JP2019066280A - Radar device and target detection method - Google Patents

Abstract

To provide a radar device and a target detection method with which it is possible to improve the accuracy of predicting particle data for a new target.SOLUTION: The radar device according to an embodiment comprises a detection unit and a filter processing unit. The detection unit detects an instantaneous value that corresponds to a target on the basis of a frequency-modulated transmitted wave and the reflected wave of the transmitted wave by the target. The filter processing unit applies a particle filter that correlates a prescribed number of particle data to the instantaneous value detected by the detection unit and thereby generates target data that corresponds to the instantaneous value. The filter processing unit also assumes a position, of a new target in which previous particle data is not allocated to the previous instantaneous value, that corresponds to a perpendicular to a vector directed to the own vehicle from the target as the latest predicted position of particle data on the basis of the relative speed of the previous instantaneous value.SELECTED DRAWING: Figure 2

Description

  The present invention relates to a radar device and a target detection method.

  Conventionally, there is a radar device which is mounted on, for example, a vehicle, receives a reflected wave in which a transmission wave transmitted from the vehicle strikes a target and receives the reflected wave, and analyzes the received signal to periodically detect the target (for example, Patent Document 1).

  In such a radar device, by analyzing reflection points detected based on the transmission wave and the reflection wave, an instantaneous value including elements such as the relative velocity of the target and the distance to the target is generated. Then, the radar device generates target data corresponding to the instantaneous value by applying a filtering process to the obtained instantaneous value.

  Moreover, it is known to use a particle filter in the above-mentioned filter processing. In the particle filter, prediction processing is performed to predict current particle data using particle data resampled in the previous cycle.

Unexamined-Japanese-Patent No. 2015-210157

  However, the prior art has room for improvement in that it improves the prediction accuracy of particle data for newly detected targets. Specifically, in the case of a target detected for the first time in the previous cycle, since there is no previous particle data, it is assumed that the previous instantaneous value moves in the vertical direction, which is the traveling direction of the vehicle, I was predicting. Therefore, when the target actually moves in the lateral direction which is the vehicle width direction, there is a possibility that the prediction error of the particle data becomes relatively large.

  The present invention is made in view of the above, and an object of the present invention is to provide a radar device and a target detection method capable of improving the prediction accuracy of particle data to a new target.

  In order to solve the problems described above and achieve the object, a radar apparatus according to the present invention includes a detection unit and a filter processing unit. The detection unit detects an instantaneous value corresponding to the target based on the frequency-modulated transmission wave and a reflected wave of the transmission wave by the target. The filter processing unit generates target data corresponding to the instantaneous value by applying a particle filter that assigns a predetermined number of particle data to the instantaneous value detected by the detection unit. In the new target to which the previous particle data can not be assigned to the previous instantaneous value, the filtering processing unit may, based on the relative velocity of the previous instantaneous value, determine the direction from the target to the vehicle. The position corresponding to the perpendicular to the vector of is the predicted position of the particle data at this time.

  According to the present invention, it is possible to improve the prediction accuracy of particle data for a new target.

FIG. 1A is a diagram showing an outline of a target detection method according to an embodiment. FIG. 1B is a diagram showing an outline of a target detection method according to an embodiment. FIG. 2 is a block diagram showing the configuration of the radar device according to the embodiment. FIG. 3 is an explanatory diagram of processing from pre-processing of the signal processing unit to peak extraction processing in the signal processing unit. FIG. 4A is a process explanatory diagram of the angle estimation process. FIG. 4B is a process explanatory diagram of the pairing process (part 1). FIG. 4C is a process explanatory diagram of the pairing process (part 2). FIG. 5 is a block diagram showing the configuration of the filter processing unit. FIG. 6 is a diagram showing the processing content of the prediction unit. FIG. 7 is a diagram showing the processing content of the prediction unit. FIG. 8 is a diagram showing the processing content of the prediction unit. FIG. 9 is a diagram showing the processing content of the prediction unit. FIG. 10 is a flowchart showing a processing procedure of processing performed by the radar device according to the embodiment. FIG. 11 is a diagram illustrating processing contents of a prediction unit according to a modification.

  Hereinafter, embodiments of a radar device and a target detection method disclosed in the present application will be described in detail with reference to the attached drawings. The present invention is not limited by this embodiment. In the following, although the case where the radar device 1 is an FM-CW (Frequency Modulated Continuous Wave) method will be described as an example, the radar device 1 may be, for example, another method such as an FCM (Fast-Chirp Modulation) method. It may be.

