JP3518363B2 - FMCW radar device, recording medium, and vehicle control device - Google Patents

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is used in collision prevention of a moving body, traveling by following a fixed distance, and the like, and an FMCW for detecting the relative speed and distance to a target existing outside the moving body by transmitting and receiving radar waves. The present invention relates to an FMCW radar device using an output of a radar device, a storage medium, and a vehicle control device.

[0002]

2. Description of the Related Art a) Conventionally, in an FMCW radar device, a transmission signal which is frequency-modulated by a triangular-wave modulation signal and whose frequency gradually increases or decreases is transmitted as a radar wave, and a radar wave reflected by a target is received. At the same time, a beat signal is generated by mixing the received signal with the transmitted signal. Then, the frequency of the beat signal (beat frequency) is specified for each section of the rising part where the frequency of the transmission signal increases and the falling part where the frequency decreases by using a signal processor, and the specified rising part. Beat frequency (upstream beat frequency) fb1 and falling beat frequency fb2
Based on (downstream beat frequency), the following equations (B), (C)
Is used to calculate the distance D to the target and the relative speed V.

V = (C / (4 * f0)) * (fb2-fb1) (B) D = (C / (8 * ΔF * fm)) * (fb1 + fb2) (C) In addition, ΔF Is the frequency displacement width of the transmission signal (frequency displacement width), f0 is the center frequency of the transmission signal, 1 / fm is the time required for one cycle of modulation (that is, fm is the triangular wave repetition frequency), and C is the speed of light.

Therefore, when the measurement is performed using the FMCW radar device, for example, changes in the frequencies of the transmission signal T and the reception signal R shown in FIG. 14 can be obtained according to the relationship between the FMCW radar device and the target. Specifically, as shown in FIG. 14A, the moving speed of the moving body to which the radar device is attached and the target that reflects the radar wave are equal (relative speed V =
In the case of 0), the radar wave reflected on the target is delayed by the time required to make a round trip to and from the target, and therefore the graph of the reception signal R is a graph obtained by shifting the graph of the transmission signal T along the time axis. , Upstream beat frequency fb1 and downstream beat frequency f
It is equal to b2 (fb1 = fb2).

On the other hand, as shown in FIG. 14 (b), when the moving speed with respect to the target is different (relative speed V ≠ 0), the radar wave reflected on the target is further changed to the relative speed V with respect to the target. As a result, the graph of the reception signal R is a graph obtained by shifting the graph of the transmission signal T along the frequency axis by the amount of the Doppler shift due to the relative speed V, and thus the upbeat frequency fb1 and the downbeat frequency fb2. Which is different from (fb1 ≠ fb2).

Therefore, the distance D to the target and the relative speed V can be calculated based on the ascending beat frequency fb1 and the descending beat frequency fb2. b) Also, in recent years,
As a technique for distinguishing a moving object from a stationary object using the FMCW radar device described above, the following Japanese Unexamined Patent Publication (Kokai) No. 7-98375.
And Japanese Patent Application Laid-Open No. 7-191133 have been proposed.

This technique uses the physical principle that a stationary object approaches -VB if the vehicle travels at a speed VB. Specifically, for example, when the direction approaching the FMCW radar device is positive, when the vehicle speed is represented by -VB,
Since the speed of the stationary object seen from the FMCW radar device is VB, the difference between the up beat frequency fb1 and the down beat frequency fb2 is given by the following equation (D).

(Fb2-fb1) = (4 * VB * f0) / C (D) When this is frequency-analyzed by a well-known Fourier transform and displayed as a spectrum, the beat signal of the rising portion including the upbeat frequency fb1 is obtained. 1 (upstream beat signal spectrum) and the falling beat signal spectrum (downstream beat signal spectrum) including the downstream beat frequency fb2 are shown in FIG.
5 (a).

At this time, if the vehicle speed VB is known, the downlink beat signal spectrum is shifted by the frequency of (fb2-fb1), as shown in FIG. 15 (b). Here, when the target object to be measured is a stationary object, that is, when the target object is approaching at the own vehicle speed VB, the spectrum frequencies of the target object and the target object are the same. Whether or not the target is a stationary object can be determined based on the state of the spectra of the upbeat signal and the downbeat signal shown in (b).

[0010]

However, according to the above-mentioned technique, the spectrum shift amount (frequency shift amount) is detected from the vehicle speed sensor or the like based on the principle that a stationary object approaches -VB when the vehicle speed is VB. There is a problem as shown below, which is merely a technique of calculating from the obtained own vehicle speed VB and utilizing the calculation result to simply distinguish between a moving object and a stationary object.

The frequency shift amount can be basically obtained from the vehicle speed VB detected by, for example, a vehicle speed sensor. However, as shown in FIG. 16, when the vehicle speed VB is calculated by another vehicle-mounted computer. The actual vehicle speed (actual vehicle speed T
VB) is delayed, and there are other causes of error such as an error of the sensor itself and a sloppy signal. for that reason,
If the frequency shift amount is simply calculated from the own vehicle speed VB, the frequency shift amount may not be accurate, resulting in incorrect determination.

That is, an accurate vehicle speed cannot be obtained due to a response delay or an error of the vehicle speed sensor, and therefore, an appropriate frequency shift amount cannot be calculated, so that there is a problem that the stationary object cannot be accurately determined. It was The present invention has been made to solve the above-mentioned problems, and for example, an FMC that can obtain an actual vehicle speed closer to an actual vehicle speed by correcting the own vehicle speed detected by a vehicle speed sensor.
An object of the present invention is to provide a W radar device, a recording medium, and a vehicle control device.

[0013]

According to a first aspect of the present invention, there is provided a transmission means for generating a transmission signal having a predetermined modulation width, the frequency of which gradually increases and decreases periodically, and transmitting it as a radar wave.
Receiving means for receiving the radar wave reflected by the target to generate a reception signal, mixing the reception signal with the transmission signal to generate a beat signal, and increasing the frequency of the transmission signal While creating an upstream spectrum from the upstream beat signal at the time of modulation, a spectrum creating means for creating a downstream spectrum from the downstream beat signal at the time of downstream modulation in which the frequency of the transmission signal drops, and for comparing the peaks of both spectra A frequency shift amount setting means for setting a frequency shift amount based on the vehicle speed measured by the speed measuring means, and at least one of the both spectra is shifted by the frequency shift amount, and a corresponding peak of the both spectra is obtained. In the FMCW radar device comprising: a detection unit that compares the two with each other to detect the moving state of the target, When setting the shift amount, a plurality of shift amount setting means for setting a plurality of frequency shift amount in consideration of measurement error, for each of the upstream spectrum and the downlink spectrum corresponding to each of the set frequency shift amount, Evaluation means for evaluating the degree of spectrum matching, based on the evaluation result, determining means for determining the highest frequency shift amount of the spectrum matching degree, using the determined frequency shift amount, by the speed measuring means Estimating means for correcting the measured own vehicle speed and estimating the actual own vehicle speed,
The gist is an FMCW radar device characterized by including the following.

In the present invention, the transmission means generates a transmission signal whose frequency is gradually increased and decreased by, for example, a triangular wave modulation signal and transmits it as a radar wave, and the reception means transmits the radar wave reflected by the target. Is received to generate a reception signal, and the reception signal is mixed with a transmission signal to generate a beat signal.

Then, the spectrum creating means has a plurality of spectrum peaks as shown in FIG. 8, for example, from the upstream beat signal at the time of upstream modulation in which the frequency of the transmission signal increases by frequency analysis by the well-known Fourier transform. In addition to creating the upstream spectrum, a downstream spectrum having a plurality of spectrum peaks is similarly created from the downstream beat signal at the time of downstream modulation in which the frequency of the transmission signal decreases.

