupler configured to combine the light, a circulator configured to output the other light branched by the first coupler to an optical fiber and output the light from the optical fiber to the second coupler, and irradiate the measurement target with the light from the optical fiber, and reflect the reflected light from the measurement target. A fiber collimator configured to be incident on the optical fiber, and the reflecting surface is an end surface of the optical fiber connected to the fiber collimator. Also, in one configuration example of the distance measuring device of the present invention, the reflectance of the reflecting surface is 1 / 3.

Effects of the Invention

[0013] According to the present invention, a reflecting surface is provided in the middle of the optical path of the reflected light of the first interferometer, and the light output from the light source and the reflected light obtained by reflecting the light output from the light source by the reflecting surface are interfered to obtain a second interference signal. The second interference signal is extracted from the first interference signal by a frequency filter, and the first interference signal is resampled in synchronization with the resampling time calculated based on the phase change curve of the second interference signal and then discretely Fourier-transformed. The distance to the measurement target is calculated based on the frequency of the peak of the first signal and the frequency of the peak of the second signal obtained by resampling the second interference signal in synchronization with the resampling time, thereby realizing high-precision distance measurement that is not affected by the jitter of the AD converter.

Brief Description of the Drawings

[0014]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Figure 7

Figure 8

Figure 9

Figure 10

Figure 11

Figure 12

Figure 13

Figure 14

Figure 15

Figure 16

Figure 17

Figure 18

Figure 19

Figure 20

Figure 21

Figure 22

Figure 23

Figure 24

Figure 25

Figure 26

DETAILED DESCRIPTION OF THE INVENTION

[0015] [First Embodiment] Hereinafter, embodiments of the present invention will be described with reference to the drawings. In the present invention, a reference interference signal is extracted from a target interference signal generated from a target interferometer by a frequency band-pass filter described later, and signal processing of FMCW LiDAR described in Non-Patent Documents 1 and 3 is performed using the target interference signal and the extracted reference interference signal. In order to realize such a thing, a reflection point for generating a reference interference signal is provided in the target interferometer separately from the reflection point on the measurement target surface. Hereinafter, specific contents of the present invention will be described.

[0016] Figure 1 is a block diagram showing the configuration of a distance measuring device according to a first embodiment of the present invention. The distance measuring device includes a wavelength sweeping light source 1, a coupler C1, a target interferometer 2 (first interferometer), a reference interferometer 3 (second interferometer), an AD converter (ADC) 4, and a signal processing device 5. Reference numeral 6 in Figure 1 is an optical fiber connecting the wavelength sweeping light source 1 and the coupler C1.

[0017] The light output from the wavelength sweeping light source 1 is split by the coupler C1. One of the lights enters the target interferometer 2, and the other light enters the reference interferometer 3.

[0018] The target interferometer 2 is an interferometer for measuring the distance to the measurement target 100. The configuration of the target interferometer 2 is shown in Figure 2. The target interferometer 2 includes couplers C2, C3, a circulator 20, a fiber collimator (FC) 21, a balanced photodetector (BPD) 22, an optical fiber 23 connecting the coupler C1 and the coupler C2, an optical fiber 24 connecting the coupler C2 and the coupler C3, an optical fiber 25 connecting the coupler C2 and the circulator 20, an optical fiber 26 connecting the circulator 20 and the FC21, an optical fiber 27 connecting the circulator 20 and the coupler C3, and optical fibers 28, 29 connecting the coupler C3 and the BPD22.

[0019] The coupler C2 splits the light from the coupler C1 into two. The circulator 20 outputs the light from the optical fiber 25 to the optical fiber 26 and outputs the light from the optical fiber 26 to the optical fiber 27. The FC21 makes the light from the optical fiber 26 parallel and irradiates the measurement target 100, and makes the reflected light from the measurement target 100 enter the optical fiber 26.

[0020] In this embodiment, among the optical paths between coupler C2 and coupler C3, the optical path passing through optical fiber 24 (the optical path not including the reflecting surface 101 of the measurement target 100) is defined as a. Also, among the optical paths between coupler C2 and coupler C3, the optical path following the path of optical fiber 25 → circulator 20 → optical fiber 26 → reflecting surface 102 existing at the output end of optical fiber 26 → optical fiber 26 → circulator 20 → optical fiber 27 is defined as b. Further, among the optical paths between coupler C2 and coupler C3, the optical path following the path of optical fiber 25 → circulator 20 → optical fiber 26 → FC21 → reflecting surface 101 of measurement target 100 → FC21 → optical fiber 26 → circulator 20 → optical fiber 27 is defined as c.

[0021] The reference surface 103 of the object interferometer 2 is a virtual surface, and it is a surface that has the same optical path length as the optical path a when light is reflected by this surface. The feature of this embodiment is that the optical path b is actively used. Conventionally, the reflected light from the end face of the optical fiber 26 connected to FC21 was unnecessary, but in this embodiment, the interference signals derived from the light passing through the optical paths b and c respectively are actively used as reference signals for creating resampling timing.

[0022] Let the distance from the reference surface 103 to the output end (reflecting surface 102) of the optical fiber 26 on the FC21 side be z S,FC , and let the distance from the reference surface 103 to the reflecting surface 101 of the measurement target 100 be z S , and let the distance from the output end (reflecting surface 102) of the optical fiber 26 to the reflecting surface 101 of the measurement target 100 be z S,FC,tgt . Among these, z S,FC,tgt is determined at the time of constructing the optical system and does not change unless the optical system is changed.

[0023] In FIG. 2, the reference surface 103 is described as being on the optical fiber 26 between the circulator 20 and FC21, but it is desirable to determine the position of the reference surface 103 such that |z S,FC |, |z S |, and |z S,FC,tgt | have significantly different values from each other. For example, |z S,FC |:|zS |:|z S,FC,tgt |=2:1:3, or|z S,FC |:|z S |:|z S,FC,tgt |=1:2:3 is desirable.

[0024] Coupler C3 multiplexes the light from optical fiber 24 and the light from optical fiber 27. BPD 22 photoelectrically converts the two outputs from coupler C3 respectively, and obtains and outputs the difference between the two analog electrical signals obtained as a result of the photoelectric conversion. At this time, from BPD 22, an analog electrical signal in which three electrical interference signals, namely, the electrical interference signal by the light passing through optical path a and the light passing through optical path b, the electrical interference signal by the light passing through optical path a and the light passing through optical path c, and the electrical interference signal by the light passing through optical path b and the light passing through optical path c, are superimposed is output.

[0025] The configuration of reference interferometer 3 is shown in FIG. 3. Reference interferometer 3 is composed of couplers C4, C5, circulator 30, FC 31, BPD 32, optical fiber 33 connecting coupler C1 and coupler C4, optical fiber 34 connecting coupler C4 and coupler C5, optical fiber 35 connecting coupler C4 and circulator 30, optical fiber 36 connecting circulator 30 and FC 31, optical fiber 37 connecting circulator 30 and coupler C5, and optical fibers 38, 39 connecting coupler C5 and BPD 32.

[0026] Reference interferometer 3 replaces measurement target 100 of target interferometer 2 with mirror 200, and has made the reflectance of the emission end of optical fiber 36 connected to FC 31 substantially zero. For example, if the emission end on the FC 31 side of optical fiber 36 is APC (Angled Physical Contact) polished and an AR (Anti Reflection) coat is formed on the emission end, the reflectance of the emission end becomes 0.1% or less. Let the distance from reference plane 203 to reflection surface 201 of mirror 200 be z R be.