  First, the outline of the target detection method according to the embodiment will be described with reference to FIGS. 1A and 1B. FIG. 1A and FIG. 1B are diagrams showing an outline of a target detection method according to an embodiment. As shown in FIG. 1, the radar device 1 is mounted, for example, in the front grille of the host vehicle MC, and detects a target (for example, a leading vehicle LC, etc.) present in the traveling direction of the host vehicle MC. In addition, the mounting location of the radar apparatus 1 may be mounted in other places, such as a windshield, a rear grille, the side part on either side (for example, the door mirror on either side).

  As shown in FIG. 1A, the radar device 1 according to the embodiment first detects an instantaneous value corresponding to a target. The instantaneous value is a value indicating the state vector of the reflection point detected based on the frequency-modulated transmission wave and the reflection wave of the transmission wave from the target. The state vector includes values such as the distance to the reflection point, the relative velocity, and the angle.

  Subsequently, as shown in FIG. 1A, the radar device 1 according to the embodiment generates target data, which is a filter value corresponding to an instantaneous value, by applying a filtering process to the detected instantaneous value. For example, a particle filter is used for the filtering process.

  The particle filter plots target data by plotting a predetermined number of particle data and an instantaneous value in a predetermined state space and analyzing the positional relationship in the state space. In the particle filter, target data is generated from instantaneous values by performing processing of “prediction”, “allocation”, “weighting”, “resampling” and “data generation”.

  Here, each process of the particle filter will be briefly described. “Prediction” is a process of predicting the distribution state of particle data used in the latest cycle in the state space. Specifically, “prediction” is prediction processing for predicting the distribution state in the current particle data, which is the latest cycle, from the distribution state in the previous particle data, which is the previous cycle.

  “Assignment” is a process of assigning instantaneous values detected in the latest cycle to particle data “predicted”. In “assignment”, for example, instantaneous values present in a predetermined assignment range from the previous target data are assigned to the current particle data.

  "Weighting" is a process of weighting each particle data based on the assigned instantaneous value. “Resampling” is a process of rearranging (resampling) each particle data based on the weight of each “weighted” particle data. “Data generation” is a process of generating the latest target data based on resampled particle data. The details of each process of the particle filter will be described later.

  Here, a conventional target detection method for a newly detected target will be described. The newly detected target refers to a target to which the previous particle data was not assigned to the instantaneous value detected in the previous cycle.

  In the conventional target detection method, when predicting the current particle data in this new target, since the previous particle data does not exist, from the position of the previous instantaneous value in the longitudinal direction which is the traveling direction of the vehicle We assumed that this particle data was assumed to move.

  However, when the new target actually moves in the lateral direction that is the vehicle width direction of the host vehicle, there is a possibility that the prediction error with the current particle data may become relatively large.

  Therefore, in the target detection method according to the embodiment, as shown in FIG. 1B, based on the relative velocity of the previous instantaneous value, the position corresponding to the perpendicular line VL to the vector RV of the direction from the target to the host vehicle MC The predicted position of the particle data 50 of The position of the start point of the vector RV is the position of the target. Here, prediction processing of particle data 50 for a new target will be described using FIG. 1B.

  FIG. 1B shows the previous instantaneous value 100 and the vector RV indicating the relative velocity of the instantaneous value 100 in the new target. As shown in FIG. 1B, in the target detection method according to the embodiment, the state vector SV is generated based on the relative velocity of the instantaneous value 100 so that the particle data 50 is arranged on the perpendicular line VL in the next process. The information of the state vector SV is given to the particle data 50 in the initial state.

  The state vector SV mentioned here assumes that the target corresponding to the instantaneous value 100 moves at the relative speed of the instantaneous value 100, and the vector when moving to other than the direction to the host vehicle MC (ie the direction of the vector RV) is Show. The method of calculating the state vector SV will be described later with reference to FIG.

  The end points of the state vector SV are all located on the perpendicular line VL of the vector RV. That is, in the target detection method according to the embodiment, the particle data 50 is arranged on the perpendicular line VL of the vector RV indicating the relative velocity of the instantaneous value 100 in the next (second) prediction process.

  That is, in the target detection method according to the embodiment, the second particle is assumed to move in various directions instead of narrowing the moving direction of the new target in the vertical direction (y-axis direction) as in the related art. Information of the state vector SV is given in the first process so that the data 50 is arranged.