Next, the frequency shift amount setting means sets the frequency shift amount based on the own vehicle speed measured by the speed measuring means (for example, a wheel speed sensor or a vehicle speed sensor). Next, the detection means shifts at least one of the peaks of the upstream spectrum and the peaks of the downstream spectrum by a predetermined frequency, and shifts the frequency of the downlink spectrum to the left side of FIG. 8, for example, so as to match the downlink spectrum with the upstream spectrum. By moving by the amount, each peak is compared, and the moving state of the target such as a stationary object or a moving object is detected.

In particular, in the present invention, when the frequency shift amount is set, the measurement error (indicating, for example, the signal shown in FIG. 16, the delay due to the filter, the error in the sensor, etc.) is measured by the plural shift amount setting means. Considering the above, multiple frequency shift amounts are set. That is, conventionally, the frequency shift amount (fb2-fb1) is set based on the above-mentioned formula (D), but in the present invention, not only a single frequency shift amount but also the following formula (E), for example, As shown in, for example, another frequency shift amount is set within a predetermined range (± Dv) on both sides of the basic frequency shift amount.

(Fb2-fb1) = (4 * VB * f0) / C (D) (fb2-fb1) = (4 * (VB ± Dv) * f0) / C (E) Next, evaluation means Thus, the spectrum matching degree is evaluated for each of the upstream spectrum and the downstream spectrum corresponding to each frequency shift amount. For example, by shifting the downlink spectrum by a certain frequency shift amount, in this state, each peak of the uplink spectrum and each peak of the downlink spectrum,
For example, the degree of matching in amplitude and phase (spectral matching degree) is evaluated. Specifically, as will be described later, for each frequency shift amount, all spectrum peaks to be evaluated are evaluated and summed, and the evaluation value (spectrum total sum Sum1) for each frequency shift amount is calculated. Obtained and evaluated by the magnitude of this value.

Then, based on the evaluation result, the determining means determines that the frequency shift amount having the highest spectrum matching degree is the true frequency shift amount which is less affected by the error of the vehicle speed sensor. For example, the total sums of the spectra are compared with each other, and the frequency shift amount corresponding to the smallest value is determined to be the true frequency shift amount. That is, it is considered that the larger the deviation between the peaks of the upstream spectrum and the respective peaks of the downstream spectrum, the greater the influence of various errors. Therefore, here, the higher the degree of coincidence between the two spectra, the greater the influence of errors. Select as less.

Then, the estimating means applies the true frequency shift amount to a predetermined arithmetic expression such as the expression (A) to correct the own vehicle speed measured by the speed measuring means. Estimate the actual vehicle speed. As a result, for example, the actual vehicle speed more accurate than the vehicle speed obtained by the normal vehicle speed sensor can be obtained.

As described above, when the vehicle speed is calculated by another vehicle-mounted computer, there is a delay between the vehicle speed obtained by the vehicle speed sensor and the actual vehicle speed due to the influence of communication delay and filtering. Although there is an error in the vehicle speed sensor itself, as in the present invention, the frequency shift amount has a range in advance (that is, a plurality of frequency shift amounts are set), and the optimum frequency shift amount is selected from them. By using the same, the error of the vehicle speed sensor can be eliminated and a more accurate actual vehicle speed can be obtained.

(2) The invention according to claim 2 is the FMCW radar device according to claim 1, characterized in that the estimating means estimates the actual vehicle speed based on the following equation (A). And TVB = TSn * C / (4 * f0) (A) where TVB is the actual vehicle speed, TSn is the determined frequency shift amount, f0 is the center frequency of the transmission signal, and C is the speed of light. The estimation means of the invention of Item 1 is illustrated. This formula (A) is the same as the formula (D) in the above formula (fb2−
fb1) is modified by replacing TSb with Vn and TVB with VB. With this, the actual vehicle speed can be calculated.

(3) The invention of claim 3 is characterized in that the plurality of frequency shift amounts are set based on the actual vehicle speed estimated by the estimating means. The FMCW radar device of is the gist.

For example, when the frequency shift amount is set by using the vehicle speed obtained from the wheel speed sensor or the vehicle speed sensor, the error of the vehicle speed caused by the characteristics of the sensor is greatly affected. By using the actual vehicle speed, the burden of the next calculation processing can be reduced. This also improves the accuracy of the next actual vehicle speed.

Considering, for example, the case where the current own vehicle speed is used to correct the next own vehicle speed and a plurality of frequency shift amounts are set using the corrected values, the own vehicle speed is considerably accurate. Therefore, the shift width (± Dv) in a plurality of frequency shift amounts can be reduced, that is, the number of frequency shift amounts can be reduced (for example, Sn ± 5 can be reduced to Sn ± 3).

(4) In the invention of claim 4, the actual relative acceleration is obtained from the difference between the actual vehicle speed obtained last time and the actual vehicle speed obtained this time, and the actual relative acceleration is used for the next automatic vehicle acceleration. The FMCW radar device according to claim 3, wherein a vehicle speed is predicted, and the predicted next vehicle speed is used to set the plurality of next frequency shift amounts.

The present invention exemplifies the invention according to claim 3, wherein the estimated actual vehicle speed is used to set the frequency shift amount with less calculation processing load and less error. can do. Specifically, the actual relative acceleration is obtained by dividing the difference between the previously obtained actual vehicle speed and the currently obtained actual vehicle speed by the calculation cycle.
Therefore, using the actual relative acceleration, for example, the value obtained by correcting the vehicle speed obtained from the vehicle speed sensor can be used as, for example, V in the equation (E).
By using it in place of B, the next own vehicle speed can be obtained more accurately.

That is, since the error of the own vehicle speed VB is reduced by this, an appropriate frequency shift amount can be set, and as a result, the actual own vehicle speed can be calculated accurately. Further, in the case of the present invention, since the error becomes small, the width (± ΔDv) of the predetermined range in the following expression (E) can be set small, and the calculation processing can be reduced.

(5) According to a fifth aspect of the invention, when the frequency shift amount is set on the basis of the vehicle speed measured by the speed measuring means, the basic frequency shift amount and the basic frequency shift amount are used. The FMCW radar device according to any one of claims 1 to 4, wherein a frequency shift amount deviated by a predetermined amount is set.

In the present invention, the basic frequency shift amount and the frequency shift amount deviated by a predetermined amount from the basic frequency shift amount are set. For example, as shown in the following formula (E) shown in claim 1, in the present invention, not only a single frequency shift amount but also within a predetermined range (± ΔDv) on both sides of the basic frequency shift amount. Another frequency shift amount is set.

(Fb2-fb1) = (4 * (VB ± Dv) * f0) / C (E) Therefore, it can be considered that the true frequency shift amount exists in these frequency shift amounts. The true frequency shift amount can be determined by the evaluation described above.

(6) The FMCW radar according to any one of claims 1 to 5, wherein the frequency shift amount is set in consideration of the beam direction of the FMCW radar device. The device is the gist.

When the radar beam is directed in a direction other than the traveling direction of the vehicle, a deviation occurs between the moving direction of the stationary object and the beam direction in which the relative speed can be detected due to the Doppler effect, and there is an error in the estimation of the actual vehicle speed. Occurs. Therefore, in the present invention, since the frequency shift amount is set in consideration of the beam direction, the actual vehicle speed can be estimated more accurately.

For example, when the direction of the beam is laterally deviated by θ from the traveling direction of the vehicle, the value obtained by multiplying the relative velocity of the target measured by this beam by cos θ is the relative velocity of the true target. I can see it. (7) In the invention of claim 7, when using the upstream spectrum and the downstream spectrum corresponding to the determined frequency shift amount, the spectrum matching degree is evaluated for each pair of spectral peaks corresponding to both spectra, The FMCW radar device according to any one of claims 1 to 6, characterized in that a stationary object determination of each target is performed based on the evaluation result.