[0027] Coupler C4 divides the light from coupler C1 into two. Circulator 30 outputs the light from optical fiber 35 to optical fiber 36, and outputs the light from optical fiber 36 to optical fiber 37. FC31 collimates the light from optical fiber 36 and irradiates mirror 200 with it, and makes the reflected light from mirror 200 enter optical fiber 36.

[0028] Similar to FIG. 2, among the optical paths between coupler C4 and coupler C5, let the optical path passing through optical fiber 34 be a. Also, among the optical paths between coupler C4 and coupler C5, let the optical path following the path of optical fiber 35 → circulator 30 → optical fiber 36 → output end of optical fiber 36 → optical fiber 36 → circulator 30 → optical fiber 37 be b. Also, among the optical paths between coupler C4 and coupler C5, let the optical path following the path of optical fiber 35 → circulator 30 → optical fiber 36 → FC31 → reflecting surface 201 of mirror 200 → FC31 → optical fiber 36 → circulator 30 → optical fiber 37 be c.

[0029] Coupler C5 multiplexes the light from optical fiber 34 and the light from optical fiber 37. BPD32 photoelectrically converts each of the two outputs from coupler C5, and obtains and outputs the difference between the two analog electrical signals obtained as a result of the photoelectric conversion.

[0030] In reference interferometer 3, the intensity of the light passing through optical paths a and c is greater than the intensity of the light passing through optical path b. Therefore, in the analog electrical signal output from BPD32, the interference signal due to the light passing through optical path a and the light passing through optical path c is large. The interference signal due to the light passing through optical path a and the light passing through optical path b, and the interference signal due to the light passing through optical path b and the light passing through optical path c are of negligible magnitude.

[0031] ADC4 performs AD conversion on the electrical signal input to the first channel from BPD22 of target interferometer 2 and the electrical signal input to the second channel from BPD32 of reference interferometer 3, respectively.

[0032] The signal processing device 5 processes the digital signal output from the ADC 4. The signal processing device 5 includes a time-frequency characteristic calculation unit 50, a band-pass filter (BPF) 51 that is a frequency filter, a coefficient calculation unit 52, and a ranging unit 53.

[0033] In the following description, the interference signal obtained from the target interferometer 2 through the ADC 4 is denoted as i S,source (t m ), and the interference signal obtained from the reference interferometer 3 through the ADC 4 is denoted as i R,source (t m ). Also, among the interference signals included in i S,source (t m ), the interference signal due to the light passing through the optical path a and the light passing through the optical path b is denoted as i S,source,z _ s _ FC (t m ), the interference signal due to the light passing through the optical path a and the light passing through the optical path c is denoted as i S,source,z _ s (t m ), and the interference signal due to the light passing through the optical path b and the light passing through the optical path c is denoted as i S,source,z _ s _ FC _ tgt (t m ). Note that t m represents time. m is an integer and is a suffix representing the discrete time of the AD-converted signal.

[0034] The time-frequency characteristic calculation unit 50 calculates a time-frequency curve showing the time evolution of the frequency of the signal included in the interference signal i R,source (t m ) obtained from the reference interferometer 3 through the ADC 4. The time-frequency characteristic calculation unit 50 includes a phase change curve calculation unit 500, a frequency calculation unit 501, and a frequency correction unit 502 as shown in FIG. 4.

[0035] The phase change curve calculation unit 500 calculates the phase change curve of the interference signal i R,source (t mCalculate a phase change curve showing the time evolution of the phase of (). The phase change curve calculation unit 500 is composed of a Fourier transform unit 5000, a negative frequency component zero unit 5001, an inverse Fourier transform unit 5002, an argument calculation unit 5003, and a phase connection unit 5004 as shown in FIG. 5.

[0036] The Fourier transform unit 5000 performs a discrete Fourier transform on the interference signal i R,source (t m ) and outputs the signal i R,source (f u ). f u represents the frequency. u is an integer and a suffix representing a discrete frequency. The negative frequency component zero unit 5001 performs a process of setting the negative frequency component of the signal i R,source (f u ) to zero, and outputs the signal I R,source + (f u ) of the processing result.

[0037] The inverse Fourier transform unit 5002 performs a discrete inverse Fourier transform on the signal I R,source + (f u ) and outputs the signal i R,source + (t m ) of the processing result. The argument calculation unit 5003 calculates the argument θ R,source + (t m ) of the signal i R,source,w (t m ).

[0038] The phase connection unit 5004 performs a phase connection (unwrapping) process to make the argument θ R,source,w (t m ) into a continuous phase value, and sets the processing result as the phase change curve θ R,source ’(t m ).

[0039] Next, the frequency calculation unit 501, based on the phase change curve θ R,source ’(t m ) calculated by the phase change curve calculation unit 500, for the interference signal iR,source (t m ) The estimated curve f of the time-frequency curve showing the time evolution of the frequency of the signal contained in R,source ’(t m ) is calculated. Specifically, the frequency calculation unit 501 performs the calculation of Equation (1).

[0040]

Equation

[0041] At this time, the unit of f R,source ’(t m ) is Hz. In reality, since the phase change curve θ R,source ’(t) is discrete data, the differentiation of Equation (1) may be calculated using differences. In that case, it becomes Equation (2).

[0042]

Equation

[0043] Here, δt is the sampling period of the ADC4. If it is to be aligned with the frequency (dimensionless) output by the discrete Fourier transform, the time-frequency curve becomes Equation (3).

[0044]

Equation

[0045] Here, T is the acquisition time width of the interference signal i R,source (t m ) and N W (=T / δt) is the number of sampling points of the interference signal i R,source (t m ) Note that the time-frequency curves f R,source ’(t m ), f R,source,DL ’(t m ) usually contain noise and may not be smooth data. Therefore, the frequency calculation unit 501 uses the calculated f R,source ’(tm )、f R,source,DL ’(t m ) is subjected to fitting by a polynomial function or B-spline function fitting, and the result is used as the final time-frequency curve f R,source ’(t m ), f R,source,DL ’(t m ).

[0046] Next, the frequency correction unit 502 corrects the time-frequency curve f R,source ’(t m ) or f R,source,DL ’(t m ) to calculate the estimated curve f S,source,z _ s _ FC (t m ) of the interference signal i S,source,z _ s _ FC (t m ) of the light passing through the optical path a and the light passing through the optical path b in the target interferometer 2. This correction utilizes the characteristic of the FMCW method that the frequency of the interference signal is proportional to the optical path length difference between the two optical paths from which the interference signal is derived. S,source,z _ s _ FC ’(t m )

[0047] For the interference signal i R,source (t m ), if the optical path length difference between the two optical paths from which the interference signal is derived is z R , and for the interference signal i S,source,z _ s _ FC (t m ) the optical path length difference between the two optical paths from which the interference signal is derived is z S,FC , then the time-frequency curve f R,source (t m ) of the interference signal i R,source (t m ) and the time-frequency curve f S,source,z _ s _ FC (t m ) of the interference signal i S,source,z _ s _ FC (tm ) has the relationship shown in Equation (4).