  For this reason, when the new target moves in the lateral direction (x-axis direction), the prediction error of the particle data 50 can be reduced compared to the conventional case. That is, in the target detection method according to the embodiment, it is possible to improve the prediction accuracy of the particle data 50 for the new target.

  In the target detection method according to the embodiment, the particle data 50 can be arranged around the perpendicular line VL in consideration of acceleration or deceleration of the new target, but such a point will be described later with reference to FIG.

  Next, the configuration of the radar device 1 according to the embodiment will be described in detail with reference to FIG. FIG. 2 is a block diagram showing the configuration of the radar device 1 according to the embodiment. In FIG. 2, only components necessary to explain the features of the present embodiment are represented by functional blocks, and descriptions of general components are omitted.

  In other words, each component illustrated in FIG. 2 is functionally conceptual and does not necessarily have to be physically configured as illustrated. For example, the specific form of the distribution and integration of each functional block is not limited to that shown in the drawings, and all or a part thereof may be functionally or physically distributed in any unit depending on various loads, usage conditions, etc. It is possible to integrate and configure.

  As shown in FIG. 2, the radar device 1 includes a transmitting unit 10, a receiving unit 20, and a processing unit 30. The radar device 1 is connected to a vehicle control device 2 that controls the behavior of the host vehicle MC.

  The vehicle control device 2 performs vehicle control such as PCS (Pre-crash Safety System) or AEB (Advanced Emergency Braking System) based on the detection result of the target by the radar device 1.

  The transmission unit 10 includes a signal generation unit 11, an oscillator 12, and a transmission antenna 13. The signal generation unit 11 generates a modulation signal for transmitting a millimeter wave frequency-modulated by a triangular wave under the control of a transmission / reception control unit 31 described later. The oscillator 12 generates a transmission signal based on the modulation signal generated by the signal generation unit 11 and outputs the transmission signal to the transmission antenna 13. As shown in FIG. 2, the transmission signal generated by the oscillator 12 is also distributed to a mixer 22 described later.

  The transmission antenna 13 converts the transmission signal from the oscillator 12 into a transmission wave, and outputs the transmission wave to the outside of the host vehicle MC. The transmission wave output from the transmission antenna 13 is a continuous wave frequency-modulated by a triangular wave. A transmission wave transmitted from the transmission antenna 13 to the outside of the host vehicle MC, for example, to the front, is reflected by a target such as the preceding vehicle LC to become a reflection wave.

  The receiving unit 20 includes a plurality of receiving antennas 21 forming an array antenna, a plurality of mixers 22, and a plurality of A / D converting units 23. The mixer 22 and the A / D converter 23 are provided for each of the receiving antennas 21.

  Each receiving antenna 21 receives a reflected wave from a target as a received wave, converts the received wave into a received signal, and outputs the signal to the mixer 22. The number of receiving antennas 21 shown in FIG. 2 is four, but may be three or less or five or more.

  The received signal output from the receiving antenna 21 is amplified by an amplifier (for example, a low noise amplifier) (not shown) and then input to the mixer 22. The mixer 22 mixes a part of the distributed transmission signal and the reception signal input from the reception antenna 21 to remove unnecessary signal components, generates a beat signal, and outputs the beat signal to the A / D conversion unit 23. .

  The beat signal is a differential wave between the transmission wave and the reflection wave, and is composed of the frequency of the transmission signal (hereinafter referred to as "transmission frequency") and the frequency of the reception signal (hereinafter referred to as "reception frequency"). It has a beat frequency that is a difference. The beat signal generated by the mixer 22 is converted to a digital signal by the A / D conversion unit 23 and then output to the processing unit 30.

  The processing unit 30 includes a transmission / reception control unit 31, a signal processing unit 32, and a storage unit 33. The signal processing unit 32 includes a detection unit 32a and a filter processing unit 32b.

  The storage unit 33 stores history data 33a. The history data 33 a is information including a history of target data in a series of signal processing performed by the signal processing unit 32.

  The processing unit 30 is a microcomputer including, for example, a central processing unit (CPU), a read only memory (ROM) corresponding to the storage unit 33, a random access memory (RAM), a register, other input / output ports, etc. The entire apparatus 1 is controlled.

  The CPU of the microcomputer reads and executes the program stored in the ROM to function as the transmission / reception control unit 31 and the signal processing unit 32. The transmission / reception control unit 31 and the signal processing unit 32 can be entirely configured by hardware such as an application specific integrated circuit (ASIC) or a field programmable gate array (FPGA).