As the evaluation here, the same evaluation as that used for the true frequency shift amount in the above-mentioned claim 1 (for example, the sum of neighborhoods Sum2 of each spectrum peak) can be adopted. In the invention of claim 1, when a true frequency shift amount is determined from a plurality of frequency shift amounts, one set of upstream spectrum and downlink spectrum is selected corresponding to the true frequency shift amount. To be done. Therefore, in the present invention, for this set of upstream spectrum and downstream spectrum,
Similar evaluation of the degree of spectral matching is performed for each pair of corresponding spectral peaks.

Here, for example, as shown by Sn in FIG.
When the evaluation value of the target corresponding to each spectrum peak exceeds a predetermined threshold value, it can be determined that the target is a moving object (or other than a stationary object). Therefore, for example, it is possible to determine a stationary object from targets whose evaluation value is equal to or less than the threshold value.
If the possibility of a combined peak is low, it is possible to determine a target whose evaluation value is less than or equal to a threshold value as a stationary object.

(8) According to the invention of claim 8, a moving object prediction flag is set by predicting a moving position of this time with respect to a target already recognized as a moving object, and a spectrum peak to be judged this time is set. On the other hand, when the moving object prediction flag is set, the FMCW radar device according to claim 7 is characterized in that the target is not determined as a stationary object.

When the FMCW radar is used as an on-vehicle radar, a stationary object such as a roadside object such as a guardrail or a moving object that is a vehicle traveling in front appears in a spectrum, and it may appear depending on the relationship between the own vehicle and the stationary object. In some cases, the spectrum peaks of a stationary object and a moving object are combined, so that the stationary object cannot be accurately determined.

Therefore, in the present invention, a moving object prediction flag is set by predicting the moving position of this time for a target that has already been recognized as a moving object, and the moving object is detected for the peak to be judged this time. When the prediction flag is set, it is regarded as a synthesized peak, and the peak is not judged as a stationary object.

As a result, even when driving in an urban area where there are many roadside objects, a composite peak obtained by combining a moving object peak and a stationary object peak is not recognized as a stationary object, and a more accurate stationary object determination can be made. It will be possible. (9) The invention of claim 9 compares the actual vehicle speed obtained by the estimation with the relative speed of the target recognized by the FMCW radar device, and reflects the comparison result in the stationary object determination. The FMCW radar device according to claim 7 or 8 is characterized in that.

The actual vehicle speed obtained by the estimation and the FMC
When comparing the relative speed with the target recognized by the W radar device, if the absolute values match, it can be determined that the target is a stationary object and the other objects are moving objects. Therefore, by using the result of this determination, it is possible to perform a more accurate stationary object determination if the target object determined to be a moving object is not evaluated for stationary object determination.

(10) The invention of claim 10 is characterized in that the evaluation of the spectrum matching degree is performed based on at least the amplitude of the spectrum peak among the amplitude of the spectrum peak and the azimuth information of the target. The gist is the FMCW radar device according to any one of Items 1 to 9.

Here, the evaluation of the degree of spectral matching means
The evaluation used when determining the true frequency shift amount and the evaluation used when determining the stationary object of the target are shown. In the present invention, the spectrum match can be evaluated simply based on the amplitude (peak level) of the spectrum peak. In particular, more accurate evaluation can be performed when the spectrum match is evaluated based on not only the amplitude of the spectrum peak but also the azimuth information of the target.

For example, in both the up and down spectrum,
Accidental stationary object determination may not be possible if peaks of the same level happen to exist. However, for example, when a phase difference monopulse radar having two receiving antennas is used, the azimuth information from the target is indicated by the phase difference between the two systems, and the phase difference is coded between the up part and the down part. Have the same value, but this can be used to evaluate the degree of coincidence of spectral peaks.

(11) According to the invention of claim 11, the amplitude evaluation value based on the amplitude of the spectrum peak is Y, and the phase evaluation value based on the azimuth information of the target is X, based on the absolute value of the evaluation vector. The FMCW radar device according to claim 10, characterized in that the spectrum is evaluated.

The present invention exemplifies a spectrum evaluation method. For example, as shown in FIG. 7, the amplitude evaluation value based on the amplitude of the spectrum peak is set to Y, and the phase evaluation value based on the direction information of the target is calculated. Since the absolute value of the evaluation vector set as X is calculated and the spectrum is evaluated using this absolute value, it is possible to more accurately determine the frequency shift amount and determine the stationary object.

Here, when the phase evaluation value X is not used, X = 0 can be set and the evaluation can be performed using only the amplitude evaluation value Y. The amplitude evaluation value can be calculated from the following formula (F), and the phase difference evaluation value can be calculated from the following formula (G).

Amplitude evaluation value = | (rising part peak level−falling part peak level) / rising part peak level | ... (F) Phase difference evaluation value = | rising part phase difference−falling part phase difference | ... (G) When the peak is expressed by a complex vector, the amplitude of the spectrum peak is shown by the absolute value (length) of the complex vector, and the phase is shown by the rotation angle. The value may be calculated.

(12) According to a twelfth aspect of the present invention, the sum of neighborhoods of the absolute values of the evaluation vectors for the spectrum peaks to be evaluated (the sum of the range of frequencies in the vicinity including the center frequency of the peaks) is assigned to each spectrum peak. The sum of the neighborhood sums of the spectrum total sum is the smallest,
The gist of the FMCW radar device according to claim 11 is that the amount of shift is a true frequency shift amount.

The present invention exemplifies a method for determining the true frequency shift amount. First, for example, the following formula (H)
Is used to find the neighborhood sum of the absolute values of the evaluation vector for each peak. Here, the reason why the sum of neighborhoods is used is that the evaluation accuracy is higher when the evaluation is performed by including the frequencies in the vicinity of each peak than when the evaluation is performed only at one point.

Neighborhood sum Sum2 = | Vp (pn) | + | Vp (p-n + 1) | + ... + | Vp (p) | + ... + | Vp (p + n) | ... (H) (p ); Peak frequency number (indicating which peak)
(n); Width of neighborhood Next, after obtaining the neighborhood sum Sum2 for each of the peak neighborhoods, each neighborhood sum Sum2 is expressed by the following formula (I), for example.
And sum the total spectrum S for each frequency shift amount.
Find um1.

Spectral total sum Sum1 = Σ neighborhood sum Sum2 (I) Then, for example, each spectral total sum Su of this evaluation value
Of m1 (| Vp |), the frequency shift amount having the smallest value can be determined as the true frequency shift amount TSn. That is, the smaller the total sum Sum1 of spectra is, the smaller the absolute value of each evaluation vector is as a whole,
Therefore, it is considered that the degree of coincidence of each spectrum peak is large.

(13) According to a thirteenth aspect of the present invention, the neighborhood sum of absolute values of the evaluation vector relating to the spectral peak to be evaluated is compared with a predetermined threshold value. The FMCW radar device according to claim 11, wherein the target is determined as a stationary object.

According to the present invention, after the true frequency shift amount is determined, in the upstream spectrum and the downlink spectrum corresponding to the true frequency shift amount, how to determine the stationary object of the target object for each pair of spectrum peaks. It is an example of how to perform. Here, the neighborhood sum of the absolute value of a certain evaluation vector is compared with a predetermined threshold, and if the neighborhood sum is less than or equal to the threshold, the amplitude evaluation value is small and the phase difference evaluation value is small, and the spectrum peak due to the same stationary object is small. Therefore, the target is judged to be a stationary object.

(14) The invention of claim 14 is summarized as a recording medium characterized by storing means for executing processing by the FMCW radar device according to any one of claims 1 to 13. . In other words, there is no particular limitation as long as it stores a means such as a program that can execute the processing of the FMCW radar device described above.
Specifically, there is an electronic control device that does not include a transmission unit and a reception unit in the FMCW radar device and can estimate the actual vehicle speed and determine a stationary object by processing a signal.

Examples of the recording medium include various types of recording media such as an electronic control unit configured as a microcomputer, a microchip, a floppy disk, a hard disk, and an optical disk. (15) The invention of claim 15 is a vehicle control device characterized by using the actual vehicle speed estimated by the invention of any one of claims 1 to 14 for various vehicle controls. .

