JP2021143854A - Device and method for estimating angle of object position and radar device - Google Patents

Abstract

To estimate the angle of an object position with high accuracy as compared with prior art.SOLUTION: The present invention comprises: a distance estimation unit for Fourier-transforming a plurality of received signals having been received by a plurality of antennas to calculate a frequency spectrum, searching for and extract a distance that corresponds to at least one frequency that is a peak in the frequency spectrum; a mode vector calculation unit for calculating a mode vector per angle that corresponds to the distance, on the basis of the extracted distance or a prescribed distance; an angle estimation unit for executing a prescribed angle estimation process on the basis of a frequency corresponding to the extracted distance so as to calculate an angle spectrum that is a spatial frequency spectrum, and extracting an angle that is a peak in the angle spectrum so as to estimate the angle of the object position; and a position information calculation unit for calculating the object position on the basis of a set of the extracted distance and the angle estimated in correspondence to the distance.SELECTED DRAWING: Figure 1

Description

The present invention relates to an object position measuring device and a method for estimating, for example, an angle of an object position at a short distance from an object position angle estimating device.

An arrival direction estimation (arrival angle estimation) device and method according to a conventional example, in which a radar device called a so-called "radio wave sensor" is used to radiate radio waves from the radar device and then estimate the arrival angle of the radio waves reflected by an object. Is disclosed in, for example, Patent Document 1.

The arrival direction estimation device according to this conventional example has the following configuration in order to estimate the arrival direction of a signal more accurately than in the prior art. The arrival direction estimation device acquires a signal for each antenna based on a received signal received by an array antenna including a plurality of antennas, generates a correlation matrix based on the acquired signal for each antenna, and obtains the correlation matrix. A first arithmetic unit that generates an extended correlation matrix based on the Katri-Lao product using the antenna and generates a spatial average extended correlation matrix that is subjected to spatial averaging processing on the extended correlation matrix, and a first arithmetic unit generated by the first arithmetic unit. It is provided with a second calculation unit that calculates the arrival direction of the signal from the target included in the received signal based on the spatial average extended correlation matrix.

JP-A-2017-090229 Japanese Patent No. 6365251 Japanese Patent No. 5783693

Nobuyoshi Kikuma, "Adaptive Signal Processing by Array Antenna", Science and Technology Publishing, published on August 1, 2004 (Sections 9.1 to 10.2, pp. 173 to 202) Matsuo Sekine, "Radar Signal Processing Technology", Corona Publishing Co., Ltd., Institute of Electronics, Information and Communication Engineers, September 20, 1991 (Chapter 5, pp. 96-157)

However, in the arrival direction estimation device according to the conventional example, the angle estimation is performed under the condition that the reflected wave from the object is a wave from infinity (plane wave). Therefore, when detecting a short-distance object, there is a problem that the assumption does not hold and the angle estimation accuracy deteriorates.

An object of the present invention is to solve the above problems and to provide an object position angle estimation device and method capable of estimating an object position angle with higher accuracy as compared with the prior art, and a radar device. ..

The object position angle estimation device according to the present invention is
It is an angle estimation device for the object position that estimates the angle with respect to the object position as seen from the array antenna.
Distance estimation that calculates the frequency spectrum by Fourier transforming a plurality of received signals received by an array antenna including a plurality of antennas, and searches for and extracts a distance corresponding to at least one frequency that peaks in the frequency spectrum. Department and
A mode vector calculation unit that calculates a mode vector for each angle corresponding to the distance based on the extracted distance or a predetermined distance.
The angle spectrum, which is a spatial frequency spectrum, is calculated by executing a predetermined angle estimation process based on the frequency corresponding to the extracted distance, and the angle at which the peak is extracted in the angle spectrum is extracted to obtain the object position. It is provided with an angle estimation unit that estimates an angle.

Therefore, according to the object position angle estimation device or the like according to the present invention, the distance information obtained by the distance estimation performed before the angle estimation is used at the time of the angle estimation, so that the distance is shorter than that of the prior art. It is possible to improve the estimation accuracy of the angle estimation of the object position in.