[0048]

Number

[0049] That is, at any time, f S,source,z _ s _ FC (t m ) is always f R,source (t m ) times (z S,FC / z R ). Because of such a relationship, the time-frequency curve f S,source,z _ s _ FC (t m )'s estimated curve f S,source,z _ s _ FC ’(t m ) can be calculated as in Equation (5) from the time-frequency curve f R,source (t m )'s estimated curve f R,source ’(t m ) and the optical path length difference z R ,z S,FC .

[0050]

Number

[0051] If it is to be aligned with the frequency (dimensionless) output by the discrete Fourier transform, Equation (5) becomes Equation (6).

[0052]

Number

[0053] Next, BPF51 will be described. Fig. 6 shows an example of the time evolution of the frequency of the interference signal. When the wavelength-sweeping light source 1 is not frequency-linear, that is, when the frequency of the light output from the light source cannot be expressed as a linear function of time, as shown in Fig. 6, the frequency of the interference signal shows a characteristic of a curve with respect to time.

[0054] BPF51 extracts the interference signal i S,source (t m ) from the interference signal i S,source,z _ s _ FC (t m ) obtained from the target interferometer 2 through the ADC4, which is the interference signal of the light passing through the optical path a and the optical path b of the target interferometer 2.

[0055] As described above, the interference signal i S,source (t m ) includes the interference signal i S,source,z _ s _ FC (t m ) of the light passing through the optical path a and the optical path b, the interference signal i S,source,z _ s (t m ) of the light passing through the optical path a and the optical path c, and the interference signal i S,source,z _ s _ FC _ tgt (t m ) of the light passing through the optical path b and the optical path c. Fig. 6 shows the time-frequency curves f S,source,z _ s _ FC (t m ), f S,source,z _ s (t m ), f S,source,z _ s _ FC _ tgt (t m ) of these interference signals, and the time-frequency curve f R,source (t m ) of the interference signal i R,source (t m ) obtained from the reference interferometer 3 through the ADC4.

[0056] Fig. 6(a) shows the interference signal i S,source,z _ s _ FC (t m )'s time-frequency curve f S,source,z _ s _ FC (t m ) is within the two cut-off frequencies (high cut-off frequency f highCut and low cut-off frequency f lowCut ) of BPF51, that is, within the passband of BPF51. Fig. 6(b) shows the case where the time-frequency curve f S,source,z _ s _ FC (t m ) is not within the passband of BPF51.

[0057] In the case of Fig. 6(a), BPF51 takes the interference signal i S,source (t m ) as input and outputs the signal passing through the band determined by the high cut-off frequency f highCut and the low cut-off frequency f lowCut as the interference signal i S,source,z _ s _ FC ’(t m ).

[0058] In the case of Fig. 6(b), the high cut-off frequency f highCut and the low cut-off frequency f lowCut are set as a dynamic BPF51 that changes proportionally to the estimated curve f S,source,z _ s _ FC ’(t m ) of the time-frequency curve obtained from the time-frequency characteristic calculation unit 50. For example, using real number coefficients such as p H , p L , the high cut-off frequency f highCut and the low cut-off frequency f lowCut are set as in equations (7) and (8).

[0059]

Equation

[0060]

Number

[0061] Coefficient p H , p L The specifiable range of p is determined by the measurable distance range of the LiDAR. The measurable distance range is determined by the device manufacturer or user.

[0062] Figures 7 to 12 show the output optical system of the target interferometer 2 and the time-frequency characteristics of the interference signal in the case of the following six conditions. The conditions at this time are the location of the reference plane 103 of the target interferometer 2 and the distance z from the reference plane 103 to the exit end (reflective surface 102) of the optical fiber 26 on the FC21 side S,FC the sign of z, and the distance z from the reference plane 103 to the reflective surface 101 of the measurement target 100 S the sign of z, and the distance z from the exit end (reflective surface 102) of the optical fiber 26 to the reflective surface 101 of the measurement target 100 S,FC,tgt and the distance z S,FC the magnitude relationship of z, and the distance z S,FC,tgt the absolute value |z S,FC,tgt | and the distance z S the absolute value |z S | are determined by the magnitude relationship. The conditions are the following six conditions I to VI.

[0063] Condition I is when z S,FC > 0, z S > 0, z S,FC,tgt < z S,FC is satisfied. Condition II is when z S,FC > 0, z S > 0, z S,FC,tgt > z S,FC is satisfied. Condition III is when z S,FC < 0, z S > 0, z S < |z S,FC | is satisfied. Also, condition IV is when z S,FC < 0, z S > 0, z S >|z S,FC | is satisfied. Condition V is when z S,FC < 0, z S<0,|z S |<|z S,FC,tgt | is satisfied, condition VI is z S,FC <0,z S <0,|z S |>|z S,FC,tgt | is satisfied.

[0064] Figures 7(a), 8(a), 9(a), 10(a), 11(a), and 12(a) show the output optical systems of the target interferometer 2 in the cases of conditions I, II, III, IV, V, and VI, respectively. Figures 7(b), 8(b), 9(b), 10(b), 11(b), and 12(b) show the time-frequency characteristics of the interference signals in the cases of conditions I, II, III, IV, V, and VI, respectively.

[0065] Distance z S,FC ,z S,FC,tgt ,z S shall have positive and negative directions. For the distance z S,FC , in the cases where the reference plane 103 of the target interferometer 2 is located between the output end (reflective surface 102) of the optical fiber 26 and the circulator 20 as shown in Figure 7 (condition I) and Figure 8 (condition II), it is considered positive, and in the cases where the reference plane 103 is located on the side from the reflective surface 102 toward the measurement target 100 as shown in Figure 9 (condition III), Figure 10 (condition IV), Figure 11 (condition V), and Figure 12 (condition VI), it is considered negative.

[0066] Distance z S , in the cases where the reference plane 103 is between the reflective surface 102 and the circulator 20, or between the reflective surface 102 and the reflective surface 101 of the measurement target 100 as shown in Figure 7 (condition I), Figure 8 (condition II), Figure 9 (condition III), and Figure 10 (condition IV), it is considered positive, and in the cases where the reference plane 103 is located at a position farther from the reflective surface 101 of the measurement target 100 than the reflective surface 102 when viewed from the reflective surface 102 as shown in Figure 11 (condition V) and Figure 12 (condition VI), it is considered negative. The distance z S,FC,tgt is all positive in conditions I to VI.

[0067] In the case of condition I, it is necessary to satisfy the conditions of equations (9) and (10).

[0068]

Number

[0069]

Number

[0070] Since the frequency and the distance are in a proportional relationship, the coefficient p H , p L and the distance z S,FC , z S,FC,tgt , z S The relationship with z is as shown in equations (11) and (12).

[0071]

Number

[0072]

Number

[0073] Here, for the distance z S,FC , z S,FC,tgt , z S There is the following relationship.

[0074]

Number

[0075] Therefore, the possible range of the coefficient p H , p L is as shown in equations (14) and (15).

[0076]

Number

[0077]

Number

[0078] Distance z S,FC,tgt The measurable range is defined as from the shortest z S,FC,tgt,min to the longest z S,FC,tgt,max Among them, the range of the coefficient p in Condition I H , p L is as shown in Formula (16) and Formula (17).

[0079]

Number

[0080]

Number

[0081] Similarly, the range of the coefficient p in Condition II H , p L is as shown in Formula (18) and Formula (19).

[0082]

Number

[0083]

Number

[0084] The coefficient p in Condition III H , p L has a range as shown in Formula (20) and Formula (21).