  The transmission / reception control unit 31 controls the transmission unit 10 including the signal generation unit 11 and the reception unit 20. The signal processing unit 32 periodically executes a series of signal processing. Subsequently, each component of the signal processing unit 32 will be described.

  The detection unit 32a includes a frequency analysis unit 321a, a peak extraction unit 322a, and an instantaneous value generation unit 323a, and corresponds to the target based on the frequency-modulated transmission wave and the reflection wave of the transmission wave by the target. The instantaneous value 100 is detected.

  The frequency analysis unit 321a performs fast Fourier transform (FFT) processing (hereinafter referred to as "FFT processing") on the beat signal input from each A / D conversion unit 23, and outputs the result as a peak. It outputs to the extraction part 322a. The result of the FFT processing is the frequency spectrum of the beat signal, and is the power value (signal level) of each frequency of the beat signal (every frequency bin set at frequency intervals according to the frequency resolution).

  The peak extraction unit 322a extracts a peak frequency that is a peak in the result of the FFT processing by the frequency analysis unit 321a, and outputs the extraction result to the instantaneous value generation unit 323a. The peak extraction unit 322a extracts peak frequencies for each of the "UP section" and the "DN section" of the beat signal described later.

  The instantaneous value generation unit 323a executes angle estimation processing for calculating the arrival angle of the reflected wave corresponding to each of the peak frequencies extracted by the peak extraction unit 322a and the power value thereof. In addition, since an arrival angle is an angle estimated that a target object exists at the time of execution of an angle estimation process, it may be described as an "estimated angle" below.

  In addition, the instantaneous value generation unit 323a executes pairing processing to determine the correct combination of the peak frequencies of the “UP section” and the “DN section” based on the calculation result of the calculated estimated angle and the power value.

  In addition, the instantaneous value generation unit 323a calculates the distance and relative velocity of each target with respect to the radar device 1 from the determined combination result. Further, the instantaneous value generation unit 323a outputs the calculated estimated angle, distance, and relative velocity of each target to the filter processing unit 32b as an instantaneous value 100 for the latest period (latest scan).

  In order to make the description easy to understand, the flow of processing up to here in the signal processing unit 32 from the pre-processing of the signal processing unit 32 is shown in FIGS. 3 to 4C. FIG. 3 is an explanatory view of processing from pre-stage processing of the signal processing unit 32 to peak extraction processing in the signal processing unit 32. As shown in FIG.

  Moreover, FIG. 4A is process explanatory drawing of an angle estimation process. Moreover, FIG. 4B and FIG. 4C are process explanatory drawing (the 1) and (the 2) of a pairing process. Note that FIG. 3 is divided into three regions by two thick downward white arrows. In the following, such regions will be referred to as the upper stage, the middle stage, and the lower stage, respectively.

  As shown in the upper part of FIG. 3, the transmission signal fs (t) is transmitted from the transmission antenna 13 as a transmission wave, and then reflected by the target to arrive as a reflected wave, and the reception signal fr (t (t) As received).

  At this time, as shown in the upper part of FIG. 3, the reception signal fr (t) is delayed by the time difference T with respect to the transmission signal fs (t) according to the distance between the host vehicle MC and the target. The beat signal has a frequency fup of the "UP section" where the frequency rises and a frequency fdn of the "DN section" where the frequency falls due to the time difference T and the Doppler effect based on the relative velocity of the vehicle MC and the target. Is obtained as a repeated signal (see the middle part of FIG. 3).

  The lower part of FIG. 3 schematically shows the result of FFT processing of the beat signal in the frequency analysis unit 321 a for each of the “UP section” side and the “DN section” side.

  As shown in the lower part of FIG. 3, after the FFT processing, waveforms in respective frequency domains on the “UP section” side and the “DN section” side are obtained. The peak extraction unit 322a extracts a peak frequency that is a peak in the waveform.

  For example, in the case shown in the lower part of FIG. 3, the peak extraction threshold value is used, and on the “UP section” side, peaks Pu1 to Pu3 are determined as peaks, and peak frequencies fu1 to fu3 are respectively extracted.

  Further, on the “DN section” side, the peaks Pd1 to Pd3 are respectively determined as peaks by the peak extraction threshold value, and the peak frequencies fd1 to fd3 are respectively extracted.