It is a block diagram which shows the structural example of the radar apparatus 100 which concerns on Embodiment 1. FIG. It is the schematic for demonstrating the radar apparatus which has a receiving array antenna apparatus, and the reflected wave from the object P. It is a schematic diagram for demonstrating the arrival angle estimation method which concerns on a prior art example. It is a schematic diagram which shows _{the distance r 1} , r ₂ , ..., r _L between the object P and the receiving antenna 1-1-1-L in the radar apparatus 100 of FIG. FIG. 5 is a waveform diagram showing a transmission signal and a reception signal used in the radar device 100 of FIG. FIG. 5 is a plan view showing the relationship between the object P in the radar device 100 of FIG. 1 and the transmitting antenna 2 and the receiving antennas 1-1 to 1-L. It is the schematic which shows the frequency spectrum with respect to the distance R which shows an example of the frequency spectrum taken out by the radar apparatus 100 of FIG. It is a figure which shows the position coordinate Pp of the object P assumed in the radar apparatus 100 of FIG. 1, and the position coordinate of each receiving antenna 1-1-1-L. It is a flowchart which shows the angle estimation and position detection processing of an object executed by the processor 30 of FIG. It is a figure which shows the conversion from polar coordinates to Cartesian coordinates when calculating the detection position information in step S10 of FIG. FIG. 5 is a coordinate diagram showing a relationship between an object P and an antenna position (origin) in a radar device that calculates three-dimensional position information of an object according to the second embodiment. It is a schematic coordinate diagram which shows the arrangement of the receiving antenna 51 in the radar apparatus of FIG. FIG. 5 is a plan view showing the relationship between the object P and the receiving antennas 1-1 to 1-L in the radar device of FIG. 6 is a coordinate diagram showing conversion from polar coordinates to Cartesian coordinates when calculating three-dimensional position detection information in the radar device of FIG. 11. It is a figure for demonstrating the position detection process of the object which concerns on a modification. It is a schematic diagram for demonstrating the mode vector used in the arrival angle estimation method which concerns on a prior art example.

Hereinafter, embodiments according to the present invention will be described with reference to the drawings. The same or similar components are designated by the same reference numerals.

(Inventor's knowledge)
FIG. 2 is a schematic diagram for explaining a radar device having a receiving array antenna device and a reflected wave from the object P. In FIG. 2, the opening surfaces and reference points R of a plurality of L receiving antennas 1-l (l = 1, 2, ..., L) are juxtaposed on a straight line. As shown in FIG. 2, the received signals of the plurality of L receiving antennas 1-l of the radar device have a distance difference r _dl (l = 1, 2, ... Depending on the direction of arrival due to the reflected wave from the object P. , L) appears.

FIG. 3 is a schematic view for explaining the arrival angle estimation method according to the conventional example. Also in FIG. 3, the opening surfaces and the reference points R of the plurality of L receiving antennas 1-1 (l = 1, 2, ..., L) are juxtaposed on a straight line. In a radar device having a receiving array antenna device composed of a plurality of L receiving antennas 1-l (l = 1, 2, ..., L), as shown in FIG. 3, a plurality of L receiving antennas 1-l There is a phase difference between them. Here, in the far-field region where the distance between the radar device and the object P is sufficiently long and plane wave approximation is possible, a distance difference r _dl peculiar to the arrival angle θ occurs in the phase of each receiving antenna 1-l, which is sufficiently far away. Since the incoming wave from is assumed, the phase difference of each reflected wave received by each receiving antenna 1-l increases linearly. In a radar device having a receiving array antenna device, _{phase scanning is performed using this linearly increasing distance difference rdl} (phase difference), and the arrival angle or arrival direction of the reflected wave reflected by the object P is estimated. However, in a region belonging to the near field where the distance between the plurality of L receiving antennas 1-l of the radar device and the object P to be detected is short, the reflected wave from the object P depends not only on the arrival angle but also on the distance. The distance difference is r _dl (l = 1, 2, ..., L). Therefore, there is a problem that the estimation accuracy of the arrival angle deteriorates because the distance difference information between the plurality of L receiving antennas 1-l estimated only from the detected arrival angle does not match.

(Embodiment 1)
FIG. 4 is a schematic view showing _{distances r 1} , r ₂ , ..., R _L between the object P in the radar device 100 of FIG. 1 and the plurality of L receiving antennas 1-1 to 1-L.