[0085]

Number

[0086]

Number

[0087] The coefficient p in Condition IV H , pL The possible range is as shown in Formula (22) and Formula (23).

[0088]

Number

[0089]

Number

[0090] The coefficient p under Condition V H , p L The possible range is as shown in Formula (24) and Formula (25).

[0091]

Number

[0092]

Number

[0093] The coefficient p under Condition VI H , p L The possible range is as shown in Formula (26) and Formula (27).

[0094]

Number

[0095]

Number

[0096] When Conditions I and III are combined into Condition VII, Condition VII is as shown in Formula (28) and Formula (29).

[0097]

Number

[0098]

Mathematics

[0099] When combining Conditions II and IV into Condition VIII, Condition VIII becomes as shown in Formula (30) and Formula (31).

[0100]

Mathematics

[0101]

Mathematics

[0102] The coefficients p in Conditions I, II, III, IV, V, VI, VII, and VIII H , p L Regarding the range, the description has been made for the real frequency with the unit of Hz. However, since the dimensionless frequency by Fourier transform is proportional to the real frequency, the range of the coefficients p H , p L for the dimensionless frequency by Fourier transform is the same as above.

[0103] The coefficient calculation unit 52 calculates the coefficients p H , p L that satisfy the above range. For example, in the case of Condition VII (Conditions I and III), the coefficient calculation unit 52 calculates the coefficients p H , p L using Formulas (32) and (33).

[0104]

Mathematics

[0105]

Mathematics

[0106] However, for z in Formulas (32) and (33)min,mgn and z max,mgn is a positive value that can be set as appropriate and must satisfy conditions such as those in Equation (34) and Equation (35).

[0107]

Number

[0108]

Number

[0109] Also, in the case of Condition VIII (Conditions II and IV), the coefficient calculation unit 52 calculates the coefficients p H , p L by Equation (36) and Equation (37).

[0110]

Number

[0111]

Number

[0112] However, the z in Equation (36) min,mgn is a positive value that can be set as appropriate and must satisfy a condition such as that in Equation (38).

[0113]

Number

[0114] The coefficient p in the case of Condition VIII L may be, for example, 0.5. Also, in the case of Condition V, the coefficient calculation unit 52 calculates the coefficients p H , p L by Equation (39) and Equation (40).

[0115]

Number

[0116]

Number

[0117] However, z in formula (40) max,mgn is a positive value that can be set as appropriate and must satisfy the conditions as in formula (41).

[0118]

Number

[0119] Coefficient p in the case of condition V H may be, for example, 1.5. Also, in the case of condition VI, the coefficient calculation unit 52 calculates the coefficient p H , p L by formulas (42) and (43).

[0120]

Number

[0121]

Number

[0122] However, z in formula (43) min,mgn is a positive value that can be set as appropriate and must satisfy the conditions as in formula (44).

[0123]

Number

[0124] Coefficient p in the case of condition VI H may be, for example, 1.5. The coefficient calculation unit 52 is externally input with z S,FC , z S,FC,tgt,min , z S,FC,tgt,max , z min,mgn , zmax,mgn The value of H , p L is input. The coefficient calculation unit 52 sets the calculated coefficients p However, the coefficients p H , p L may be directly input to the BPF 51 from outside the signal processing device 5. In this case, the coefficient calculation unit 52 becomes unnecessary.

[0125] The high cut-off frequency f highCut and the low cut-off frequency f lowCut are shown in FIG. 13 as a configuration example of the dynamic BPF 51 that varies. In FIG. 13, as an example of the BPF 51, a FIR (Finite Impulse Response) filter is shown. The BPF 51 is composed of a weight coefficient calculation unit 510, a delay unit 511, a weight integration unit 512, and an addition unit 513. The BPF 51 performs calculations as shown in Equation (45).

[0126]

Equation

[0127] The weight coefficient calculation unit 510 calculates the filter coefficient w m every time (every 1 sampling clock) at time t. Here, as an example, a method of combining two low-pass filters will be described. Equation (46) represents that the coefficients p tm (k) (where k = 0 to N W - 1) are calculated. Here, as an example, a method of combining two low-pass filters will be described. Equation (46) is an equation showing that the coefficients p H , p L of the BPF 51 are composed of the difference between the coefficient of the low-pass filter with a high cut-off frequency and the coefficient of the low-pass filter with a low cut-off frequency.

[0128]

Equation

[0129] In Equation (46), the cut-off frequency (dimensionless) is represented by q c and D qc(·) is used as the coefficient of the low-pass filter. p in Equation (46) H ·q(t m ) and p L ·q(t m ) each represent the cut-off frequency. Coefficients p H and p L are real numbers such that p H > p L . Also, q(t m ) represents the non-dimensional frequency in the discrete Fourier transform and is as shown in Equation (47).

[0130]

Number

[0131] However, f f (t m ) is represented by Equation (48).

[0132]

Number

[0133] Δf in Equation (47) is the unit frequency of the discrete Fourier transform (the reciprocal of the acquisition time width T of the interference signal i R,source (t m ). An example of the above D f (·) is shown in Equation (49). In this example, D f (·) is in the form of the Dirichlet Kernel.

[0134]

Number

[0135] However, M is represented by Equation (50).

[0136]

Number

[0137] floor(x) is the largest integer not exceeding the real number x. Thus, for frequency f S,source,z _ s _ FC (t m ) is converted to the dimensionless frequency q(t m ) by discrete Fourier transform, enabling filtering using Equation (45) for both frequency and dimensionless frequency.

[0138] The filter with D p·q(tm) (k) in Equation (49) as the filter coefficient is a low-pass filter that transmits frequency components in the range of -p·q(t m ) ~ +p·q(t m ) with the same intensity. When p·q(t m ) is an integer, D p·q(tm) (k) becomes the Dirichlet kernel of period N w .

[0139] The filter coefficients w tm (k) in Equations (45) and (46) are the difference between the filter coefficients D pH·q(tm) (k) and D pL·q(tm) (k) of two low-pass filters with different cut-off frequencies. Therefore, the filter with w tm (k) as the filter coefficient is a band-pass filter that transmits -p H ·q(t m ) ~ -p L ·q(t m ) and +p L ·q(t m ) ~ +p H ·q(t m ) with the same intensity.

[0140] The delay unit 511 sequentially captures the interference signal i S,source (t m ) for each sample, and outputs the N W -point data shifted by one sampling clock for each sampling clock to the weighted integration unit 512. The delay unit 511 consists of N W -1 delay units 5110. When k = N WThe data output from the first delay unit 5110 is the data (N W −1) samples before the sampling clock for the k = 0th data that is output directly without passing through the delay unit 5110.

[0141] The weight integration unit 512 consists of N W integrating units 5120. The kth (k = 0 to N W −1) integrating unit 5120 calculates the product of the kth filter coefficient w tm (k) calculated by the weight coefficient calculation unit 510 and the kth data output from the delay unit 511 at each time t m .

[0142] The addition unit 513 outputs the sum of the products calculated by the weight integration unit 512 as the interference signal i S,source,z _ s _ FC ’(t m ).

[0143] Next, the ranging unit 53 performs the same processing as the FMCW LiDAR described in Non-Patent Documents 1 and 3, using the interference signal i S,source (t m ) obtained from the target interferometer 2 through the ADC 4 as the target interference signal and the interference signal i S,source,z _ s _ FC ’(t m ) as the reference interference signal, to obtain the ranging result z S ’.

