JP2003177177A - Data processing device of radar - Google Patents

Abstract

<P>PROBLEM TO BE SOLVED: To improve the followability when an actual relative speed fluctuates more greatly than a fixed level during extrapolation processing. <P>SOLUTION: If the extrapolation processing is not executed this time but was executed the last time (step 1,010, 1,012) after continuity determination (step 1,008), a filter operation of the relative speed is executed not this time (step 1,014) but another time (step 1,016). <P>COPYRIGHT: (C)2003,JPO

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a radar, and more particularly to a data processing device for a millimeter wave radar. 2. Description of the Related Art A millimeter wave radar (FM-CW radar) radiates forward a continuous wave beam that is FM-modulated by a triangular wave that alternately rises and falls, and reflects a reflected signal from a target. Is mixed with a part of the transmission signal to generate a beat signal, and a peak on the frequency spectrum is associated with a peak when the frequency modulation rises and a peak when the frequency modulation falls (hereinafter, referred to as pairing). Then, the distance and the relative speed of each target are calculated from the sum and difference of the frequencies of the paired peak pairs. Further, by scanning the beam angle physically or electronically, the direction in which each target exists can be known. [0003] In the above pairing process, a peak (up peak) in a beat signal (up beat) at the time of rising frequency modulation and a peak (down peak) in a beat signal (down beat) at the time of falling frequency modulation are used. Then, for example, those having close angles and intensities are associated as peaks based on the same target. In addition, continuity determination is performed to determine that the data is from the same target by comparing the data with past data, and the distance and relative speed determined to be from the same target are determined. Is subjected to a filter operation by the following equation, for example. [Relative speed] = ([previous value] × 3 + [current value]) / 4 [distance] = ([previous value] + [current value]) / 2 The past correspondence pairing that preferentially associates peaks that exist in the frequency range where the target is expected to exist at the position and speed predicted from, and the pairing that remains after the past correspondence pairing ends There is a new pairing to do. In the past correspondence pairing, even if one or both peaks are not temporarily observed in the prediction range, extrapolation is performed within a certain period on the assumption that the target exists at the prediction position. In addition, when the target relatively approaches, the up peak having a low frequency often disappears. Therefore, the extrapolation at this time is particularly called downbeat extrapolation. [0006] The predicted value of the target data is calculated by the following equation, for example, assuming that the previous relative speed continues this time. [Distance] = [Previous distance] + [Previous relative speed] × [Time] [Relative speed] = [Previous relative speed] [0008] FIG. FIG. 6 shows a temporal change in the frequency of the down peak 10 and the up peak 12 until the traveling target approaches and runs a fixed distance after a little distance. The sum of the two frequencies corresponds to the distance to the target, and the difference corresponds to the relative speed. As shown in FIG. 2, it is assumed that at time t ₁ , the frequency of the up-peak is lowered and cannot be detected, and extrapolation of the down-beat is started. Since the distance is the sum of the two frequencies, it can be represented by a dashed line with a reference numeral 14 between the two frequencies if the scale is appropriately taken. As described above, the estimated position of the target is calculated with the relative speed being constant at the time of downbeat extrapolation. Therefore, even if the filter operation is performed, the estimated value (14) of the distance decreases substantially linearly, and The estimated frequency 16 and the estimated frequency 18 of the up-peak decrease almost linearly with almost the same slope. Therefore the actual frequency difference is reduced relative velocity decreases, new pair a decrease of the frequency of the down-peak level off the difference between the estimated value spread, eventually down peak at time t ₂ also outside the expected range It becomes a ring. At time t ₃ , even if an up-peak is detected, down-beat extrapolation starts again because it is still outside the prediction range. Thereafter, at time t ₄ , the down peak is out of the prediction range and new pairing is performed. However, at time t ₅ , the up peak is out of the prediction range, and down beat extrapolation starts again. As described above, when the actual relative speed fluctuates by a certain amount or more during the extrapolation process, there is a problem that it takes time until the output value reflects the actual value. Accordingly, an object of the present invention is to provide a radar data processing apparatus capable of quickly returning even if the actual relative speed fluctuates during extrapolation processing. According to the radar data processing apparatus of the present invention, the target data obtained up to the previous time and the target data obtained this time are based on the same target. Means for performing continuity determination with respect to the target, and means for performing a filter operation on target data determined to be based on the same target as a result of the continuity determination. When the data is calculated from the actual measurement values obtained this time and the previous target data is not the actual measurement value but calculated by extrapolation, the relative speed of the target included in the target data Is characterized in that no filter operation is performed for. The reason why it takes a long time to return is that the estimated value becomes too close to the actual value due to the filter calculation of the relative speed even if new pairing is performed after the extrapolation process as seen in the example of FIG. Because there is no. Therefore, when the new pairing is performed this time and the extrapolation processing was performed last time, the filter operation of the relative speed is not performed, so that it is possible to quickly return. FIG. 3 shows a configuration of an example of a millimeter wave radar device to which the present invention is applied. In FIG. 3, the transmitter control circuit 20 outputs a triangular wave that alternates between rising and falling. In the RF unit 22, a millimeter wave frequency-modulated by the triangular wave is generated.
