JP2004069467A - Object detector for vehicle - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the detection accuracy of an object detector for vehicle by accurately discriminating a peak signal caused by an object from another peak signal caused by noise. <P>SOLUTION: The object detector for vehicle uses an FM-CW wave, generates a beat signal by mixing a transmission wave with a reception wave by means of a mixer, and discriminates a peak signal having a detection threshold of ≥Th1 of peak signals obtained by performing frequency analysis on the beat signal as the detection peak of the object. Since the detector extracts noise regions Fa, Fc, and Fe obtained by removing peak regions Fb and Fd and sets a new detection threshold Th2 by adding a margin THRESH MARGIN to the mean value Lave of the detection levels of the noise regions Fa, Fc, and Fe, this device can successively update the detection threshold Th1 in accordance with the state of the latest noise level, and can be improved in detection accuracy by surely discriminating the detection peak from the background noise. <P>COPYRIGHT: (C)2004,JPO

Description

[0001]
TECHNICAL FIELD OF THE INVENTION
The present invention relates to an object detection device for a vehicle for detecting an object such as a preceding vehicle with a radar device using an FM-CW wave (frequency modulated continuous wave).
[0002]
[Prior art]
Such an object detection device for a vehicle is known from Japanese Patent Application Laid-Open No. 11-38129.
[0003]
As shown in FIG. 4, in the conventional object detection device using the FM-CW wave, the oscillation operation of the oscillator 3 is modulated and controlled by the FM modulation control circuit 2 based on the timing signal input from the timing signal generation circuit 1. 5A, a transmission wave whose frequency is modulated in a triangular waveform is transmitted from the transmission / reception antenna 6 via the amplifier 4 and the circulator 5. As shown in FIG. When the FM-CW wave is reflected by an object such as a preceding vehicle and the reflected wave is received by the transmission / reception antenna 6, the received wave is, for example, as shown in FIG. ), The rising side where the frequency of the transmitting wave increases linearly appears at a lower frequency than the transmitting wave and lags behind the transmitting wave, and the descending side where the frequency of the transmitting wave decreases linearly as shown by the broken line. Appears later than the transmitted wave at a higher frequency than the transmitted wave.
[0004]
The received wave received by the transmitting / receiving antenna 6 is input to the mixer 7 via the circulator 5. The mixer 7 receives, via an amplifier 8, a scanning wave distributed from a transmission wave output from the oscillator 3, in addition to the reception wave from the circulator 5, via the amplifier 8. The mixer 7 mixes the transmission wave and the reception wave. As a result, as shown in FIG. 5B, the peak frequency Fup is on the rising side where the frequency of the transmission wave increases linearly, and the peak frequency Fup is on the falling side where the frequency of the transmission wave decreases linearly. A beat signal having Fdn is generated.
[0005]
The beat signal obtained by the mixer 7 is amplified to a required level of amplitude by the amplifier 9, A / D converted by the A / D converter 10 at each sampling time, and the digitized amplified data is stored in the memory 11 in time series. Is stored. A timing signal from the timing signal generation circuit 1 is input to the memory 11, and in response to the timing signal, the memory 11 stores data for each of the rising side where the frequency of the transmission / reception wave increases and the falling side where the frequency decreases. Will be stored.
[0006]
The data stored and held in the memory 11 is input to a CPU 12 including a frequency analysis unit 13, a detection peak determination unit 14, and an object detection unit 15, and the CPU 12 executes an arithmetic process based on the input data.
[0007]
The frequency analysis means 13 performs frequency analysis of the beat signal data stored in the memory 11 to obtain a spectrum distribution, and an FFT (Fast Fourier Transform) is used as a frequency analysis method.
[0008]
The detection peak determination means 14 detects a spectrum (peak signal) having a maximum value when the detection level is equal to or higher than a predetermined detection threshold, based on the spectrum data obtained by the frequency analysis by the frequency analysis means 13. The rising-side peak signal shown in FIG. 6A and the falling-side peak signal shown in FIG. 6B are detected symmetrically with respect to the peak position when the relative speed with respect to the object is “0”. Is done.
[0009]
The object detecting means 15 calculates the relative distance and the relative speed of the object based on the ascending peak frequency Fup and the descending peak frequency Fdn obtained by the detected peak determining means 14. Specifically, the distance to the object is calculated based on the sum of the two peak frequencies Fup and Fdn, and the relative speed with the object is calculated based on the difference between the two peak frequencies Fup and Fdn.