[0144] An example of the configuration of the ranging unit 53 is shown in FIG. 14. The ranging unit 53 includes a phase change curve calculation unit 530, a resampling time calculation unit 531, resampling units 532 and 533, a δθ calculation unit 534, frequency intensity distribution calculation units 535 and 536, peak position frequency calculation units 537 and 538, a range calculation unit 539, and a distance calculation unit 540.

[0145] For simplicity of explanation, hereinafter, the target interference signal i S,source (t m ) is denoted as i S (t m) is replaced with the reference interference signal i S,source,z _ s _ FC ’(t m ) is replaced with i R (t m ) and will be described.

[0146] The phase change curve calculation unit 530 calculates a phase change curve showing the time transition of the phase of i R (t m ) from the target interferometer 2. The phase change curve calculation unit 530 performs the same operation as the phase change curve calculation unit 500. The phase change curve calculation unit 530 is composed of a Fourier transform unit 5300, a negative frequency component zero unit 5301, an inverse Fourier transform unit 5302, an argument calculation unit 5303, and a phase connection unit 5304 as shown in FIG. 15. R (t m ) as shown in FIG. 15.

[0147] The Fourier transform unit 5300 outputs a signal I R (t m ) obtained by performing a discrete Fourier transform on the reference interference signal i R (f u ). The negative frequency component zero unit 5301 performs a process of setting the negative frequency component of the signal I R (f u ) to zero, and outputs the signal I R + (f u ) of the processing result.

[0148] The inverse Fourier transform unit 5302 outputs a signal i R + (f u ) obtained by performing a discrete inverse Fourier transform on the signal I R + (t m ). The argument calculation unit 5303 calculates the argument θ R + (t m ) of the signal i R,w (t m ).

[0149] The phase connection unit 5304 is the argument θR,w (t m ) is subjected to an unwrapping process to make it a continuous phase value, and the processing result is used as a phase change curve θ R ’(t m ).

[0150] Next, the resampling time calculation unit 531 generates a resampling time τ R ’(t m ) to be used in the resampling units 532 and 533 from the phase change curve θ n .

[0151] FIG. 16 is a diagram showing the relationship between the phase change curve θ R ’(t m ) and the resampling time τ n . The resampling time τ n is the time when the phase change curve θ R ’(t m ) becomes an equal phase interval δθ. At this time, the number N n of the resampling time τ res is as shown in Equation (51).

[0152]

Equation

[0153] Δθ is the phase change width of the phase change curve θ R ’(t m ). In FIG. 16, an example of the case where N res = 15 is shown. The distance to the measurement target 100 is obtained by measuring the distance z S,FC,tgt or z S of the target interferometer 2. When measuring the distance z S,FC,tgt , the phase interval δθ is made to satisfy the condition of Equation (52) from the sampling theorem.

[0154]

Equation

[0155] Similarly, when measuring the distance z S the phase interval δθ satisfies the condition of Equation (53) according to the sampling theorem.

[0156]

Number

[0157] z S,FC,tgt,max is the maximum measurable distance of z S,FC,tgt z S,max is the maximum measurable distance of z S z. Also, max(x, y) means a function that outputs the larger of x and y.

[0158] Next, the δθ calculation unit 534 calculates the phase interval δθ that satisfies Equation (52) or Equation (53), and passes the value of the phase interval δθ to the resampling time calculation unit 531. Specifically, when measuring the distance z S,FC,tgt the δθ calculation unit 534 calculates the phase interval δθ according to Equation (54).

[0159]

Number

[0160] Also, when measuring the distance z S the δθ calculation unit 534 calculates the phase interval δθ according to Equation (55).

[0161]

Number

[0162] N 1per is a positive real number greater than or equal to 1, and represents the number of samples for half a period of the target interference signal i S (t m ) or the reference interference signal i R (t m ). In this embodiment, an example where the δθ calculation unit 534 calculates the phase interval δθ has been shown, but the value of δθ may be directly input to the resampling time calculation unit 531 from outside the distance measurement unit 53. In this case, the δθ calculation unit 534 becomes unnecessary.

[0163] The resampling unit 532 samples the reference interference signal i R (t m ) synchronously with the resampling time τ n and outputs the interference signal i R ’(τ n ) after the resampling process. The resampling unit 533 samples the target interference signal i S (t m ) synchronously with the resampling time τ n and outputs the interference signal i S ’(τ n ) after the resampling process.

[0164] The frequency intensity distribution calculation unit 535 calculates the frequency intensity distribution of the interference signal i R ’(τ n ) obtained from the resampling unit 532. The frequency intensity distribution calculation unit 536 calculates the frequency intensity distribution of the interference signal i S ’(τ n ) obtained from the resampling unit 533. The configuration of the frequency intensity distribution calculation unit 535 is shown in FIG. 17, and the configuration of the frequency intensity distribution calculation unit 536 is shown in FIG. 18.

[0165] The frequency intensity distribution calculation unit 535 is composed of a window function unit 5350, a discrete Fourier transform unit 5351, and an intensity calculation unit 5352. The frequency intensity distribution calculation unit 536 is composed of a window function unit 5360, a discrete Fourier transform unit 5361, and an intensity calculation unit 5362.

[0166] The window function unit 5350 multiplies the interference signal i R ’(τ n ) by a predetermined window function. The discrete Fourier transform unit 5351 performs a discrete Fourier transform on the signal i R,win (τn ) The signal I obtained by performing the discrete Fourier transform on R,win (f v ) is output. The intensity calculation unit 5352 calculates the intensity of the signal I R,win (f v ) and outputs a signal I R,dis ’(f v ) indicating the intensity.

[0167] The window function unit 5360 multiplies the interference signal i S ’(τ n ) by a predetermined window function. The discrete Fourier transform unit 5361 performs the discrete Fourier transform on the signal i S,win (τ n ) output from the window function unit 5360 to obtain a signal I S,win (f v ) and outputs it. The intensity calculation unit 5362 calculates the intensity of the signal I S,win (f v ) and outputs a signal I S,dis ’(f v ) indicating the intensity.

[0168] Note that, as shown in FIGS. 19 and 20, the discrete Fourier transform unit 5351 and the intensity calculation unit 5352 may be replaced with a power spectrum calculation unit 5353, and the discrete Fourier transform unit 5361 and the intensity calculation unit 5362 may be replaced with a power spectrum calculation unit 5363.

[0169] As the window function, a Hanning window function, a Hamming window function, a Blackman window function, a Gaussian window function, etc. can be considered. In particular, a window function with both ends being zero is desirable in the sense that it is less likely to generate pseudo noise (spurious). Such window functions include a Hanning window function and a Blackman window function.

[0170] The peak position frequency calculation unit 537 calculates the frequency of the peak position within a desired frequency range of the signal I R,dis ’(f v ) obtained from the frequency intensity distribution calculation unit 535. The peak position frequency calculation unit 538 calculates the frequency of the peak position within a desired frequency range of the signal I S,dis ’(f vCalculate the frequency at the peak position within the desired frequency range of

[0171] Fig. 21(a) shows the signal I R,dis ’(f v )'s frequency intensity distribution, and Fig. 21(b) shows the signal I S,dis ’(f v )'s frequency intensity distribution. As shown in Fig. 21, the signal I R,dis ’(f v ) has one peak, while the signal I S,dis ’(f v ) has three peaks. Therefore, the peak position frequency calculation unit 537 and the peak position frequency calculation unit 538 are slightly different in configuration.