Radiated from 4. The reflected wave from the target is received by the antenna 24 and mixed with a part of the transmission signal in the RF unit 22 to generate a beat signal having the frequency of the difference between the transmission signal and the reflection signal. The generated beat signal is converted to a baseband in the receiving circuit 26 and then input to the DSP 28. In DSP28,
The conversion from the time domain to the frequency domain is performed by performing an FFT operation on each of the rising section (upbeat) and the falling section (downbeat) of the triangular wave. The DSP 28 further extracts a peak (up-peak) and a peak (down-peak) in the frequency spectrum of the rising section and sends it to the microcomputer 20. On the other hand, the drive circuit 32
Drives the motor 34 to scan the direction of the antenna 24,
The angle data at that time is input to the microcomputer 30. In microcomputer 30, one-to-one between each of the up-peaks and each of the down-peaks based on the intensity of each peak, the angle of the antenna at which it was obtained, and the continuity with the previous data. (Pairing processing) is performed. Paired 2
The distance to each target is calculated from the sum of the frequencies of the two peaks, and the relative speed with each target is calculated and output from the difference between the frequencies of the two peaks. FIG. 4 is a schematic flowchart of an example of processing from input of peak data to output of data in the microcomputer 30. In FIG. 4, first, peak data is input from the DSP 28 (step 1000). A predicted value of the target data (distance, relative speed) of the current time and a predicted value of the peak frequency are calculated from the previous data (step 1002, details are described above), and pairing corresponding to the past is attempted based on the calculated values (step 1004, details). Is mentioned above). Next, a new pairing is attempted for the remaining peak (step 1006, details described above). The distance and the relative speed are calculated from the frequency of the paired peak, and the continuity with the previous data is determined (step 1008, details described above). If there is no data in the prediction range from the previous data, target data is calculated by extrapolation for a certain period. This time, instead of extrapolation (step 10
10) When the target data was calculated by extrapolation the previous time (step 1012), the current value is used as the final value of the current relative speed without performing the filtering operation on the relative speed (step 1014). In other cases, a filtering operation is performed for the relative speed (step 1016). FIG. 5 is a diagram for explaining the operation in the situation described in FIG. 1 and FIG. 2 is the same as FIG. ₂ until time t ₂ , but when a new pairing is performed at t ₂ , the predicted value of the peak frequency is closer to the actual value because the filtering operation is not performed for the relative speed.
Even downbeat extrapolation is resumed at time t _3, becomes within the expected range from the new pairing later time t ₄ since the predicted value is closer to the measured value up peak, down peak both the past correspondence paired It becomes possible, and the predicted value converges to the actually measured value. As described above, according to the present invention, as in the case where a vehicle traveling ahead approaches and then travels at a constant inter-vehicle distance, the actual relative position is determined during extrapolation processing. Provided is a radar data processing device that can quickly follow even when the speed fluctuates beyond a certain value.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a diagram showing a transition of a peak frequency when a target in front travels at a constant inter-vehicle distance after approaching. FIG. 2 is a diagram illustrating a problem of the related art. FIG. 3 is a block diagram illustrating an example of a configuration of a millimeter wave radar to which the present invention is applied. FIG. 4 is a flowchart showing processing of the microcomputer 30. FIG. 5 is a diagram showing a processing result of data processing according to an embodiment of the present invention.

   ────────────────────────────────────────────────── ─── Continuation of front page    (72) Inventor Yasuhiro Sekiguchi             1-2-2, Goshodori, Hyogo-ku, Kobe City, Hyogo Prefecture               Fujitsu Ten Limited F term (reference) 5J070 AB19 AB24 AC02 AC06 AD01

Claims (1)

Claims: 1. A means for determining continuity that target data obtained up to the previous time and target data obtained this time are based on the same target, Means for performing a filter operation on the target data determined to be based on the same target as a result of the gender determination, wherein the filter operation means calculates the current target data from the actually measured values obtained this time. And when the previous target data is not actually measured values but calculated by extrapolation, the filter operation is not performed on the relative speed of the targets included in the target data. Radar data processing device.