  Here, in the frequency components of the respective peak frequencies extracted by the peak extraction unit 322a, there are cases where reflected waves from a plurality of targets are mixed. Therefore, the instantaneous value generation unit 323a performs angle estimation processing for performing azimuth calculation on each of the peak frequencies, and analyzes the presence of a target corresponding to each of the peak frequencies.

  Note that the azimuth calculation in the instantaneous value generation unit 323a can be performed using, for example, a known arrival direction estimation method such as ESPRIT (Estimation of Signal Parameters via Rotational Invariance Techniques).

  FIG. 4A schematically shows the azimuth calculation result of the instantaneous value generation unit 323a. The instantaneous value generation unit 323a calculates an estimated angle of each target (each reflection point) corresponding to each of the peaks Pu1 to Pu3 from each peak Pu1 to Pu3 of the direction calculation result. Further, the magnitudes of the respective peaks Pu1 to Pu3 are power values. As shown in FIG. 4B, the instantaneous value generation unit 323a performs such angle estimation processing for each of the “UP section” side and the “DN section” side.

  Then, in the direction calculation result, the instantaneous value generation unit 323a performs a pairing process of combining the peaks near the estimated angle and the power value. Further, from the combination result, the instantaneous value generation unit 323a calculates the distance and the relative velocity of each target (each reflection point) corresponding to the combination of each peak.

  The distance can be calculated based on the relationship of “distance ∝ (fup + fdn)”. The relative velocity can be calculated on the basis of the relationship of "velocity (fup-fdn)". As a result, as shown in FIG. 4C, it is possible to obtain a pairing processing result indicating the estimated angle, distance, and relative velocity of the reflection point RP with respect to the radar device 1.

  Next, the filter processing unit 32b will be described with reference to FIG. FIG. 5 is a block diagram showing the configuration of the filter processing unit 32b. As shown in FIG. 5, the filter processing unit 32b includes a prediction unit 321b, an assignment unit 322b, a new particle setting unit 323b, a weighting unit 324b, a resampling unit 325b, and a target data generation unit 326b. .

  The filter processing unit 32 b generates target data corresponding to the instantaneous value 100 by applying a particle filter that associates a predetermined number of particle data 50 with the instantaneous value 100 detected by the detection unit 32 a.

The prediction unit 321 b performs prediction processing to predict the current particle data 50 from the previous particle data 50 in the particle filter. Specifically, the prediction unit 321b includes the latest period time t, when the distribution X _t of each particle data 50 at time t, the distribution X _t-1 time t-1 of the previous cycle The N particle data 50 are arranged (sampled) based on the probability density function based on it.

  Note that, in the case of the second processing for a new target, the prediction unit 321b places the perpendicular line VL (see FIG. 1B) on the basis of the relative velocity given in the first processing by the new particle setting unit 323b described later. Perform processing to move. In addition, the prediction process of the particle data 50 with respect to a novel target is later mentioned using FIGS. 6-9.

  The allocation unit 322 b performs processing of allocating the instantaneous value 100 in the latest cycle to the current particle data 50 that is the prediction result of the prediction unit 321 b. Specifically, the allocation unit 322 b predicts the present target data from the previous target data, and assigns an instantaneous value 100 existing within a predetermined allocation range based on the prediction result.

  In addition, when there is an instantaneous value 100 which does not exist in the allocation range of any target data, the allocation unit 322 b treats the instantaneous value 100 as a new target.

  The new particle setting unit 323b sets particle data 50 in the initial state around the instantaneous value corresponding to the new target. The initial state means, for example, a state in which a plurality of particle data 50 are arranged in a predetermined positional relationship with respect to a new target.

  Furthermore, the new particle setting unit 323b determines, based on the relative velocity of the instantaneous value, the information of the relative vector (state vector SV) indicating the predicted position predicted to move in the next process with respect to the particle data 50 in the initial state. give. The processing by the weighting unit 324 b and the resampling unit 325 b is not performed on the particle data 50 in the initial state set by the new particle setting unit 323 b.

  The weighting unit 324 b weights each of the current particle data 50 based on the assigned current instantaneous value 100. Specifically, the weighting unit 324 b increases the weight of particles close to the current instantaneous value 100 in the current particle data 50 and decreases the weight of particles distant from the current instantaneous value 100. Here, “close” and “far” mean that the Mahalanobis distance is “close” and “far”.

  The resampling unit 325 b rearranges (resamples) the particle data 50 based on the weight of each particle data 50 of this time. Specifically, the resampling unit 325 b moves the particle data 50 with small weight to near the instantaneous value 100.