[0172] The peak position frequency calculation unit 538 selects one of the three peaks of the signal I S,dis ’(f v ) and then performs the same processing as the peak position frequency calculation unit 537 to calculate the frequency at the peak position. First, the peak position frequency calculation unit 537 will be described with reference to Fig. 22.

[0173] As shown in Fig. 22, the peak position frequency calculation unit 537 is composed of a maximum position frequency search unit 5370, a quadratic function fitting unit 5371, a window function unit 5372, a zero-padding interpolation unit 5373, and a maximum position frequency search unit 5374.

[0174] The maximum position frequency search unit 5370 detects the frequency f R,dis ’ at which the signal I v ) reaches its maximum value. The signal I R ’(f R,dis ) is discrete data. Therefore, the frequency f v at which the signal I R,dis ’(f v ) reaches its true maximum value is different from f R ’. The frequency f R ’ can be said to be an approximate value of the true frequency f R . R

[0175] ​The quadratic function fitting unit 5371 uses the frequency f detected by the maximum position frequency search unit 5370 R ’ as the center, and ±f R,w,fit in the region, that is, f R ’ - f R,w,fit ~ f R ’ + f R,w,fit (a region with 2f R,w,fit + 1 data points) of the signal I R,dis ’(f v ) and performs logarithmic conversion on its intensity. Then, the quadratic function fitting unit 5371 obtains an approximate curve of the logarithmically converted signal through quadratic function fitting, and calculates the frequency f R ” at which the approximate curve reaches the maximum value. The frequency f R ” can be regarded as an approximate value of the frequency f R ’ at which the signal I R,dis ’(f v ) reaches the true maximum value f R . The value f R,w,fit specifying the frequency range is input from the outside.

[0176] The window function unit 5372 uses the frequency f R ” calculated by the quadratic function fitting unit 5371 as the center, and ±f R,w,win in the region, that is, f R ” - f R,w,win ~ f R ” + f R,w,win (a region with 2f R,w,win + 1 data points) of the signal I R,dis ’(f v ) and multiplies it by a predetermined window function, and outputs the processed signal I R,dis,win ’(f v ). The value f R,w,win specifying the frequency range is input from the outside.

[0177] The zero-padding interpolation unit 5373 performs zero-padding interpolation on the signal I R,dis,win ’(f v ) output from the window function unit 5372, and outputs the processed signal I R,dis,win,0pad ’(f v ). The signal I R,dis,win ’(f v) is discrete data. The zero-padding interpolation unit 5373 adds N R,dis,win,0pad ’(f v ) - 1 zeros between the data of the signal I R,0pad . As a result, the number of data of the signal I R,dis,win,0pad ’(f v ) becomes N R,dis,win ’(f v ) times that of the signal I R,0pad . N R,0pad is an integer of 2 or more.

[0178] The maximum position frequency search unit 5374 detects the frequency f R,dis,win,0pad ’(f v ) at which the signal I R becomes the maximum value within the desired frequency range.

[0179] Next, the peak position frequency calculation unit 538 will be described. As shown in FIG. 21(b), there are three peaks in the signal I S,dis ’(f v ). The frequencies of the peak positions respectively correspond to the optical path length differences between the optical paths a and b, the optical path length differences between the optical paths a and c, and the optical path length differences between the optical paths b and c shown in FIG. 2.

[0180] The frequency f S,dis ’(f v ) of the peak position corresponding to the optical path length difference between the optical paths a and b, which is f R (=f S,FC ), is the same as the frequency of the peak position of the signal I R,dis ’(f v ) and is proportional to the distance z S,FC which is half of the optical path length difference between the optical paths a and b.

[0181] The peaks related to the position of the measurement object 100 are the two peaks related to the optical path c. The frequency of the peak position corresponding to the optical path length difference between the optical paths b and c is proportional to the distance z S,FC,tgt . The frequency of the peak position corresponding to the optical path length difference between the optical paths a and c is proportional to the distance z S . In distance measurement, the frequency f S,FC,tgt of the peak position related to the distance z S,FC,tgt、or the distance z which is half of the optical path length difference between optical paths a and c S The frequency f of the peak position related to S will be calculated.

[0182] Distance z S,FC,tgt When measuring, the measurable range z S,FC,tgt,min ~z S,FC,tgt,max The frequency z corresponding to S,FC,tgt,min ·f R / |z S,FC |~z S,FC,tgt,max ·f R / |z S,FC |, within the range of the signal I S,dis ’(f v ) the frequency of the peak position may be searched for.

[0183] Distance z S When measuring, the measurable range z S,min ~z S,max The frequency z corresponding to S,min ·f R / |z S,FC |~z S,max ·f R / |z S,FC |, within the range of the signal I S,dis ’(f v ) the frequency of the peak position may be searched for. Here, the shortest measurable distance z S of distance z S,min and the longest distance z S,max are the shortest measurable distance z S,FC,tgt of distance z S,FC,tgt,min and the longest distance z S,FC,tgt,max and distance z S,FC can be calculated as in equations (56) and (57).

[0184]

Equation

[0185]

Equation

[0186] The range calculation unit 539 of the distance measurement unit 53 calculates the distance z S,FC,tgt When measuring, the distance z S,FC and the shortest measurable distance z S,FC,tgt,min and the longest distance z S,FC,tgt,max and the frequency f obtained by the peak position frequency calculation unit 537 R From this, the lowest frequency z of the above frequency range S,FC,tgt,min ·f R / |z S,FC | and the highest frequency z S,FC,tgt,max ·f R / |z S,FC | are calculated. Also, the range calculation unit 539 calculates the distance z S When measuring, the distance z S,FC and the shortest measurable distance z S,min and the longest distance z S,max and the frequency f obtained by the peak position frequency calculation unit 537 R From this, the lowest frequency z S,min ·f R / |z S,FC | and the highest frequency z S,max ·f R / |z S,FC | are calculated.

[0187] As shown in FIG. 23, the peak position frequency calculation unit 538 is composed of a maximum position frequency search unit 5380, a quadratic function fitting unit 5381, a window function unit 5382, a zero-padding interpolation unit 5383, and a maximum position frequency search unit 5384.

[0188] The difference from the peak position frequency calculation unit 537 is the maximum position frequency search unit 5380. When measuring the distance z S,FC,tgt , the lowest frequency z calculated by the range calculation unit 539 S,FC,tgt,min ·f R / |z S,FC | and the highest frequency z S,FC,tgt,max ·f R / |z S,FC |, the frequency range z determined by S,FC,tgt,min ·f R / |z S,FC |~z S,FC,tgt,max ·f R / |zS,FC The signal I in S,dis ’(f v ) at the peak position frequency f S ’ is detected.

[0189] Also, when measuring the distance z S , the minimum frequency z S,min ·f R / |z S,FC | and the maximum frequency z S,max ·f R / |z S,FC | determine the frequency range z S,min ·f R / |z S,FC |~z S,max ·f R / |z S,FC | of the signal I S,dis ’(f v ) at the peak position frequency f S ’ is detected.