In the case of vehicle control using the own vehicle speed which is the ground speed, it is important to use an accurate own vehicle speed. Therefore, by using the accurate actual vehicle speed estimated by the inventions of the above-mentioned claims for various vehicle controls (for example, cruise control, anti-skid control), more precise and suitable control can be performed. it can.

In particular, during vehicle control, there are problems in error, smoothing, and responsiveness, such as a vehicle speed sensor that detects the vehicle speed from the rotational speed of the differential gear and the rotational speed of the crank, a wheel speed sensor that detects the vehicle speed from the wheel speed, etc. If a sensor that seems to be present is used, or if the detection of the vehicle body speed is delayed due to communication, etc., it is possible to perform more precise control by using the estimated actual vehicle speed. .

[0059]

BEST MODE FOR CARRYING OUT THE INVENTION Embodiments (embodiments) of an FMCW radar device, a recording medium and a vehicle control device of the present invention will be described below with reference to the drawings. (Embodiment) a) First, FIG. 1 shows a basic configuration of an FMCW radar device and a vehicle control device of the present embodiment.

As shown in FIG. 1, the vehicle control device 1 is an on-vehicle computer that performs various vehicle controls such as anti-skid control and constant-speed cruise control. FMCW radar device (hereinafter simply referred to as a radar device) 2 is connected, and a wheel speed sensor 3a and (detects the rotation speed of the crank angle).
A sensor 3 such as a vehicle speed sensor 3b is connected. Also,
For example, various actuators 5 such as a throttle valve 5a, an injector 5b, and a solenoid valve 5c for hydraulic control are connected.

Then, as will be described later in detail, the vehicle control device 1 outputs the vehicle speed VB obtained from the signal from the vehicle speed sensor 3b etc. to the radar device 2 and the radar device 2
From this, the actual vehicle speed TVB obtained by correcting the vehicle speed VB is output to the vehicle control device 1. Therefore, the vehicle control device 1 can perform various vehicle controls using the actual vehicle speed TVB after the actual vehicle speed TVB is obtained.

Next, FIG. 2 shows the overall configuration of the radar device 2, which is a so-called phase difference monopulse radar device. As shown in FIG. 2, the radar device 2
Is transmitted from a transmitter 12 that transmits a radar wave modulated to a predetermined frequency according to the modulation signal Sm, and a pair of receivers 14 and 16 that receives the radar wave emitted from the transmitter 12 and reflected by an obstacle. In order to detect the obstacle and determine the stationary object based on the intermediate frequency beat signals B1 and B2 output from the receivers 14 and 16 while supplying the modulated signal Sm to the transmitter / receiver 10 and the transmitter 12 And a signal processing unit 20 that executes the processing of.

Here, the transmitter 12 is the transmitting means of the present invention,
The receivers 14 and 16 correspond to receiving means, and the signal processing unit 20 corresponds to other means (for example, spectrum creating means, frequency shift amount setting means, detecting means, etc.). And in this embodiment,
In order to detect obstacles in front of the vehicle by the radar device 2, the transmitting / receiving unit 10 is attached to the front of the vehicle,
The signal processing unit 20 is attached at a predetermined position in or near the vehicle compartment.

Here, first, the transmitter 12 outputs a voltage controlled oscillator (VCO) 12b for generating a high frequency signal in the millimeter wave band as a transmission signal and a modulation signal Sm to the voltage controlled oscillator 1.
A modulator (MOD) 12a that converts the adjusted level to 2b and supplies the voltage to the voltage controlled oscillator 12b;
power distributor (COU) that distributes the power of the transmission signal from b to generate a local signal to be supplied to each of the receivers 14 and 16.
P) 12c and 12d, and a transmission antenna 12e that radiates a radar wave according to a transmission signal.

Further, the receiver 14 amplifies the outputs of the receiving antenna 14a for receiving the radar wave, the mixer 14b for mixing the received signal from the receiving antenna 14a with the local signal from the power distributor 12d, and the mixer 14b. A pre-amplifier 14c and a low-pass filter 1 for removing an unnecessary high-frequency component from the output of the pre-amplifier 14c and extracting a beat signal B1 which is a difference component between frequencies of a transmission signal and a reception signal.
4d and a post-amplifier 14e for amplifying the beat signal B1 to a required signal level. The receiver 16 has the same configuration as the receiver 14 (14a to 1).
4e corresponds to 16a to 16e), receives a local signal from the power distributor 12c, and receives a beat signal B2.
Is output. Then, the receiver 14 receives the reception channel CH
1. The receiver 16 is called a reception channel CH2.

On the other hand, the signal processing unit 20 is activated by the activation signal C1 and generates the triangular wave modulation signal Sm, and the activation signal C2 activates the beat signal B1 from the receivers 14 and 16. , B2 into digital data D1, D2 A / D converters 24a, 24b
The CPU 26a, the ROM 26b, and the RAM 26c are mainly configured to send start signals C1 and C2 to operate the triangular wave generator 22 and the A / D converters 24a and 24b.
At the same time, an obstacle that detects the distance to the obstacle, the relative speed, and the azimuth of the obstacle based on the digital data D1 and D2 obtained through the A / D converters 24a and 24b and determines the stationary object. The microcomputer 26 includes a well-known microcomputer 26 that executes an object detection process (which will be described later) and an arithmetic processing unit 28 that executes an operation of a fast Fourier transform (FFT) based on a command from the microcomputer 26.

When the A / D converters 24a and 24b start operating by the activation signal C2, the beat signals B1 and B2 are A / D converted at predetermined time intervals, and the RAM 26c.
When the A / D conversion has been performed a predetermined number of times while writing to a predetermined area of No. 3, an end flag (not shown) set on the RAM 26c is set and the operation is stopped.

When the triangular wave generator 22 is activated by the activation signal C1 and the modulation signal Sm is input to the voltage controlled oscillator 12b via the modulator 12a, the voltage controlled oscillator 12b generates a triangular waveform of the modulation signal Sm. The frequency increases at a predetermined rate according to the rising slope of the waveform (hereinafter, this section is referred to as the rising section), and the frequency decreases according to the subsequent downward slope (hereinafter, this section is referred to as the falling section). The transmission signal thus modulated is output.

FIG. 3 is an explanatory diagram showing the modulation state of the transmission signal. As shown in FIG. 3, the frequency of the transmission signal is modulated by the modulation signal Sm so as to increase or decrease by ΔF in the period of 1 / fm, and the center frequency of the change is f0. The frequency is modulated at 100 ms intervals because the obstacle detection process described later is executed at 100 ms cycles, and the activation signal C1 is generated in the process.

A radar wave corresponding to this transmission signal is transmitted by the transmitter 1.
The radar waves transmitted from the antenna 2 and reflected by the obstacles are received by the receivers 14 and 16. Then, the receivers 14 and 16
Then, the beat signals B1 and B2 are generated by mixing the reception signals output from the reception antennas 14a and 16a with the transmission signals from the transmitter 12. It should be noted that the reception signal is delayed with respect to the transmission signal by the time for the radar wave to make a round trip to the obstacle, and if there is a relative velocity with the obstacle, the reception signal undergoes Doppler shift accordingly. . Therefore, the beat signals B1 and B2 include the delay component fr and the Doppler component fd (see FIG. 14).
Becomes

As shown in FIG. 4, the digital data D1 obtained by A / D converting the beat signal B1 by the A / D converter 24a is the data block D on the RAM 26c.
Sequentially stored in B1 and DB2, while the A / D converter 24
Similarly, the digital data D2 obtained by A / D converting the beat signal B2 by b is similarly used in the data blocks DB3 and DB4.
Stored in. By the way, the A / D converters 24a and 24b
Is started when the triangular wave generator 22 is started, and A / D conversion is performed a predetermined number of times while the modulation signal Sm is being output. Therefore, the data block DB1 in which the first half of the data is stored , DB3 stores the rising portion data corresponding to the rising portion of the transmission signal, and the data blocks DB2 and DB4 storing the latter half of the data store the falling portion data corresponding to the falling portion of the transmission signal. Will be. In this way, each data block DB1 to DB4
The data stored in is processed by the microcomputer 26 and the arithmetic processing unit 28, and is used for detecting obstacles and stationary objects.