  The target data generation unit 326 b generates target data based on the average of the current particle data 50 rearranged by the resampling unit 325 b. In addition, although target object data generation part 326b generated target object data based on an average of a probability density function, it may generate target data based on a median of each parameter of particles, for example. In addition, target data may be generated based on the maximum value of the probability density function.

  In addition, the target data generation unit 326b outputs the instantaneous value 100 corresponding to the new target as it is as target data. That is. The target data generation unit 326 b outputs the instantaneous value 100 as target data.

  Next, the prediction process of the particle data 50 for the new target of the prediction unit 321b will be described in detail with reference to FIGS. 6 to 9 are diagrams showing the processing content of the prediction unit 321b. First, the method of calculating the state vector SV will be described with reference to FIG. Although one state vector SV is illustrated in FIG. 6, in actuality, a plurality of state vectors SV having different vector orientations are calculated.

  As shown in FIG. 6, the new particle setting unit 323b determines a longitudinal component RVy that is a relative velocity of the traveling direction (y axis) of the host vehicle MC based on the vector RV indicating the relative velocity, and a vehicle width direction of the host vehicle MC. A state vector SV, which is a combination with the transverse component RVx, which is a relative velocity to, is generated, or information of the state vector SV is given to particle data 50 in the initial state.

  Specifically, first, the new particle setting unit 323b generates a perpendicular line VL passing through the end point of the vector RV. The perpendicular line VL is a straight line represented by the longitudinal component RVy and the transverse component RVx. That is, the linear expression of the perpendicular line VL is represented by RVy = a × RVx + b. The coefficient a is a = −1 / tan θ, and the coefficient b is expressed as b = RV / sin θ using a vector RV which is a relative velocity.

  Subsequently, the new particle setting unit 323b extracts a combination of the longitudinal component RVy and the horizontal component RVx using the linear expression of the perpendicular line VL described above, and gives the state data SV of the combination to the particle data 50. As a result, even if the new target moves in any direction, the particle data 50 can be easily allocated in the subsequent processing, so that the prediction error can be reduced.

  When the angle θ of the instantaneous value 100 is −270 °, −90 °, 0 °, 180 °, 360 °, the prediction unit 321 b performs exception processing without using the linear equation of the perpendicular line VL.

  Specifically, when the angle θ is 0 ° or 360 °, RVx = RV, 180 °, RVx = −RV, and −270 ° or 90 °, RVy = RV, −90 ° Or in the case of 270 °, RVy = −RV.

  Next, with reference to FIG. 7, a method of generating particle data 50 by limiting the longitudinal component RVy and the transverse component RVx will be described. FIG. 7 shows the case where thresholds are provided in the positive and negative directions of the longitudinal component RVy and in the positive and negative directions of the lateral component RVx.

  The new particle setting unit 323b generates the state vector SV based on a combination in which the relative velocity of the longitudinal component RVy and the relative velocity of the transverse component RVx are each equal to or less than a predetermined threshold. For example, the threshold value of the longitudinal component RVy is 120 km / h, and the threshold value of the transverse component RVx is 60 km / h.

  In this case, the upper limit (“RVy” in FIG. 7) of the longitudinal component RVy is 120 km / h, and the lower limit (“−RVy”) is −240 km / h. Further, the upper limit (“RVx”) of the lateral component RVx is 60 km / h, and the lower limit (“−RVx”) is −60 km / h.

  Then, the new particle setting unit 323b sets the particle data 50a and the particle data 50b, which are intersection points of the threshold value (the rectangular frame shown in FIG. 7) and the perpendicular line VL, to both ends, A state vector SV is generated so that particle data 50 is arranged.

  In other words, the particle data 50a is the upper limit of the longitudinal component RVy and the lower limit of the transverse component RVx, and the particle data 50b is the lower limit of the longitudinal component RVy and the upper limit of the transverse component RVx.

  As described above, by providing the threshold value for the vertical component RVy and the horizontal component RVx, it is possible to prevent the range of the particle data 50 from being extended excessively, that is, the particle data 50 can be densely arranged, thereby improving the prediction accuracy. it can.

  In the example shown in FIG. 7, the threshold is set to both the vertical component RVy and the horizontal component RVx, but the threshold may be set to only one of them.

  Further, the new particle setting unit 323b is not limited to the relative velocity threshold value, and may generate the state vector SV based on the threshold value of another parameter. This point will be described with reference to FIG. In FIG. 8, the threshold value Vth for the velocity (vector V) of the target with respect to the road surface will be described as an example.