[0190] The quadratic function fitting unit 5381 uses the frequency f S ’ detected by the maximum position frequency search unit 5380 as the center, within the range of ±f S,w,fit , that is, the region from f S ’ - f S,w,fit to f S ’ + f S,w,fit (a region with 2f S,w,fit + 1 data points). The intensity of the signal I S,dis ’(f v ) in this region is logarithmically transformed. Then, the quadratic function fitting unit 5381 obtains an approximate curve of the logarithmically transformed signal by quadratic function fitting, and calculates the frequency f S ” at which the approximate curve reaches its maximum value. The frequency f S ” can be said to be an approximate value of the frequency f S ’ at which the signal I S,dis ’(f v ) reaches its true maximum value, similar to f S . The value f S,w,fit specifying the frequency range is input from the outside.

[0191] The window function section 5382 applies a predetermined window function to the signal I' (f) in the region centered on the frequency f calculated by the quadratic function fitting section 5381 and within the range of ±f, that is, in the region of f - f to f + f (a region with 2f + 1 data points), and outputs the processed signal I' (f). The value f specifying the frequency range is input from the outside. S centered on ±f S,w,win in the region of, that is, f S - f S,w,win to f S + f S,w,win (a region with 2f + 1 data points), and outputs the processed signal I' (f). The value f specifying the frequency range is input from the outside. S,w,win in the region of f - f to f + f (a region with 2f + 1 data points), and outputs the processed signal I' (f). The value f specifying the frequency range is input from the outside. S,dis ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The value f specifying the frequency range is input from the outside. S,dis,win ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The value f specifying the frequency range is input from the outside. S,w,win is input from the outside.

[0192] The zero-padding interpolation section 5383 performs zero-padding interpolation on the signal I' (f) output from the window function section 5382, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win,0pad ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win ’(f v ) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,0pad - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win,0pad ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,dis,win ’(f v ) by multiplying it with a predetermined window function, and outputs the processed signal I' (f). The signal I' (f) is discrete data. The zero-padding interpolation section 5383 adds N - 1 zeros between the data points of the signal I' (f). As a result, the number of data points of the signal I' (f) becomes N times that of the signal I' (f). N is an integer of 2 or more. S,0pad times that of the signal I' (f). N is an integer of 2 or more. S,0pad is an integer of 2 or more.

[0193] The maximum position frequency search section 5384 detects the frequency f at which the signal I' (f) becomes the maximum value within the desired frequency range. S,dis,win,0pad ’(f v ) becomes the maximum value within the desired frequency range. S is detected.

[0194] Next, the distance calculation section 540 of the distance measurement section 53 calculates the distance z based on the frequency f calculated by the peak position frequency calculation section 537 and the frequency f calculated by the peak position frequency calculation section 538. R calculated by the peak position frequency calculation section 537 and the frequency f S calculated by the peak position frequency calculation section 538 and the distance zS,FC Based on this, the estimated value z of the distance from the output end (reflective surface 102) of the optical fiber 26 to the reflective surface 101 of the measurement target 100 S,FC,tgt ’ is calculated by Equation (58).

[0195]

Equation

[0196] Alternatively, the distance calculation unit 540 calculates the estimated value z of the distance from the reference surface 103 to the reflective surface 101 of the measurement target 100 S ’ by Equation (59).

[0197]

Equation

[0198] Note that since the signs of the estimated values z S,FC,tgt ’ and z S ’ are positive, the estimated values z S,FC,tgt ’ and z S ’ are the absolute values of the distances from the reference surface 103. Since the distance z S,FC,tgt is always positive, there is no problem with the sign for the estimated value z S,FC,tgt ’. On the other hand, since the distance z S can be either positive or negative, it is necessary to check in advance which sign the estimated value z S ’ will take before the measurement.

[0199] Next, as an effect of this embodiment, an example of improving the ranging accuracy is shown. Here, the end face of the FC21 of the target interferometer 2 is PC (Physical Contact) polished, a 50% reflective coating is formed on the end face, and the end face reflectivity is set to 0.356. 2z S,FC for 1.2 m of the optical fiber, 2z S for 4.2 m of the optical fiber, and z S,FC,tgt for 3.0 m of the optical fiber. The optical path length difference of the reference interferometer 3 is set to 2.8 m of the optical fiber. For the wavelength sweep light source 1, the sweep frequency is 1 kHz, the sweep wavelength width is 0.5 nm, and the sweep center wavelength is 1308 nm.

[0200] Regarding the configuration of the conventional distance measuring device, the end face of the FC of the target interferometer is APC polished, an AR coat is formed on the end face, and the end face reflectance is 1.01×10 -7 is set. Other parameters are the same as those of the distance measuring device of this embodiment.

[0201] The number of measurements is set to 1000 times. When the distance measurement accuracy of this embodiment is expressed by the standard deviation, it is 7.72 μm. The conventional distance measurement accuracy was 12.8 μm. According to this result, since the standard deviation of the distance measurement result of this embodiment decreased to 60.3% when the conventional distance measurement result was taken as 100%, it can be considered that the distance measurement accuracy of this embodiment is 1.66 times higher than that of the prior art.

[0202] [Second Embodiment] Next, a second embodiment of the present invention will be described. In the first embodiment, the interference signal i S,source,z _ s _ FC (t m ) of the time-frequency curve f S,source,z _ s _ FC (t m ) of the estimated curve f S,source,z _ s _ FC ’(t m ) was calculated based on the interference signal i R,source (t m ) obtained from the reference interferometer 3 through the ADC 4.

[0203] On the other hand, before distance measurement, a shielding object is installed between the measurement target 100 of the target interferometer 2 and the FC 21 to block the optical path c, so that only the interference signal i S,source,z _ s _ FC (t m ) of the light passing through the optical path a and the light passing through the optical path b is output. Then, the interference signal i S,source,z _ s _ FC (t m) is input, and the interference signal i S,source,z _ s _ FC (t m )'s time-frequency curve f S,source,z _ s _ FC (t m )'s estimated curve f S,source,z _ s _ FC ’(t m ) is calculated.

[0204] During ranging, the obstacle between the measurement target 100 and FC21 is removed. Then, the estimated curve f S,source,z _ s _ FC ’(t m ) calculated in advance by the time-frequency characteristic calculation unit 50 is used to operate the BPF 51.

[0205] The configuration of the ranging device of this embodiment is shown in FIG. 24. In this embodiment, since the coupler C1 and the reference interferometer 3 are not required, the ranging device can be miniaturized compared with the first embodiment.

[0206] [Third Embodiment] In the first and second embodiments, when measuring the distance z S,FC,tgt , it is desirable that the amplitude of the interference signal i S,source,z _ s _ FC _ tgt (t m ) passing through the optical path b and the optical path c is large. In this embodiment, in the configuration of the target interferometer 2 as shown in FIG. 25, the reflectivity R FC of the end face of the optical fiber 26 connected to FC21 is set so that the amplitude of the interference signal i S,source,z _ s _ FC _ tgt (t m ) passing through the optical path b and the optical path c becomes maximum.

[0207] Let the power splitting ratio of coupler C2 be r:1 - r (0 < r < 1), the power splitting ratio of coupler C3 be 0.5:0.5, the coupling efficiency from the measurement target 100 side of FC21 be κ, and the surface reflectivity of the measurement target 100 be R tgt be defined as such.

[0208] Assuming the power of the light input to coupler C2 is P, the intensity of the reflected light from the end face of the optical fiber 26 connected to FC21 is P·r·R FC Also, the intensity of the light that passes through the end face of the optical fiber 26 connected to FC21 and enters the optical fiber 26 from the reflected light from the measurement target 100 is P·r·R tgt ·κ·(1 - R FC ) 2 The interference signal intensity i 0 of the FMCW LiDAR is proportional to the product of the electric field intensities of the two lights, so the relationship is as shown in Equation (60).