In the present embodiment, in order to solve the above problem, the distance between the object P and the radar device is estimated before the angle estimation of the object position, and the object P and the plurality of objects P related to the reflected wave from the object P are estimated. _{The distance r l} (l = 1, 2, ..., L) between the L receiving antennas 1-l is measured. Then, when estimating the angle of the object position, in addition to the angle between the receiving antenna 1-l, the reference line (horizontal line) of the reference point R, and the reflected wave, the object P and the plurality of L related to the reflected wave It is characterized in that the estimation accuracy at the time of angle estimation is improved by performing phase scanning including _{the distance r l} between the receiving antennas 1 l.

FIG. 1 is a block diagram showing a configuration example of the radar device 100 according to the first embodiment.

In FIG. 1, the radar device 100 according to the embodiment is, for example, an FMCW (Frequency Modulation Continuous Wave) type radar device, and a plurality of L receiving antennas 1-l (l = 1, 2, ..., L; hereinafter, The same applies to l), the receiving antenna device 1, the plurality of L low noise amplifiers 11-l, the plurality of L mixers 12-l, and the plurality of L low frequency pass filters 13-. 1 and a plurality of L AD converters (ADCs) 14-l. Further, the radar device 100 further includes a transmitting antenna device 2 including one transmitting antenna, a local oscillator 15, a mixer 16, a bandpass filter (BPF) 17, a circulator 18, a power amplifier 19, and a signal. It includes a processing device 20 and a display unit 40. Here, the signal processing device 20 includes a received signal memory 21, a transmission signal generator 22, a memory 23 for storing data during signal processing, and a processor 30.

The transmission control unit 31 controls the transmission operation of the radar device 100 and outputs a transmission instruction signal to the transmission signal generator 22. In response to this, the transmission signal generator 22 generates one chirped pulse signal and outputs it to the mixer 16. The mixer 16 mixes the input 1-charp pulse signal with the local oscillation signal from the local oscillator 15 to up-convert it to a predetermined transmission frequency to generate a transmission signal, and passes through a band having a predetermined transmission frequency band. It is output to the power amplifier 19 via the filter (BPF) 17 and the circulator 18. After amplifying the input transmission signal, the power amplifier 19 radiates the radio wave of the transmission signal from the transmission antenna device 2 in a predetermined radiation direction toward the object to be detected.

In FIG. 1, the opening surfaces of a plurality of L receiving antennas 1-l (l = 1, 2, ..., L) are juxtaposed on a straight line to form a receiving antenna device 1 which is a straight array antenna device. .. The transmission signal reflected by the object P is received by the receiving antenna device 1 including the plurality of receiving antennas 1-l (l = 1, 2, ..., L) described above. The plurality of L received signals received by each receiving antenna 1-l are low-noise amplified by the low-noise amplifier 11-l and then input to the mixer 12-l. The mixer 12-l mixes the input multiple L received signals with the transmitted signal from the circulator 18, and then passes the low-pass filter (LPF) 13-l to pass the plurality of L baseband signals. obtain. The plurality of L baseband signals are AD-converted into digital baseband signals by the AD converter 14-l, and then sequentially stored in the received signal memory 21 in the signal processing device 20 in chronological order.

In the radar device 100 of FIG. 1, a low-pass filter 13-l is used by using a direct conversion system, but the present invention is not limited to this, and when a heterodyne conversion system is used, an intermediate frequency signal having an intermediate frequency is used. A bandpass filter (BPF) that allows only to pass through may be used.

For example, in the case of the FMCW type radar device 100, the frequency difference between the transmission signal and the reception signal increases or decreases in proportion to the distance between the object and the radar device 100, and this frequency difference becomes a variation component of the distance. Further, in the case of the FCM (Fast Chirp Modulation) method, the phase difference (phase shift) between the transmitted signal and the received signal increases or decreases in proportion to the distance between the object and the radar device 100, so that the beat signal fluctuates due to this phase difference. The component becomes the variable component of the distance. Further, when reflected by an object, the received signal is affected by the speed of the object, and the frequency difference between the pulses increases or decreases in proportion to the relative speed (Doppler frequency) between the object and the radar device 100. Therefore, this pulse The fluctuation component of the beat signal due to the frequency difference between them becomes the fluctuation component of the velocity. When there are a plurality of targets having different relative speeds and distances, each receiving antenna 1-l receives a plurality of reflected waves having different phase shift amounts and Doppler shift amounts, and a base obtained from each mixer 12-l. The band signal will include various components including a beat signal corresponding to each object.