  The new particle setting unit 323b calculates the velocity of the target relative to the road surface based on the relative velocity of the instantaneous value, generates the state vector SV based on the calculated velocity, and gives it to the particle data 50 in the initial state. The velocity of the target relative to the road surface is a velocity component obtained by removing the velocity component of the host vehicle MC from the relative velocity of the instantaneous value 100, and is a velocity of the substantial target. When the velocity of the target relative to the road surface is V, the relative velocity that is the length of the vector RV is RV, and the velocity of the host vehicle MC is MV, the following equation is obtained. That is, V = RV + MV × cos θ.

  Then, the new particle setting unit 323b generates a perpendicular line VLa corresponding to the vector V. The perpendicular line VLa is generated in substantially the same manner as the perpendicular line VL described above. That is, the perpendicular line VLa is a line connecting the end points of the state vector (not shown) corresponding to the vector V.

  Then, the new particle setting unit 323b sets the threshold value Vth of the velocity V regardless of the direction. That is, the threshold value Vth is a circle centered on the instantaneous value 100. Then, the new particle setting unit 323b generates the state vector SV so that the current particle data 50 is arranged between the particle data 50a and the particle data 50b which are intersections of the threshold value Vth and the perpendicular line VLa. Therefore, the prediction error can be reduced by predicting the particle data 50 in consideration of the velocity of the target.

  Although the case where the particle data 50 is arranged on the perpendicular line VL has been described above, the particle data 50 may be arranged around the perpendicular line VL. This point will be described with reference to FIG.

  As shown in FIG. 9, the prediction unit 321b arranges the particle data 50 (not shown) around the perpendicular line VL in accordance with the probability density function P centering on an arbitrary point on the perpendicular line VL. Specifically, the probability density function P follows a normal distribution centered on any point on the perpendicular line VL.

  That is, the new particle setting unit 323b generates the state vector SV such that the current particle data 50 is arranged based on the probability density function P centering on the point indicating the combination of the vertical component RVy and the horizontal component RVx. Thereby, even when the relative velocity of the new target changes, it is possible to make the prediction error less likely to occur.

  Next, a processing procedure of processing performed by the radar device 1 according to the embodiment will be described with reference to FIG. FIG. 10 is a flowchart showing a processing procedure of processing performed by the radar device 1 according to the embodiment.

  As shown in FIG. 10, first, the detection unit 32a detects an instantaneous value 100 corresponding to a target based on the frequency-modulated transmission wave and the reflected wave of the transmission wave by the target (step S101).

  Subsequently, the prediction unit 321b of the filter processing unit 32b performs prediction processing to predict the current particle data 50 based on the previous particle data 50 (step S102).

  Subsequently, the allocating unit 322 b allocates the current instantaneous value 100 to the current particle data 50 (step S 103). Subsequently, the allocating unit 322b determines whether there is a new target whose instantaneous value 100 does not exist within the predetermined allocation range (step S104).

  When there is a new target (Yes at step S104), the new particle setting unit 323b sets particle data 50 in the initial state for the new target, and also obtains information on the state vector SV based on the relative velocity. Give (step S105).

  On the other hand, in step S104, if the weighting unit 324b is not a new target (step S104, No), that is, based on the current instantaneous value 100 within the predetermined allocation range, weights the current particle data 50 respectively. The operation is performed (step S106).

  Subsequently, the resampling unit 325b resamples the current particle data 50 based on the weighting by the weighting unit 324b (step S107). Subsequently, the target data generation unit 326b updates the probability density function of the resampled current particle data 50, generates target data based on the probability density function (step S108), and ends the processing. .

  As described above, the radar device 1 according to the embodiment includes the detection unit 32a and the filter processing unit 32b. The detection unit 32a detects an instantaneous value 100 corresponding to a target based on the frequency-modulated transmission wave and the reflected wave of the transmission wave by the target. The filter processing unit 32 b generates target data corresponding to the instantaneous value 100 by applying a particle filter that associates a predetermined number of particle data 50 with the instantaneous value 100 detected by the detection unit 32 a. In addition, the filter processing unit 32b generates a vector of the direction from the target to the vehicle based on the relative velocity of the previous instantaneous value 100 in a new target to which the previous particle data 50 is not allocated to the previous instantaneous value 100. The position corresponding to the perpendicular to the point is taken as the predicted position of the current particle data 50. Thereby, the prediction accuracy of particle data 50 to a new target can be improved.