[0010]
[Problems to be solved by the invention]
Meanwhile, the detection level detected by the above-described conventional object detection device decreases as the distance between the vehicle and the object increases (see FIG. 8). FIG. 7A shows spectral data at time t. An object moving away from the own vehicle has a peak signal Pa1 equal to or higher than a detection threshold indicated by a dashed line. The peak signals Pc1 and Pc2 are smaller than the detection threshold. FIG. 7B shows the spectrum data at time t + 1. Since the peak signal Pa1 of the object drops below the detection threshold indicated by the dashed line due to the increase in the distance to the object, the object is There is a problem that cannot be detected. If the detection threshold is lowered to a level at which the object can be detected in order to solve such a problem, the object can be detected, but the other two peak signals Pc1 and Pc2 due to noise are erroneously recognized as the object. New problems arise.
[0011]
The present applicant has already proposed a method for solving the above problem in Japanese Patent Application No. 2001-108925, but the present invention solves the above problem by a method different from that in the above Japanese Patent Application No. 2001-108925.
[0012]
That is, an object of the present invention is to accurately identify a peak signal due to an object and a peak signal due to noise to enhance the detection accuracy of the object detection device.
[0013]
[Means for Solving the Problems]
In order to achieve the above object, according to the invention described in claim 1, a transmitting / receiving means for transmitting an FM-CW wave and receiving a reflected wave of the FM-CW wave from an object, a transmitting wave and a receiving means A mixer for mixing the waves to generate a beat signal, frequency analysis means for frequency-analyzing the beat signal obtained by the mixer, and a peak signal obtained based on the frequency analysis results on the ascending and descending sides by the frequency analyzing means. Among them, a detection peak determination unit that determines a peak signal equal to or greater than a detection threshold as a detection peak, and a distance to the object and a relative speed to the object based on the detection peaks on the ascending and descending sides obtained by the detection peak determination unit. In a vehicle object detection device including an object detection unit that calculates at least one, a frequency domain extraction unit that extracts a frequency domain obtained by removing the detection peak, Object detecting apparatus for a vehicle characterized by comprising a detection threshold changing means for changing the detection threshold based on the signal strength of the frequency domain obtained by the wavenumber region extracting means is proposed.
[0014]
According to the above configuration, since the detection threshold is changed based on the signal intensity in the frequency domain obtained by excluding the detection peak, the detection threshold can be sequentially updated according to the latest noise level, and the detection peak can be changed. Can be reliably identified from the background noise, and the detection accuracy of the object detection device can be increased.
[0015]
According to the second aspect of the present invention, in addition to the configuration of the first aspect, the frequency domain extracting means extracts the frequency domain based on a detected peak position and a predetermined detected peak width. A vehicle object detection device characterized by the following is proposed.
[0016]
According to the above configuration, the frequency region obtained by excluding the detected peak is extracted based on the detected peak position and the predetermined detected peak width, so that the frequency region of the background noise excluding the detected peak is accurately extracted. can do.
[0017]
According to a third aspect of the present invention, in addition to the configuration of the first or second aspect, the detection threshold value changing unit is configured to detect the detection threshold value based on an average value of the frequency spectrum in the extracted frequency domain. Is proposed.
[0018]
According to the above configuration, since the detection threshold is changed based on the average value of the frequency spectrum in the extracted frequency domain, the detection threshold is changed without being affected by the fluctuation of the background noise or the peak-like noise which differs for each frequency. Can be performed accurately.
[0019]
According to a fourth aspect of the present invention, in addition to the configuration of the third aspect, there is provided an abnormality detecting means for storing at least one of an upper limit value for signal determination and a lower limit value for signal determination. An object detection device for a vehicle is proposed, wherein an abnormality is determined when the average value of the frequency spectrum is equal to or more than the upper limit for signal determination or equal to or less than the lower limit for signal determination.
[0020]
According to the above configuration, the abnormality determination is performed when the average value of the frequency spectrum in the extracted frequency domain is equal to or more than the upper limit for signal determination or equal to or less than the lower limit for signal determination, so that the average of the frequency spectrum indicating an abnormal value is determined. It is possible to prevent the detection threshold from being inappropriately changed based on the value.
[0021]
The transmitting / receiving antenna 6 of the embodiment corresponds to the transmitting / receiving means of the present invention, the noise regions Fa, Fb, and Fc of the embodiment correspond to the frequency region of the present invention, and the distance average value Ln of the embodiment corresponds to the frequency of the present invention. Corresponds to the average value of the spectrum.
[0022]
BEST MODE FOR CARRYING OUT THE INVENTION
Hereinafter, embodiments of the present invention will be described based on examples of the present invention shown in the accompanying drawings.