The processor 30 is a computer such as a digital computer, and includes a transmission control unit 31, a distance estimation unit 32, a mode vector calculation unit 33, a mode vector memory 34, an angle estimation unit 35, and a position information calculation unit 36. To be equipped.

Hereinafter, the angle estimation and position detection processing of the object executed by the processor 30 will be described with reference to FIGS. 5 to 10.

FIG. 5 is a waveform diagram showing a transmission signal and a reception signal used in the radar device 100 of FIG. Further, FIG. 6 is a plan view showing the relationship between the object P in the radar device 100 of FIG. 1 and the transmitting antenna 2 and the receiving antennas 1-1 to 1-L.

Here, consider the FMCW type radar device 100. For the sake of simplicity, it is assumed that the reflected signal is one wave and the receiver thermal noise is sufficiently small to be ignored. As shown in FIG. 5, the sweep time of 1 chirped pulse signal and T _SW, the center frequency is f _c, the frequency bandwidth is B, a delay time from transmission to reception at each receiving antenna 1-l tau _l Let (l = 1, 2, ..., L). Here, the delay time _{τ l (l = 1,2, ...} , L) , as shown in FIG. 6, the receiving antenna the radio wave radiated from the transmitting antenna 2 is separated by a distance r _l is reflected from an object P to 1-l receives a delay time required for the radio wave propagation distance r _l. The distance r _l is the distance from the transmitting antenna 2 to the object P (common) is the distance obtained by adding the distance from the object P to the receiving antennas 1-1 to 1-L. At this time, l-th radar receiver signal x _l, which is converted into a baseband signal to be analyzed is expressed by the following equation.

（１）

(1)

here,

（ｌ＝１，２，…，Ｌ）
である。

(L = 1, 2, ..., L)
Is.

_{Here, the frequency spectrum X l} of the received signal corresponding to each receiving antenna 1-l obtained by Fourier transforming the _{received signal data x l} related to one charp pulse signal (time signal) is expressed by the following equation.

（２）

(2)

Here, l = 1, 2, ..., L.

FIG. 7 is a diagram showing a frequency spectrum with respect to a distance R showing an example of the frequency spectrum taken out by the radar device 100 of FIG. The frequency data that becomes the frequency peak and the distance information thereof are extracted from the frequency spectrum X _l. Here, when the maximum value is taken out because of the reflected signal of one wave, the frequency data S _l of the following equation formed from the delay time τ _{l corresponding to the distance l l} can be taken out.

Here, max [・] is a function indicating the maximum value of the argument.

The frequency data y to be used to calculate the spatial frequency spectrum removed from the frequency spectrum X _l obtained from the receiving antennas 1-l is expressed by the following equation.

（４）

(4)

Here, [・] ^T represents transpose.

FIG. 8 is a diagram showing the position coordinates Pp of the object P assumed in the radar device 100 of FIG. 1 and the position coordinates of each of the receiving antennas 1-1 to 1-L. Hereinafter, a method of generating a mode vector used for angle estimation at a short distance will be described with reference to FIG.

The following procedure is used to generate a mode vector containing distance and angle information.
(1) Of the plurality of L receiving antennas 1-1 to 1-L, any one of the receiving antennas is defined as the reference point R. In FIG. 8, the position of the receiving antenna 1-1 is set as the reference point R.
(2) determining the distance r _l measured in advance distance estimation, the position coordinates Pp envisaged object P from an angle for scanning (reflected wave source) a (x, y).
(3) From the assumed position coordinates Pp (x, y) _{of the object P and the position coordinates (xl} , y _{l ) of each of the remaining receiving antennas 1-l, the distance R l} between the receiving antenna 1-l and the object P. Is calculated, and the mode vector is calculated using the following formula.

Here, the position coordinates Pp (x, y) of the object P are expressed by the following equation.

x = r _ref・ cos (90 ° −θ) = r _ref・ sinθ
y = r _ref・ sin (90 ° −θ) = r _ref・ cosθ

Here, θ is the scanning angle, and r _{ref is the distance r l} (l = 1, 2, ..., L) with respect to any one of the receiving antennas 1-l obtained by distance estimation.

_{The distance R l} (l = 1, 2, ..., L) between each receiving antenna 1-1 to 1-L and the object P is expressed by the following equation.