  Although the radar apparatus 1 is provided in a vehicle in the above-described embodiment, it is needless to say that the radar apparatus 1 may be provided in a mobile other than the vehicle, for example, a ship or an aircraft.

  Further, in the above-described embodiment, the ESPRIT is described as an example of the direction of arrival estimation method used by the radar device 1, but the present invention is not limited to this. For example, digital beam forming (DBF), propagator method based on an improved spatial-smoothing matrix (PRISM), multiple signal classification (MUSIC) or the like may be used.

  In the above-described embodiment, when the instantaneous value 100 is at the same height as the mounting position of the radar device 1, that is, the target detection method on the XY plane is described, the height of the instantaneous value 100 is taken into consideration. May be This point will be described with reference to FIG.

  FIG. 11 is a diagram showing the processing content of the new particle setting unit 323b according to the modification. In FIG. 11, the z-axis direction is described as the height direction. As shown in FIG. 11, when the instantaneous value 100 is detected from a height different from the mounting position of the radar device 1, the vector RV indicating the relative velocity is a longitudinal component RVy, a transverse component RVx, and a height component RVz. And will be included.

  The prediction unit 321 b generates a state vector SV such that the particle data 50 (not shown) is disposed on a vertical plane VS having the vector RV as a normal vector and the end RVc of the vector RV as a center. Specifically, the prediction unit 321 b generates the particle data 50 by extracting the state vector SV indicating the combination of the vertical component RVy corresponding to the vertical plane VS, the horizontal component RVx, and the height component RVz.

  Thereby, even when the height of the instantaneous value 100 is different from the height of the radar device 1, the prediction accuracy of the particle data 50 can be improved.

  Further effects and modifications can be easily derived by those skilled in the art. Thus, the broader aspects of the invention are not limited to the specific details and representative embodiments represented and described above. Accordingly, various modifications may be made without departing from the spirit or scope of the general inventive concept as defined by the appended claims and their equivalents.

Reference Signs List 1 radar apparatus 32 signal processing unit 32a detection unit 32b filter processing unit 321a frequency analysis unit 322a peak extraction unit 323a instantaneous value generation unit 321b prediction unit 322b allocation unit 323b new particle setting unit 324b weighting unit 325b resampling unit 326b target data generation Part 33 Storage part 33a Historical data 50 Particle data 100 Instantaneous value

Claims (6)

A detection unit that detects an instantaneous value corresponding to the target based on the frequency-modulated transmission wave and a reflected wave of the transmission wave from the target;
A filter processing unit that generates target data corresponding to the instantaneous value by applying a particle filter that assigns a predetermined number of particle data to the instantaneous value detected by the detection unit;
The filter processing unit
In a new target to which the previous particle data can not be assigned to the previous instantaneous value, a position corresponding to a perpendicular to the vector of the direction from the target to the vehicle based on the relative velocity of the previous instantaneous value A radar apparatus characterized in that is a predicted position of the particle data at this time. The filter processing unit
The particle data in the initial state is set with respect to the instantaneous value of the new target, and based on the relative velocity, a longitudinal component which is a relative velocity of the traveling direction of the vehicle and a vehicle width direction of the vehicle The radar apparatus according to claim 1, wherein a state vector, which is a combination with a transverse component that is a relative velocity of V, is given to particle data in the initial state. The filter processing unit
The radar apparatus according to claim 2, wherein the state vector of the combination in which the relative velocity of the longitudinal component and the relative velocity of the lateral component are respectively equal to or less than predetermined threshold values is provided to particle data in the initial state. The filter processing unit
The speed of the target relative to the road surface is calculated based on the relative speed, the state vector is generated based on the calculated speed, and is provided to particle data in the initial state. Radar apparatus as described. The filter processing unit
The state vector is generated such that the current particle data is arranged based on a probability density function centered on a point indicating a combination of the longitudinal component and the lateral component. The radar apparatus according to any one. Detecting an instantaneous value corresponding to the target based on the frequency-modulated transmission wave and a reflected wave of the transmission wave by the target;
Applying a particle filter that assigns a predetermined number of particle data to the instantaneous value detected in the detecting step to generate target data corresponding to the instantaneous value;
The filtering step is
In a new target to which the previous particle data can not be assigned to the previous instantaneous value, a position corresponding to a perpendicular to the vector of the direction from the target to the vehicle based on the relative velocity of the previous instantaneous value A target detection method characterized in that is the predicted position of the particle data at this time.

Images