[0023]
1 to 3 show an embodiment of the present invention. FIG. 1 is an overall configuration diagram of an object detection device, FIG. 2 is a diagram illustrating a method of changing a detection threshold, and FIG. 3 is a flowchart illustrating an operation. It is.
[0024]
FIG. 1 shows a circuit configuration of the CPU 12 of the object detection device according to the present embodiment. The configuration and operation other than the CPU 12 are the same as those in FIG. The conventional CPU 12 described with reference to FIG. 4 includes a frequency analysis unit 13, a detection peak determination unit 14, and an object detection unit 15, but the CPU 12 of the present embodiment further includes a frequency domain extraction unit 16, a detection threshold A change unit 17 and an abnormality detection unit 18 are provided.
[0025]
In step S1 of the flowchart of FIG. 3, as shown in FIG. 2A, the detection spectrum analyzed by the frequency analysis unit 13 is compared with the detection threshold Th1, and the peak position of the detection peak at which the detection level is equal to or higher than the detection threshold Th1 Is extracted. In the following step S2, peak areas Fb and Fd obtained by adding a preset detection peak width before and after the peak position of the detection peak are defined as peak areas Fb and Fd, and frequency areas excluding the peak areas Fb and Fd are defined as noise areas Fa, Fc and Extract as Fe. As described above, since the peak areas Fb and Fd are set based on the detected peak position and the predetermined detected peak width, the noise areas Fa, Fc and Fe obtained by excluding the peak areas Fb and Fd are accurately extracted. be able to. In the following step S3, as shown in FIG. 2B, a distance average value Ln (T) at a time T of a detection level (noise level) in the extracted noise regions Fa, Fc, and Fe is calculated. As described above, by using the distance average value Ln (T) of the noise levels in the extracted noise regions Fa, Fc, and Fe, the variation in the background noise and the peak-like noise that are different for each frequency (distance) are affected. Instead, the detection threshold Th1 described later can be changed accurately. The above steps S1 to S3 are executed by the frequency domain extracting means 16.
[0026]
In the following step S4, the distance average value Ln (T) of the noise level calculated in step S3 is set to the signal determination upper limit MAX. 　 NOISE 　 LEVEL and signal determination lower limit MIN 　 NOISE 　 Compared with LEVEL, the distance average value Ln (T) of the noise level is the signal determination upper limit value MAX. 　 NOISE 　 LEVEL or more, or signal judgment lower limit MIN 　 NOISE 　 If it is lower than LEVEL, it is determined in step S7 that the noise level is abnormal. As described above, since the abnormality determination is performed when the distance average value Ln (T) of the noise level indicates an abnormal value, the detection threshold Th1 can be changed based on the distance average value Ln (T) indicating the abnormal value. Can be prevented. The steps S4 and S7 are executed by the abnormality detecting means 18.
[0027]
In step S4, the distance average value Ln (T) of the noise level is set to the signal determination upper limit MAX. 　 NOISE 　 LEVEL and signal determination lower limit MIN 　 If it is between NOISELEVEL, in step S5, the noise level time average value Lave (T-1) at the previous sampling time (time T-1) and the noise level distance average value at the current sampling time (time T). Using the value Ln (T) and the weighting coefficient K (0 <K <1), the time average value Lave (T) of the noise level at the time of the current sampling is calculated as
Lave (T) = K × Ln (T) + (1−K) × Lave (T−1)
It is calculated by: As described above, by calculating the time average value of the noise level Lave (T), the fluctuation of the noise level due to the influence of the temporary noise or the like can be reduced.
[0028]
Then, in step S6, a margin THRESH set in advance to the time average value Lave (T) of the current noise level is set. 　 MARGIN is added to calculate a new detection threshold Th2 (see FIG. 2B). As described above, since the detection threshold value Th1 is changed based on the noise levels of the noise regions Fa, Fc, and Fe obtained by excluding the peak regions Fa and Fc, the detection threshold value Th1 is sequentially updated according to the latest noise level situation. This makes it possible to reliably identify the detection peak from the background noise, thereby improving the detection accuracy of the object detection device. The steps S5 and S6 are executed by the detection threshold value changing means 17.
[0029]
The embodiments of the present invention have been described above. However, various design changes can be made in the present invention without departing from the gist thereof.
[0030]
For example, in the embodiment, the abnormality detection unit 18 determines that the signal determination upper limit MAX 　 NOISELEVEL and signal determination lower limit MIN 　 NOISE 　 Although the abnormality determination is performed using both LEVEL, the abnormality determination may be performed using either one of them.