（５）

(5)

The mode vector a (r _ref , θ) used for angle estimation is expressed by the following equation.

（６）

(6)

Here, k (= 2π / λ) represents the wave number, and λ is the wavelength.

FIG. 8 shows a case where the reference point R is set to the receiving antenna 1-1, the assumed position coordinates Pp (x, y) of the object P are calculated with _{r ref} = r _{1, and each receiving antenna 1-1. The distance R l} between ~ 1-L and the assumed position coordinates Pp (x, y) of the object P can be obtained, and the mode vector a (r _ref , θ) can be generated.

The above-mentioned calculation of the spatial frequency spectrum (angle spectrum) can be applied to any known algorithm for performing angle scanning, and is, for example, a Beamformer method, a Capon method, a MUSIC (Multiple SIGNAL Classification) method, and an LP (Linear Prediction) method. It can be applied to various methods for estimating the direction of arrival (see, for example, Non-Patent Document 1). Here, a method of calculating the spatial frequency spectrum (angle spectrum) when a predetermined algorithm is used will be described below.

For example, in the Beamformer method, the spatial frequency spectrum (angle spectrum) _PBF (θ) can be calculated using the following equation.

（７）

(7)

Here, R is a correlation matrix expressed by the following equation.

R = yy ^H (8)

Here, [・] ^H represents a complex conjugate transpose.

When a _{plurality of incoming waves are included in the distance r l} , the calculation performance deteriorates when the spatial frequency spectrum (angle spectrum) is calculated using the following algorithm. Therefore, for example, it is desirable to use the spatial averaging process also mentioned in Patent Document 1.

In the Capon method, the spatial frequency spectrum (angle spectrum) P _CP (θ) can be calculated using the following equation.

（９）

(9)

Here, [・] ^-1 represents an inverse matrix operator.

Further, in the linear prediction method, the spatial frequency spectrum (angle spectrum) _PLP (θ) can be calculated by using the following equation.

（１０）

(10)

Here, the vectors w and t are expressed by the following equations.

w = R ^-1 t (11)
t = [1,0, ..., 0] ^T (12)

Further, in the MUSIC method, the spatial frequency spectrum (angle spectrum) _{PLP (} θ) can be calculated by using the following equation. In the MUSIC method, it is necessary to change the number of eigenvectors used in the calculation according to the number of incoming waves, and a known wavenumber such as AIC (Akaike Information Criterion) or MDL (Minimum Description Algorithm) (see, for example, Patent Document 3). The output value of the incoming wavenumber calculated by the estimation algorithm is used.

（１３）

(13)

Here, E _N is expressed by the following equation.

_{E N = [e M + 1} , ..., e L] (14)

Here, e is an eigenvector of the correlation matrix R, and M is an integer in the range of 1 ≦ M ≦ L-1.

Using the angle spectrum P (θ), which is the spatial frequency spectrum calculated as described above, the angle θ _l (l = 1, 2, ..., L) at which the angle spectrum P (θ) peaks is extracted. When there is only one incoming wave, the angle θ _max that takes the maximum value of the angle spectrum is taken out as shown in the following equation.

（１５）

(15)

_{When the angles θ l} corresponding to a plurality of peaks are taken out, it can be searched by using a signal detection method such as CFAR (Constant False Alarm Rate) processing (see, for example, Non-Patent Document 2). It should be noted that the _{distance r l} corresponding to a plurality of peaks can be extracted even in the distance estimation by the above-mentioned method.

The detection position information in the radar device 100 is generally expressed in polar coordinates, but it is conceivable that the polar coordinates are converted into orthogonal coordinates and expressed. FIG. 10 is a diagram showing conversion from polar coordinates to Cartesian coordinates when calculating the detection position information in step S10 of FIG. 9, which will be described later. As an example of the detection position information output process, when _{the two-dimensional information of the distance r and the angle θ max} is converted into the two-dimensional Cartesian coordinates as shown in FIG. 10, the conversion formula is expressed by the following formula.

x = r · cos (90 ° −θ _max ) = r · sinθ _max
y = r · sin (90 ° −θ _max ) = r · cosθ _max
(16)

FIG. 9 is a flowchart showing an angle estimation and position detection process of an object executed by the processor 30 of FIG. In the angle estimation and position detection processing of the objects in FIG. 9, it is assumed that a plurality of N objects exist.