B) Next, the obstacle detection process executed by the microcomputer 26 will be described with reference to the flowchart of FIG. It should be noted that this obstacle detection processing is 10
It is activated in a 0 ms cycle. As shown in FIG. 5, when this process is started, first, in step 110, the start signal C
1 is output to activate the triangular wave generator 22, and in step 120, the end flag on the RAM 26c is cleared and the activation signal C2 is output to output the A / D converter 24a,
24b is activated.

As a result, the transmitter 12, which receives the modulation signal Sm from the triangular wave generator 22, transmits the frequency-modulated radar wave and receives the radar wave reflected by the obstacle, thereby receiving the receiver 14. , 16 output beat signals B1, B2 from the A / D converter 24a,
R is converted to digital data D1 and D2 via 24b.
Written to AM 26c.

At the following step 130, it is determined whether or not the A / D conversion is completed by checking the end flag on the RAM 26c. If the end flag is not set, it is determined that the A / D conversion is not completed, and the process waits by repeatedly executing the same step 130. On the other hand, if the end flag is set, the A / D conversion is completed. Is completed, and the process proceeds to step 140.

At step 140, one of the data blocks DB1 to DB4 on the RAM 26c is sequentially selected, and the data of the data block DBi (i = 1 to 4) is input to the arithmetic processing unit 28 to perform the FFT operation. To execute. The data input to the arithmetic processing unit 28 is F
Well-known window processing using a Hanning window, a triangular window, or the like is performed in order to suppress side lobes that appear due to the calculation of FT. Then, as a result of this calculation, a complex vector for each frequency is obtained.

In step 150, based on the absolute value of the complex vector, that is, the amplitude of the frequency component indicated by the complex vector, all frequency components having a peak on the frequency spectrum (hereinafter referred to as peak frequency components) are detected, The frequency is specified as the peak frequency, and the process proceeds to step 160. As a method of detecting the peak, for example, the amount of change in the amplitude with respect to the frequency is sequentially obtained, and it is assumed that there is a peak in the frequency at which the sign of the amount of change is inverted before and after that.
The frequency should be specified.

In step 160, the phase of the peak frequency component specified in step 150 is calculated. This phase is equal to the angle formed by the complex vector and the real axis, and can be easily obtained from the complex vector. In the following step 165, it is determined whether or not there is an unprocessed data block DBi. If there is an unprocessed data block DBi, the process returns to step 140,
The processes of steps 140 to 160 are executed for the unprocessed data block DBi, while if there is no unprocessed data block DBi, the process proceeds to step 170.

At step 170, as will be described in detail later,
An actual vehicle speed estimation process for estimating the actual vehicle speed TVB is performed.
In the following step 175, as will be described in detail later, a stationary object determination process for determining whether the detected obstacle is a stationary object is performed.

In the following step 180, by comparing the amplitudes of the peak frequency components, that is, the powers respectively, those having the same power in the ascending portion and the descending portion are identified as the peak frequency components based on the reflected wave from the same obstacle. A pairing process that identifies a pair is executed.

However, in this case, the pairing processing is not performed for the peaks which are clearly determined not to be a pair in the stationary object determination processing in step 175. The pairing process is the same as the process shown in FIG. 7 of Japanese Patent Application No. 8-179227 and the description thereof, and the detailed description thereof will be omitted.

In the following step 190, step 180
Using the peak frequency component paired in, the distance / velocity azimuth calculation processing for calculating the distance to the obstacle, the relative speed, and the azimuth of the obstacle is executed, and this processing ends. For example, each receiving channel CH1,
The phase difference between CH2 is calculated, and when the signs of the phase difference are not equal, the distance D to the obstacle and the relative speed V are calculated using the following equations (1) and (2).

V = (C / (4 * f0)) * (fb2-fb1) (1) D = (C / (8 * ΔF * fm)) * (fb1 + fb2) (2) However, ΔF Is the frequency displacement width of the transmission signal (frequency displacement width), f0 is the center frequency of the transmission signal, 1 / fm is the time required for one cycle of modulation (that is, fm is the triangular wave repetition frequency), C is the speed of light, and fb1 is the rising portion. , Fb2 represents the beat frequency of the descending part (downbeat frequency), and C represents the speed of light.

The distance / speed / azimuth calculation processing is the same as the processing shown in, for example, FIG. 5 of Japanese Patent Application No. 8-179227 and its description, so a detailed description thereof will be omitted. c) Next, the basic principle of the actual vehicle speed estimation processing performed in step 170 will be described.

Here, to estimate the actual vehicle speed TVB,
First, a plurality of frequency shift amounts (Sn-1, Sn, Sn + 1) are set in consideration of the error of the vehicle speed sensor 3b and the like, and the true frequency shift amount TSn is obtained from them by using the evaluation function. Then, the actual vehicle speed is calculated using the true frequency shift amount TSn. The details will be described below.

(I) First, the procedure for calculating the frequency shift amount (hereinafter also simply referred to as the shift amount) will be described. First, a basic frequency shift amount calculation formula for calculating a basic frequency shift amount (basic shift amount) is set by using the speed of the own vehicle (own vehicle speed) VB obtained by the vehicle speed sensor 3b and the like.

That is, the formula (1) is modified to set the formula (3) for calculating the basic amount (basic shift amount = (fb2-fb1)) for determining how much the spectrum is shifted. Basic shift amount = (fb2-fb1) = (4 * VB * f
0) / C (3) At this time, if the error of the vehicle speed sensor 3b or the like is known or already learned, the own vehicle speed VB corrected by the correction coefficient, map calculation or the like is used.

Next, correction is performed according to the beam angle of the laser device 12. As shown in FIG. 6, when the beam angle from the sensor front direction or the traveling direction in the case of a beam steer or a scan sensor is θ, in the case of a radar that detects the relative velocity using the Doppler effect, the detectable velocity The component is a velocity component equal to the beam direction (-VB * C
Since it is OS θ), it is necessary to perform correction. This component becomes smaller as the angle at which the beam is steered and scanned becomes larger, and the deviation from the true moving speed (approaching speed; -VB) becomes larger, and accurate frequency shift cannot be performed.

Therefore, here, the following equation (4) for the first correction in which the angle correction coefficient (COSθ) is added to the basic shift amount is set. As a result, the larger θ becomes,
The basic shift amount is corrected to be small. First post-correction shift amount = (4 * COS (θ) * VB * f0) / C (4) Next, correction is performed in consideration of the response delay of the vehicle speed sensor 3b.

The wheel speed sensor 3a and the vehicle speed sensor 3b (hereinafter, mainly the vehicle speed sensor 3b is taken as an example) generally measure the time intervals of pulse signals from the drive system and the wheel system, and use the values to measure the actual values. Detect the vehicle speed. However, in reality, there is a response delay with respect to the actual vehicle speed because the time is filtered due to stability, noise, and the like. For example, in a situation where the vehicle travels at a speed of 100 km / h, the delay is not a concern, but a time delay of several kilometers per hour occurs during acceleration and deceleration. Therefore, in this embodiment,
In consideration of the delay, processing with an allowable value is performed.

Specifically, since the time delay width differs depending on the type of actual vehicle, it is possible to give each vehicle a basic delay width in advance. Although this varies depending on the vehicle speed VB, the time constant of the filter of the vehicle speed sensor 3b, etc., it is set as the speed delay value Dv here. Therefore, using this value, the following equation (5) for the second correction is set.