[0209]

Equation

[0210] Differentiating the right side of Equation (60) with respect to the reflectivity R FC of the end face of the optical fiber 26 results in Equation (61).

[0211]

Equation

[0212] Equation (60) is a convex - up function with respect to the reflectivity R FC of the end face of the optical fiber 26. The reflectivity R FC at which the maximum value occurs is the R FC for which Equation (61) becomes 0, and at that time, R FC = 1 / 3. Therefore, by setting the reflectivity R FC of the end face of the optical fiber 26 to 1 / 3, the interference signal i S,source,z _ s _ FC _ tgt (t mso that the amplitude of S,FC,tgt can be improved in the measurement of noise resistance and the ranging accuracy can be improved.

[0213] The signal processing device 5 described in the first to third embodiments can be realized by a computer including a CPU (Central Processing Unit), a storage device, and an interface, and a program for controlling these hardware resources. A configuration example of this computer is shown in FIG. 26.

[0214] The computer includes a CPU 300, a storage device 301, and an interface device (I / F) 302. An ADC 4 or the like is connected to the I / F 302. A program for realizing the ranging method of the present invention is stored in the storage device 301. The CPU 300 executes the processes described in the first to third embodiments according to the program stored in the storage device 301.

Industrial Applicability

[0215] The present invention can be applied to a ranging device.

Explanation of Signs

[0216] 1…Wavelength-sweeping light source, 2…Target interferometer, 3…Reference interferometer, 4…AD converter, 5…Signal processing device, 6, 23~29, 33~39…Optical fiber, C1~C5…Coupler, 20, 30…Circulator, 21, 31…Fiber collimator, 22, 32…Balanced photodetector, 50…Time-frequency characteristic calculation unit, 51…Band-pass filter, 52…Coefficient calculation unit, 53…Distance measurement unit, 100…Measurement target, 200…Mirror, 500…Phase change curve calculation unit, 501…Frequency calculation unit, 502…Frequency correction unit, 510…Weight coefficient calculation unit, 511, 5110…Delay unit, 512…Weight integration unit, 513…Addition unit, 530…Phase change curve calculation unit, 531…Resampling time calculation unit, 532…Resampling unit, 533…Resampling unit, 534…δθ calculation unit, 535, 536…Frequency intensity distribution calculation unit, 537, 538…Peak position frequency calculation unit, 539…Range calculation unit, 540…Distance calculation unit, 5000, 5300…Fourier transform unit, 5001, 5301…Negative frequency component zero unit, 5002, 5302…Inverse Fourier transform unit, 5003, 5303…Deflection angle calculation unit, 5004, 5304…Phase connection unit, 5120…Integration unit, 5350, 5360, 5372, 5382…Window function unit, 5351, 5361…Discrete Fourier transform unit, 5352, 5362…Intensity calculation unit, 5353, 5363…Power spectrum calculation unit, 5370, 5374, 5380, 5384…Maximum position frequency search unit, 5371, 5381…Quadratic function fitting unit, 5373, 5383…Zero-padding interpolation unit.

Claims

1. A first interferometer configured to convert a first interference light, which is obtained by interfering continuous light output from a light source with its wavelength swept over time and reflected light obtained by reflecting the light output from the light source by a measurement target, into an electrical signal and output a first interference signal; A signal processing device configured to calculate a distance to the measurement target in the first interferometer; The first interferometer includes a reflecting surface in the middle of the optical path of the reflected light; The signal processing device: Extracts a second interference signal obtained from a second interference light, which is obtained by interfering the light output from the light source with the reflected light obtained by reflecting the light output from the light source by the reflecting surface, from the first interference signal by a frequency filter, resamples the first interference signal in synchronization with a resampling time calculated based on a phase change curve of the second interference signal, and then calculates a distance to the measurement target in the first interferometer based on a frequency of a peak of a first signal obtained by performing a discrete Fourier transform on the resampled first interference signal and a frequency of a peak of a second signal obtained by performing a discrete Fourier transform on the second interference signal resampled in synchronization with the resampling time. A distance measuring device characterized by the above.

2. In the distance measuring device according to Claim 1, Further includes a second interferometer configured to convert a third interference light, which is obtained by interfering the light output from the light source with light having a predetermined optical path length difference from the light output from the light source, into an electrical signal and output a third interference signal; The signal processing device: The frequency filter; A time-frequency characteristic calculation unit configured to calculate an estimation curve showing a time transition of the frequency of the second interference signal from the third interference signal; The frequency filter is characterized in that a high cut-off frequency and a low cut-off frequency change in proportion to the estimation curve calculated by the time-frequency characteristic calculation unit. A distance measuring device characterized by the above.

3. In the distance measuring device according to Claim 1, The signal processing device: The frequency filter; A time-frequency characteristic calculation unit configured to calculate an estimation curve showing a time transition of the frequency of the second interference signal from an interference signal output from the first interferometer in a state where light incident on the measurement target from the light source is blocked before distance measurement; The frequency filter is characterized in that a high cut-off frequency and a low cut-off frequency change in proportion to the estimation curve calculated by the time-frequency characteristic calculation unit. A distance measuring device characterized by the above.

4. In the distance measuring device according to claim 2 or 3, the signal processing device includes a coefficient calculation unit configured to calculate a coefficient determined by a measurable distance range, wherein a high cut-off frequency and a low cut-off frequency of the frequency filter are determined by the estimation curve and the coefficient, the distance measuring device being characterized thereby.

5. In the distance measuring device according to any one of claims 1 to 4, the signal processing device includes a phase change curve calculation unit configured to calculate a phase change curve indicating a time transition of a phase of the second interference signal, a resampling time calculation unit configured to calculate the resampling time based on the phase change curve and a predetermined phase interval, a first resampling unit configured to resample the first interference signal in synchronization with the resampling time, a second resampling unit configured to resample the second interference signal in synchronization with the resampling time, a first peak position frequency calculation unit configured to detect a frequency of a peak of a first signal obtained by performing a discrete Fourier transform on the resampled first interference signal, a second peak position frequency calculation unit configured to detect a frequency of a peak of a second signal obtained by performing a discrete Fourier transform on the resampled second interference signal, and a distance calculation unit configured to calculate a distance to the measurement target in the first interferometer based on the frequency of the peak of the discretely Fourier-transformed first signal and the frequency of the peak of the discretely Fourier-transformed second signal, the distance measuring device being characterized thereby.

6. In the distance measuring device according to any one of claims 1 to 5, the first interferometer includes a first coupler configured to divide light from the light source into two, a second coupler configured to combine one of the lights branched by the first coupler and the other light branched by the first coupler and reflected by the measurement target, a circulator configured to output the other light branched by the first coupler to an optical fiber and output light from the optical fiber to the second coupler, and a fiber collimator configured to irradiate the measurement target with light from the optical fiber and allow reflected light from the measurement target to enter the optical fiber, wherein the reflecting surface is an end surface of the optical fiber connected to the fiber collimator, the distance measuring device being characterized thereby.

7. In the distance measuring device according to claim 6, the distance measuring device, characterized in that the reflectance of the reflecting surface is 1 / 3.

Images