In FIG. 9, the distance estimation unit 32 _{reads and acquires the radar reception signal x l} (l = 1, 2, ..., L) of the equation (1) from the reception signal memory 21 (S1), and one chap pulse signal is obtained. By performing a fast Fourier transform (FFT) on the received data of _{the radar received signal x l} transmitted by the object P after being transmitted from the radar device 100, _{the frequency spectrum X l} (l = 1, 2, ..., L) was calculated (S2), the frequency spectrum _X and the frequency data _{S l} of the formula (3) in which the frequency reaches a peak (maximum value) from _l, the distance information _{r l} corresponding to the frequency data _{S l} N pairs of and are searched and extracted (S3).

Next, 0 is set in the parameter n (S4), and it is determined whether or not the parameter n is n ≧ N in comparison with the predetermined object numerical value N (S5). If, the process proceeds to step S6. In step S6, the parameter n is incremented by 1, and then the process proceeds to step S7.

The mode vector calculation unit 33 uses the equation (6) to generate a mode vector a (r, θ) for each angle using the phase data corresponding to _{the distance information r l, and stores the mode vector a (r, θ) in the mode vector memory 34 (} S7). Next, the angle estimation unit 35 calculates an angle spectrum, which is a spatial frequency spectrum, by performing a predetermined angle estimation process based on the frequency data y of the equation (4) corresponding to _{the distance information l (S8).} The peak angle information θ _l is extracted from the calculated angle spectrum and stored in the memory 23 in correspondence with the distance information _l (S9), and the process returns to step S5.

Next, the position information calculation unit 36 calculates _{the detection position information (xl} , y _l ) of the object P from the information of the set of _{the distance information l l} and the angle information θ _l stored in the memory 23, and the display unit 40. Is output to and displayed (S10), and the angle estimation and position detection processing of the object P is completed.

In the present embodiment, the detection position information ( _xl , y _l ) of the object P is output to the display unit 40 for display, but the present invention is not limited to this, and the object P is output to a printer for printing. Alternatively, it may be output to a voice synthesizer and output as a voice signal for notification.

(Embodiment 2)
In the above first embodiment, the position of the object in the two-dimensional coordinates on the plane space is detected, but the present invention is not limited to this, and the present invention is also applied to the object position detection in the three-dimensional space to which the elevation angle information is added. You may. In the following second embodiment, the differences between the radar device 100 applied to the object position detection in the three-dimensional space and the first embodiment will be described below.

FIG. 11 is a coordinate diagram showing the relationship between the object P and the antenna position (origin) in the radar device for calculating the three-dimensional position information of the object according to the second embodiment. Further, FIG. 12 is a schematic coordinate diagram showing the arrangement of a plurality of receiving antennas 51 in the radar device of FIG.

In the second embodiment, as an example of the coordinate system, a case where a reflected wave from an object P located at a distance r, an azimuth angle θ, and an elevation angle φ as shown in FIG. 11 arrives is considered. Further, as shown in FIG. 12, the receiving antenna device 1 which is an array antenna is configured to include a plurality of receiving antennas 51 juxtaposed on the XZ plane.

FIG. 13 is a plan view showing the relationship between the object P and the receiving antenna 51 (1-1 to 1-L) in the radar device of FIG. In the second embodiment, the assumed position coordinates of the object P, which is the reflected wave source, are obtained from the distance r measured by the distance estimation performed in advance and the scanning angles θ and φ for scanning, and the assumed position coordinates of the object P and each reception are obtained. The mode vector a (r, θ, φ) can be calculated from each coordinate of the antenna 51 (1-1 to 1-L) by using the following equation.

（１７）

(17)

Here, r _l (l = 1, 2, ..., L) is between the coordinates of each receiving antenna 51 (1-1 to 1-L) and the position coordinates determined from the distance r and the scanning angles θ and φ. Represents the length (ie, the distance between the antenna coordinates and the coordinates to be scanned).

Also in the second embodiment, the calculation of the spatial frequency spectrum (angle spectrum) can be applied to any algorithm such as the Beamformer method, the Capon method, and the MUSIC method, which performs angular scanning.

For example, in the Beamformer method, the spatial frequency spectrum (angle spectrum) _PBF (θ, φ) can be calculated using the following equation.