When setting the speed delay value Dv,
It is desirable to take into consideration the error due to other factors (error of the spoiler or the sensor itself). Second corrected shift amount = (4 * COS (θ) * (VB ± Dv) * f0) / C (5) Here, the value of the speed delay value Dv is the vehicle speed sensor 3b.
It should be a value considering the resolution of. For example, when the vehicle speed sensor 3b has an error of ± 5 km / h, Dv = (− 5,
It is possible to set it in three ways, such as 0, +5).

That is, in reality, there is a possibility that the vehicle speed sensor 3b will be delayed by the width of the speed delay value Dv.
According to the equation (5), the basic shift amount can have a certain width. That is, as will be described in detail later, the frequency shift amount is set to have a certain range, a plurality of frequency shift amounts are set, and the most matched frequency shift amount is selected from among the frequency shift amounts. The frequency shift amount can be set.

In order to simplify the processing, it is possible to set Dv to a value that absorbs the influence of the angle component. (ii) Next, the evaluation method using the evaluation function will be described.
Here, a plurality of frequency shift amounts obtained by the above equation (5) are used to shift the spectra of the descending portion and the degree of coincidence with the spectra of the ascending portion is compared.

Conventionally, only the corresponding spectral peaks (peaks of peak frequency components) of the upstream and downstream spectra are subtracted using the uniquely determined frequency shift amount, but in this embodiment, a plurality of them are subtracted. The following evaluation function is used in order to obtain the optimum frequency shift amount from the frequency shift amounts (thus, the shift width when shifting).

This evaluation function is expressed by the following equations (6) and (7).
As shown in, not only the amplitude of the spectrum peak but also the phase difference indicating the azimuth information obtained by the phase difference monopulse radar is used. The phase difference is a value obtained by subtracting the respective phase information received by the monopulse radar having two receiving systems, and the method of finding the azimuth of the object using this phase difference is the phase difference monopulse radar. .

Amplitude evaluation value = | (Rising peak level−Descent peak level) / Rising peak level | ... (6) Phase difference evaluation value = | Rising phase difference−Descent phase difference | ... (7) In the case of the phase difference monopulse radar, the sign is reversed in the ascending part and the descending part due to its configuration, so that if the sum is 0, they match.

Then, as shown in FIG. 7, the length of the evaluation vector Vp having the amplitude evaluation value Y and the phase difference evaluation value X is |
Let Vp | be the evaluation value. If the phase difference evaluation value is not used, the accuracy is somewhat lowered, but there is an advantage that the calculation is reduced.

Next, the evaluation value | Vp | is calculated for each peak frequency component and the sum is calculated. In that case, as shown in the following equation (8), the target peak frequency component is obtained. Not only for the frequencies around it,
Similarly, the evaluation value | Vp | is obtained, and the sum (neighborhood sum S
um2). In addition, the range of the neighborhood sum Sum2 is
As shown in FIG. 10, it is a band-shaped range sandwiched by right and left broken lines with the alternate long and short dash line as the center, and varies depending on the resolution of the FFT.

Here, the reason for obtaining the neighborhood sum Sum2 is that the accuracy is higher than that in the case where a single peak frequency component is used. Neighborhood sum Sum2 = | Vp (pn) | + | Vp (p-n + 1) | + ... + | Vp (p) | + ... + | Vp (p + n) | ... (8) where P; ( Peak frequency number, which indicates the order of peaks to be evaluated, n; Neighborhood width (when neighborhood is divided into n) In addition, in this case, the neighborhood sum is not calculated for all the peak frequency components, but a stationary state described later. The neighborhood sum Sum2 is calculated only for the peak frequency component for determining the object and the moving object. This is because if the processing is performed on all of them, not only a correct result may not be detected due to peaks such as noise and clutter, but also a large amount of calculation time is required.

Then, as described above, for the up and down spectrums corresponding to the respective frequency shift amounts,
That is, the evaluation value Sum2 is calculated in the vicinity of each peak frequency component for a pair of spectral peaks corresponding to both the upstream and downstream spectra.

Here, as shown in FIG. 8, for example, the peaks existing in the rising portion (upstream spectrum) are identified as Pu1 and Pu.
2, Pu3, Pu4, and the peaks existing in the descending part (downstream spectrum) are represented by Pd1, Pd2, Pd3, Pd4 (where 1,
2 and 3 are stationary objects and 4 is a peak of a moving object).

At this time, considering the case where the downlink spectrum is shifted by a predetermined frequency shift amount as shown in FIG. 9, when the vehicle speed VB including many errors and the like is shifted, FIG. ), The peak of the upstream spectrum (shown by the solid line) and the peak of the downstream spectrum (shown by the broken line) are largely deviated from each other and do not match. On the other hand, when the shift is performed at the actual vehicle speed TBV, the error is small, and as a result, as shown in FIG. 9B, naturally, the degree of coincidence between the peak of the up spectrum and the peak of the down spectrum is high.

Therefore, it can be considered that the peak of both spectra having a high degree of coincidence is shifted at the actual vehicle speed TBV, and conversely, the frequency shift amount having a high degree of coincidence (hence the true frequency). The actual vehicle speed TVB can be estimated from the shift amount TSn).

FIG. 10 shows a method for selecting the true frequency shift amount from the plurality of frequency shift amounts described above. In FIG. 10, for the basic shift amount Sn, S
There were three types of frequency shift amounts of n ± 1. Next, the neighborhood sum Sum for each of the spectral peaks
After 2 is calculated, each neighborhood sum Sum2 is calculated by the following equation (9).
And sum the total spectrum S of a certain frequency shift amount.
Find um1. This is obtained for each of a plurality of frequency shift amounts.

Total spectrum sum Sum1 = Σ neighborhood sum Sum2 (9) Then, the frequency shift amount having the smallest total spectrum sum Sum1 (| Vp |) corresponding to each frequency shift amount is determined as the true frequency shift amount TSn. And Therefore, for example, in FIG. 10, the frequency shift amount Sn + 1 corresponding to the right spectrum is selected as the true frequency shift amount TSn.

Thus, the total sum of spectra Sum1 (|
The reason why the smallest Vp |) is taken as the true frequency shift amount TSn is that the smaller the overall spectrum sum Sum1 is, the smaller the absolute value of each evaluation vector is as a whole, and therefore the greater the degree of coincidence of each spectrum peak is. Because it will be done.

Although three types of shift widths are given as an example here, the true frequency shift amount is actually calculated based on the width obtained by the equation (3) indicating the second corrected shift amount. TSn is determined. Also, once the true frequency shift amount T
After Sn is obtained, the frequency shift amount may be calculated in consideration of the true frequency shift amount TSn in the next calculation.

(Iii) Next, a method of calculating the actual vehicle speed TVB using the true frequency shift amount TSn will be described.
In the above calculations (i) and (ii), the true frequency shift amount TSn was obtained by excluding the influence of the vehicle speed sensor 3b such as delay, error, and beam direction. Here, the true frequency shift TVB is calculated by applying the true frequency shift amount TSn to the equation (11) obtained by modifying the following equation (10) obtained from the equation (3).

TSn = (4 * TVB * f0) / C (10) TVB = TSn * C / (4 * f0) (11) Incidentally, the previous value TVBn- of the actual own vehicle speed TBV one cycle before.
The actual relative acceleration dTVB of the vehicle is obtained by dividing the difference between 1 and the current value TVBn by the period ΔT, but the actual relative acceleration dTVB may be used to correct the frequency shift amount.

For example, when setting a plurality of frequency shift amounts using the above equation (5), the vehicle speed VB obtained from the vehicle speed sensor 3b is not used as it is, but the vehicle speed predicted by using the actual relative acceleration dTVB is also used. YVB may be taken into consideration and, for example, the average value of VB and YVB may be used as the vehicle speed in Expression (5).

D) Next, referring to FIG. 11, the actual vehicle speed estimation processing of step 170 performed based on the above principle will be described.
A description will be given based on the flowchart. In this processing, the true frequency shift amount TSn is obtained based on the above-described principle, and the actual vehicle speed TV is calculated from the true frequency shift amount TSn.
This is a process for obtaining B.