[0031]
Further, a configuration in which a circulator is not used may be adopted by separately installing a transmitting antenna and a receiving antenna.
[0032]
【The invention's effect】
As described above, according to the invention described in claim 1, since the detection threshold is changed based on the signal strength in the frequency domain obtained by excluding the detection peak, the detection threshold is changed according to the latest noise level situation. The update can be performed sequentially, and the detection peak can be reliably identified from the background noise, and the detection accuracy of the object detection device can be improved.
[0033]
According to the second aspect of the present invention, the frequency region obtained by excluding the detected peak is extracted based on the detected peak position and the predetermined detected peak width, so that the background noise excluding the detected peak is extracted. Can be accurately extracted.
[0034]
According to the third aspect of the present invention, since the detection threshold is changed based on the average value of the frequency spectrum in the extracted frequency domain, the variation of the background noise and the peak-like noise which are different for each frequency are affected. Instead, the detection threshold can be changed accurately.
[0035]
According to the invention described in claim 4, when the average value of the frequency spectrum in the extracted frequency domain is equal to or more than the upper limit for signal determination or equal to or less than the lower limit for signal determination, abnormality determination is performed. It is possible to prevent the detection threshold from being inappropriately changed based on the average value of the frequency spectrum indicating the value.
[Brief description of the drawings]
FIG. 1 is an overall configuration diagram of an object detection device. FIG. 2 is a diagram illustrating a method of changing a detection threshold. FIG. 3 is a flowchart illustrating an operation. FIG. 4 is an overall configuration diagram of a conventional object detection device. FIG. 6 is a graph showing the waveform and peak frequency of a transmitted / received wave when an object is moving close to the transmitting / receiving antenna. FIG. 6 is a graph showing a peak signal detected by a detected peak determining means. FIG. 8 is a graph showing the relationship between the distance between the vehicle and the object and the detection level.
6 transmitting / receiving antenna (transmitting / receiving means)
7 Mixer 13 Frequency analysis means 14 Detection peak determination means 15 Object detection means 16 Frequency domain extraction means 17 Detection threshold change means 18 Abnormality detection means Fa Noise area (frequency area)
Fc noise domain (frequency domain)
Fe noise area (frequency area)
Ln Distance average value (average value of frequency spectrum)
MAX 　 NOISE 　 LEVEL signal judgment upper limit MIN 　 NOISE 　 LEVEL signal determination lower limit value Th1 detection threshold value

Claims (4)

Transmitting and receiving means (6) for transmitting an FM-CW wave and receiving a reflected wave of the FM-CW wave from an object;
A mixer (7) for mixing the transmission wave and the reception wave to generate a beat signal;
Frequency analysis means (13) for frequency-analyzing the beat signal obtained by the mixer (7);
Detection peak determination means (14) for determining, as a detection peak, a peak signal equal to or greater than a detection threshold (Th1) among peak signals obtained based on the results of frequency analysis on the ascending and descending sides by the frequency analysis means (13);
Object detection means (15) for calculating at least one of the distance to the object and the relative speed to the object based on the detection peaks on the ascending and descending sides obtained by the detection peak determination means (14);
In an object detection device for a vehicle having
Frequency domain extraction means (16) for extracting a frequency domain (Fa, Fc, Fe) obtained by excluding the detection peak;
A detection threshold value changing unit (17) for changing the detection threshold value (Th1) based on the signal strength in the frequency domain (Fa, Fc, Fe) obtained by the frequency domain extraction unit (16);
An object detection device for a vehicle, comprising: The vehicle according to claim 1, wherein the frequency domain extracting means (16) extracts the frequency domain (Fa, Fc, Fe) based on a detected peak position and a predetermined detected peak width. Object detection device. The said detection threshold value change means (17) changes the said detection threshold value (Th1) based on the average value (Ln) of the frequency spectrum in the extracted frequency area (Fa, Fc, Fe). The vehicle object detection device according to claim 1 or 2. Upper limit for signal judgment (MAX 　 NOISE 　 LEVEL) and the signal determination lower limit (MIN) 　 NOISE 　 LEVEL), and an abnormality detecting means (18) for storing at least one of
The abnormality detecting means (18) determines that the average value (Ln) of the frequency spectrum is equal to the signal determination upper limit (MAX). 　 NOISE 　 LEVEL) or higher or the lower limit for signal judgment (MIN) 　 NOISE 　 4. The object detection device for a vehicle according to claim 3, wherein an abnormality is determined when the difference is equal to or less than (LEVEL).

Images