（１８）

(18)

When a _{plurality of incoming waves are included in the distance r l} , the calculation performance deteriorates when the spatial frequency spectrum (angle spectrum) is calculated using the following algorithm. Therefore, for example, it is desirable to use the spatial averaging process also mentioned in Patent Document 1.

In the Capon method, the spatial frequency spectrum (angle spectrum) P _CP (θ) can be calculated using the following equation.

（１９）

(19)

Further, in the linear prediction method, the spatial frequency spectrum (angle spectrum) _PLP (θ) can be calculated by using the following equation.

（２０）

(20)

Further, in the MUSIC method, the spatial frequency spectrum (angle spectrum) _PMU (θ) can be calculated by using the following equation. In the MUSIC method, it is necessary to change the number of eigenvectors used in the calculation according to the number of incoming waves, and a known wavenumber such as AIC (Akaike Information Criterion) or MDL (Minimum Description Algorithm) (see, for example, Patent Document 3). The output value of the incoming wavenumber calculated by the estimation algorithm is used.

（２１）

(21)

Also in the radar device 100 according to the second embodiment, the detected position information is generally expressed in polar coordinates, but it is conceivable that the polar coordinates are converted into orthogonal coordinates and expressed. FIG. 14 is a coordinate diagram showing conversion from polar coordinates to Cartesian coordinates when calculating three-dimensional position detection information in the radar device of FIG. As an example of the detection position information output processing, when _{the three-dimensional phase detection information of the distance r and the angles θ max} and φ _max is converted into the three-dimensional Cartesian coordinates as shown in FIG. 14, the conversion formula is expressed by the following formula. Will be done.

x = r · cosφ _max · cos (90 ° −θ _max ) = r · cosφ _max · sinθ _max
y = r · cosφ _max · sin (90 ° −θ _max ) = r · cosφ _max · cosθ _max
z = r · sinφ _max
(22)

By using the above mathematical formula and similarly executing the process of FIG. 9 of the first embodiment in the second embodiment, the three-dimensional position coordinates of the object P can be detected.

(Modification example)
FIG. 15 is a diagram for explaining the position detection process of the object according to the modified example. As shown in FIG. 15, a predetermined distance range is divided into an arbitrary plurality of sections, a representative distance such as a median or an average value is set in each section, and the representative distance is used for the distance data included in each section. By using it for angle estimation, it becomes unnecessary to calculate the mode vector using the distance information obtained from the distance estimation, and the total calculation time can be shortened. That is, for each distance when the mode vector a (r, θ) is calculated in advance, the distance range to be detected is divided into a plurality of distance sections, and each of the divided distance sections is, for example, the median value or the average. A predetermined representative distance, which is a value, is used. Therefore, the estimation accuracy can be improved as compared with the conventional angle estimation that does not refer to the distance information.

According to this modification, since the mode vector corresponding to the representative distance can be prepared in advance, the process of step S7, which is calculated each time the angle is estimated, can be omitted. However, the estimation accuracy is deteriorated as compared with the case of performing the angle estimation using the distance information corresponding to the peak frequency.

In the above embodiment, when estimating only the angle of the object, the process of step S10 in FIG. 9 may be omitted.

(Effects of Embodiments and Modifications)
As described above, according to the object position angle estimation device or the like according to the present embodiment, the distance information obtained by the distance estimation performed before the angle estimation is used at the time of the angle estimation to compare with the conventional technique. Therefore, the estimation accuracy of the angle estimation of the object position at a short distance can be improved.

(Supplementary information on the difference from the arrival angle estimation method according to the conventional example)
FIG. 16 is a schematic diagram for explaining a mode vector used in the arrival angle estimation method according to the conventional example. In FIG. 16, the distance from the reference point R of each receiving antenna 1-l (l = 1, 2, ..., L) is _rl (l = 1, 2, ..., L).

(Mode vector used in the angle estimation of the far field assumption of the conventional method)
For example, in the radar device disclosed in Patent Document 1, a linear array antenna is used, and the phase difference of each antenna is calculated based on the arrangement of the antennas (element spacing) according to the scanning angle, and the mode at this time. The vector a (θ) is expressed by the following equation (see paragraph 0061 and equation 22 of Patent Document 1).

（２３）

(23)

Further, for example, in the radar device disclosed in Patent Document 2, the mode vector a (θ) when an evenly spaced linear array antenna is used is expressed by the following equation (see paragraph 0143 and Equation 16 of Patent Document 2).