First, in step 200 of FIG. 11, the basic shift amount is calculated using the own vehicle speed VB including the error obtained by the vehicle speed sensor 3b on the basis of the equation (5) set in the procedure from to. Then, the width of the second post-correction shift amount, that is, the frequency shift amount is corrected.

In the following step 210, the frequency shift is performed from the shift width Sn-1 (which has not been frequency-shifted yet), for example. For example, the entire downlink spectrum is shifted. Therefore, when this processing is passed the second time, the frequency shift of the next shift width is performed.

In the following step 220, the step 2
In the spectrum frequency-shifted by 10, for the predetermined peak to be evaluated, the equation (6),
Based on (7), the evaluation value | Vp | near the peak (hence the peak frequency component) is sequentially calculated.

In the following step 230, the evaluation value | Vp | in the vicinity of the predetermined peak calculated in step 220 is summed based on the above equation (8) to calculate the sum Sum2 of the peaks. In the following step 240, it is determined whether or not the calculation of the neighborhood sum Sum2 has been completed by the number of peaks for which determination is desired. For example, as shown in FIG. 9, for example, in the shift width Sn-1, it is determined whether or not all of the neighborhood sums Sum2 have been calculated for four peaks. If an affirmative determination is made here, the process proceeds to step 250, and if a negative determination is made, the process returns to step 220 and subsequent steps, and the evaluation value | Vp | of the other peaks and the neighborhood sum Su.
Calculate m2.

At step 250, the step 220 is executed.
At ~ 240, the calculation of the sum of neighborhoods Sum2 of all peaks is completed, so that they are summed based on the above equation (9) to calculate the sum of spectra Sum1. In subsequent step 260, it is determined whether or not the shift width has been shifted. For example, if the frequency shift amount is Sn-1, Sn, Sn
If there are three types of +1, it is determined whether or not the above-described total spectrum sum Sum1 or the like has been performed for each frequency shift amount. If an affirmative judgment is made here, the routine proceeds to step 270, while if a negative judgment is made, step 210 is carried out.
The process returns to the subsequent steps, and the process of calculating another total spectrum sum Sum1 is performed in the same manner as described above.

At step 270, the values of all the spectrum sums Sum1 (for example, three ways in FIG. 9) are compared,
The frequency shift amount corresponding to the smallest value is set as the true frequency shift amount TSn. In the following step 280,
The true frequency shift amount TSn is applied to the above equation (11) to calculate the actual vehicle speed TVB, and the present process is terminated.

By this actual vehicle speed estimation processing, it is possible to obtain a more accurate actual vehicle speed TVB than the own vehicle speed VB obtained by the vehicle speed sensor 3b. e) Next, the basic principle of the stationary object determination process performed in step 175 will be briefly described.

Here, the up and down spectra corresponding to the true frequency shift amount TSn are used to determine the stationary object, and it is determined whether the target corresponding to the peak frequency component is a moving object or a stationary object. The details will be described below. First stationary object determination Conventionally, only spectrum subtraction was performed. However, in the present embodiment, there is a possibility that spectral peaks having the same peak level happen to exist, that is, a moving object and a stationary object have peaks. The processing is performed in consideration of the case of composition.

As shown in FIG. 9, when the neighborhood sum Sum2 of a certain spectrum peak is less than or equal to the threshold Thp, the peaks of the rising portion and the falling portion match with respect to the peak level and the azimuth (phase difference) of the beam. Therefore, the stationary object is determined to be that the matched peak is a stationary object. On the other hand, when the neighborhood sum Sum2 of a certain spectrum peak exceeds the threshold Thp, it is determined that the combination of the peaks of the moving object and the stationary object, or the combination of the peaks due to noise or the like, and the stationary object determination is not performed.

Second stationary object determination Here, the actual vehicle speed TVB and the relative speed RVB of the target recognized by the radar device 2 are compared, and when the actual vehicle speed TVB and the relative speed RVB match. In addition, the judgment that it is likely to be a stationary object is added.

That is, when the target is a stationary object, it should approach at the same speed as the own vehicle speed VB, but the own vehicle speed VB and the relative speed R obtained simply by the vehicle speed sensor 3b.
There is a possibility that the determination of the stationary object is inaccurate just by comparing with VB, but by comparing the accurate actual vehicle speed TVB and the relative speed RVB obtained by this embodiment,
It is possible to determine still objects with higher accuracy.

Judgment Using Moving Object Prediction Flag Whether a peak to be judged is a combination of moving object and stationary object peaks is judged by using the moving object prediction flag. That is, as shown in FIG. 12, a peak that was not previously determined to be a stationary object and was determined to be a moving object by another means will appear after Δt depending on the motion state of the moving object. The peak position is predicted. The flag indicating this predicted position is called a moving object prediction flag. Therefore, even if a stationary object is determined, if the moving object prediction flag is set at the position of the peak, it is determined to be a combined peak of the moving object and the stationary object, and the stationary object determination is not performed. Incidentally, similarly, the moving object prediction flag is sequentially set.

F) Next, the stationary object determination processing of step 175, which is performed based on the principle of stationary object determination, will be described with reference to the flowchart of FIG. This process is a process for determining whether or not the spectral peak of the obstacle (target) recognized by the radar device 2 corresponds to a stationary object based on the principle described above.

First, in step 300 of FIG. 13, with respect to the spectrum frequency-shifted by the true frequency shift amount TSn (the spectrum of the shift amount Sn + 1 on the right side in FIG. 10), a predetermined peak (specifically, upstream and downstream). Sum 2 of the pair of peaks corresponding to each other in the spectrum of
Whether it is less than or equal to hp is determined. If an affirmative decision is made here, the operation proceeds to step 310, whereas if a negative decision is made, the step 3
Go to 20.

In step 350, the peak neighborhood sum Su
Since m2 is larger than the threshold value Thp, that is, the degree of coincidence of evaluation is low to determine a stationary object, it is determined that the object marker corresponding to the peak is a moving object and a moving object flag indicating a moving object is set. Set and proceed to step 360.

On the other hand, in step 3100, the relative speed RVB of the target detected by the radar device 2 and the actual vehicle speed TVB are compared to determine whether they match (within a certain allowable range). . Here, if an affirmative judgment is made, the routine proceeds to step 320, while if a negative judgment is made, the routine proceeds to step 350.

In other words, this process is the same as step 300 above.
For targets that are judged to be highly likely to be stationary,
This is a process for more accurately determining a stationary object.
In step 320, it is determined whether or not the moving object prediction flag is set at the peak position of the target to be determined. If an affirmative judgment is made here, the routine proceeds to step 340, while if a negative judgment is made, the routine proceeds to step 330.

At step 340, since the moving object prediction flag is set, it is determined that the peak is a combined peak of a stationary object and a moving object, and a combined peak flag indicating a combined peak is set, and step 360 is performed. Proceed to. On the other hand, in step 330, since the moving object prediction flag is not set, that is, the neighborhood sum Sum2 is less than or equal to the threshold Thp, the target relative speed RVB and the actual vehicle speed TVB match, and the moving object prediction flag is set. Is not set, it is judged that the peak is the peak of a stationary object,
Set a stationary object flag indicating stationary object, step 360
Proceed to.

In step 360, it is determined whether or not the processing in steps 300 to 350, that is, the processing for determining what the peak means has been completed by the number of peaks to be determined. If a negative decision is made here, the processing from step 300 onward is repeated, while if an affirmative decision is made, this processing is once terminated.

This makes it possible to accurately determine whether the target object to be measured is a stationary object or a moving object. g) As described above, in this embodiment, the vehicle speed sensor 3
A plurality of frequency shift amounts (Sn-1,
Sn, Sn + 1) is set, and the true frequency shift amount T is selected from among them by using the evaluation function composed of the equations (6) to (9).
Sn is calculated, and the actual vehicle speed TVB is calculated based on the true frequency shift amount TSn.