（２４）

(24)

This equation (24) corresponds to the above-mentioned equation (23) normalized by the first term, and the element spacings between the receiving antennas are all the same distance d.

The mode vector a (θ) of the equations (23) and (24) is clearly completely different from the mode vector a (r, θ) of the equation (6) according to the present embodiment. In the radar devices according to Patent Documents 1 and 2, the results of distance and speed estimation are used in the formula defined as importance, and the closer to the radar device and the faster the moving speed, the earlier the angle is estimated. There is. That is, while the distance information and the speed information are used to calculate the formula for determining the importance, in the present embodiment, the distance information is used for the angle estimation of the post-processing.

As described in detail above, the object position angle estimation device and method according to the present invention are object position angle estimation devices and methods capable of performing object position angle estimation with higher accuracy than conventional techniques. It is useful for radar devices using.

1 Receiving antenna device 1-1-1-L Receiving antenna 2 Transmitting antenna device 11-1 to 11-L Low noise amplifier 12-1 to 12-L Mixer 13-1 to 13-L Low-pass filter (LPF)
14-1 to 14-LA AD converter (ADC)
15 Local oscillator 16 Mixer 17 Bandpass filter (BPF)
18 Circulator 19 Power amplifier 20 Signal processing device 21 Received signal memory 22 Transmission signal generator 23 Memory 30 Processor 31 Transmission control unit 32 Distance estimation unit 33 Mode vector calculation unit 34 Mode vector memory 35 Angle estimation unit 36 Position information calculation unit 40 Display Part 51 Receiving antenna P Object Pp Position coordinates of object P R Reference point S Equal phase wave plane

Claims (8)

It is an angle estimation device for the object position that estimates the angle with respect to the object position as seen from the array antenna.
Distance estimation that calculates the frequency spectrum by Fourier transforming a plurality of received signals received by an array antenna including a plurality of antennas, and searches for and extracts a distance corresponding to at least one frequency that peaks in the frequency spectrum. Department and
A mode vector calculation unit that calculates a mode vector for each angle corresponding to the distance based on the extracted distance or a predetermined distance.
The angle spectrum, which is a spatial frequency spectrum, is calculated by executing a predetermined angle estimation process based on the frequency corresponding to the extracted distance, and the angle at which the peak is extracted in the angle spectrum is extracted to obtain the object position. It is equipped with an angle estimation unit that estimates an angle.
An angle estimation device for the position of an object. The angle estimation device for the object position further
It includes a position information calculation unit that calculates the position of the object based on the set of the extracted distance and the angle estimated corresponding to the distance.
The angle estimation device for an object position according to claim 1. The position information calculation unit calculates the position of an object represented by two-dimensional coordinates or three-dimensional coordinates.
The angle estimation device for an object position according to claim 2. When the mode vector calculation unit calculates a mode vector for each angle corresponding to the distance based on the predetermined distance, the distance divides the distance range to be detected into a plurality of distance sections, and the division divides the distance range to be detected. A predetermined representative distance is used for each distance section.
The angle estimation device for an object position according to any one of claims 1 to 3. The object position angle estimation device is an object position angle estimation device for an FMCW (Frequency Modulation Continuous Wave) type radar device.
The received signal is a signal received by reflecting a transmitted signal which is a chirped pulse signal by the object.
The angle estimation device for an object position according to any one of claims 1 to 4. The angle estimation device for an object position according to any one of claims 1 to 5 is provided.
Radar device. It is an angle estimation method of the object position that estimates the angle with respect to the object position seen from the array antenna.
A step of calculating a frequency spectrum by Fourier transforming a plurality of received signals received by an array antenna including a plurality of antennas, and searching for and extracting a distance corresponding to at least one frequency peak in the frequency spectrum. ,
A step of calculating a mode vector for each angle corresponding to the distance based on the extracted distance or a predetermined distance, and
The angle spectrum, which is a spatial frequency spectrum, is calculated by executing a predetermined angle estimation process based on the frequency corresponding to the extracted distance, and the peak angle in the angle spectrum is extracted to obtain the object position. Including steps to estimate the angle,
How to estimate the angle of the object position. The method for estimating the angle of the object position further
A step of calculating the object position based on a set of the extracted distance and an angle estimated corresponding to the distance.
The angle estimation method for an object position according to claim 7.

Images