Therefore, as in the conventional case, the vehicle speed sensor 3b is simply used.
A more accurate actual vehicle speed TVB can be obtained than the own vehicle speed VB obtained from the above. Therefore, by using this accurate actual vehicle speed TVB, various vehicle controls using the own vehicle speed for control, such as anti-skid control and constant speed cruise, can be more accurately and suitably performed.

Further, there is an advantage that more accurate stationary object determination can be performed by using the accurate actual vehicle speed TVB for stationary object determination. It is needless to say that the present invention is not limited to the above-mentioned embodiments and can be implemented in various modes without departing from the technical scope of the present invention.

(1) For example, although the FMCW radar device and the vehicle control device have been described in the above embodiments, a recording medium storing means for executing various processes and controls by these devices is also within the scope of the present invention. Is. Examples of the recording medium include various recording media such as an electronic control device configured as a microcomputer, a microchip, a floppy disk, a hard disk, and an optical disk.

(2) As the evaluation value, only the amplitude evaluation value Y is used without using the phase difference evaluation value X, and its absolute value is | V.
It may be p |.

[Brief description of drawings]

FIG. 1 is a block diagram showing an overall configuration of a radar device and a vehicle control device of this embodiment.

FIG. 2 is a block diagram showing the overall configuration of a radar device of this embodiment.

FIG. 3 is a graph showing a change in frequency of a transmission signal.

FIG. 4 is an explanatory diagram showing data stored in a RAM.

FIG. 5 is a flowchart showing an obstacle detection process.

FIG. 6 is an explanatory diagram showing correction based on a beam direction of a laser radar.

FIG. 7 is an explanatory diagram showing evaluation values.

FIG. 8 is an explanatory diagram showing spectra of a rising portion and a falling portion.

FIG. 9 shows spectra of a rising part and a falling part, (a)
Is an explanatory diagram showing a case where VB is used, and (b) is an explanatory diagram showing a case using TVB.

FIG. 10 is an explanatory diagram showing spectra when shifting is performed with three types of shift widths.

FIG. 11 is a flowchart showing an actual vehicle speed estimation process of the embodiment.

FIG. 12 is an explanatory diagram showing a method of setting a moving object prediction flag.

FIG. 13 is a flowchart showing a stationary object determination process of the embodiment.

FIG. 14 is an explanatory diagram showing the principle of an FMCW radar.

FIG. 15 is an explanatory diagram showing a beat signal spectrum by the FMCW radar.

FIG. 16 is an explanatory diagram showing a cause of an error in vehicle speed measurement.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 ... Vehicle control device 2 ... Radar device 3 ... Sensor 3a ... Wheel speed sensor 3b ... Vehicle speed sensor 10 ... Transceiver 12 ... Transmitter 12a ... Modulator 12b ... Voltage controlled oscillator 12c, 12d ... Electric power distributor 12e ... Transmitting antenna 14 , 16 ... Receivers 14a, 16a ... Receiving antennas 14b, 16b ... Mixers 14c, 16c ... Preamplifiers 14d, 16d ... Low pass filters 14e, 16e ... Postamplifier 20 ... Signal processing unit 22 ... Triangular wave generators 24a, 24b ... A
/ D converter 26 ... Microcomputer 28 ... Arithmetic processing device

─────────────────────────────────────────────────── ─── Continuation of the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) G01S ⁷ /00-7/42 G01S 13/00-13/95

Claims (15)

(57) [Claims] 1. A transmission unit that generates a transmission signal having a frequency that gradually increases and decreases periodically with a predetermined modulation width and transmits the signal as a radar wave; and a reception signal that receives the radar wave reflected by a target and outputs the reception signal. Receiving means for generating a beat signal by mixing the received signal with the transmission signal as well as generating an upstream spectrum from an upstream beat signal at the time of upstream modulation in which the frequency of the transmission signal increases; A spectrum creating means for creating a downlink spectrum from a downlink beat signal at the time of downlink modulation in which the frequency of the signal falls, and for comparison of the peaks of the both spectra, based on the vehicle speed measured by the speed measuring means, the frequency A frequency shift amount setting means for setting a shift amount, and at least one of the two spectra is shifted by the frequency shift amount, In an FMCW radar device comprising: a detecting means for comparing the corresponding peaks with each other to detect the moving state of the target; and a plurality of frequency shifts considering a measurement error when setting the frequency shift amount. A plurality of shift amount setting means for setting the amount, for each of the uplink spectrum and the downlink spectrum corresponding to each of the set frequency shift amount, evaluation means for evaluating the spectrum matching degree, based on the evaluation result, the Determining means for determining the frequency shift amount with the highest spectrum matching degree, and estimating means for estimating the actual vehicle speed by correcting the own vehicle speed measured by the speed measuring means using the determined frequency shift amount. An FMCW radar device comprising: Wherein said estimating means, the claim 1, characterized in that for estimating the actual vehicle speed based on the following formula (A)
The FMCW radar device according to 1. TVB = TSn * C / (4 * f0) (A) where TVB is the actual vehicle speed, TSn is the determined frequency shift amount, f0 is the center frequency of the transmission signal, and C is the speed of light. 3. The FMCW radar device according to claim 1, wherein the plurality of frequency shift amounts are set based on the actual vehicle speed estimated by the estimating means. 4. The actual relative acceleration is obtained from the difference between the previously obtained actual vehicle speed and the currently obtained actual vehicle speed, the next own vehicle speed is predicted using the actual relative acceleration, and the prediction is performed. The FMC according to claim 3, wherein the next frequency shift amount is set by using the next own vehicle speed.
W radar device. 5. When setting the frequency shift amount based on the vehicle speed measured by the speed measuring means,
The FMCW radar device according to claim 1, wherein a basic frequency shift amount and a frequency shift amount deviated from the basic frequency shift amount by a predetermined amount are set. 6. The FMCW radar device according to claim 1, wherein the frequency shift amount is set in consideration of the beam direction of the FMCW radar device. 7. When using an upstream spectrum and a downstream spectrum corresponding to the determined frequency shift amount,
For each pair of corresponding spectral peaks of both spectra,
The FMCW radar device according to any one of claims 1 to 6, wherein a spectrum matching degree is evaluated, and a stationary object of each target is determined based on the evaluation result. 8. A moving object prediction flag is set by predicting a current moving position for a target already recognized as a moving object, and the moving object prediction flag is set for a spectrum peak to be judged this time. 8. If is set, the target is not determined to be a stationary object.
The FMCW radar device according to 1. 9. The actual speed of the vehicle obtained by the estimation is compared with the relative speed of the target recognized by the FMCW radar device, and the comparison result is reflected in the stationary object determination. The FMCW according to claim 7 or 8.
Radar equipment. 10. The method according to claim 1, wherein the evaluation of the degree of spectrum matching is performed based on at least the amplitude of the spectrum peak among the amplitude of the spectrum peak and the azimuth information of the target. The described FMCW radar device. 11. The spectrum matching degree is evaluated based on the absolute value of the evaluation vector, where Y is an amplitude evaluation value based on the amplitude of the spectrum peak and X is a phase evaluation value based on the azimuth information of the target. The FMCW radar device according to claim 10, which is performed. 12. The neighborhood sum of absolute values of the evaluation vector for the spectrum peak to be evaluated is obtained for each spectrum peak, and the sum of the neighborhood sums of the spectrum total sums is the minimum. The FMCW radar device according to claim 11, wherein a shift amount is used. 13. Comparing a neighborhood sum of absolute values of the evaluation vector relating to the spectral peak to be evaluated with a predetermined threshold value, and determining the target as a stationary object when the neighborhood sum is less than or equal to a threshold value. 11. The above-mentioned claim 11 or
12. The FMCW radar device according to item 12 . 14. A recording medium storing means for executing processing by the FMCW radar device according to any one of claims 1 to 13. 15. A vehicle control device, wherein the actual vehicle speed estimated according to any one of claims 1 to 14 is used for various vehicle controls.