Radar device, target detection method, and target detection program

Abstract

To provide a radar device, a target detection method and a target detection program with which it is possible to improve responsiveness while acquiring detail information.SOLUTION: A radar device 1 transmits and receives a signal and processes a received signal so as to detect a target. The radar device 1 includes: a distance process of executing a processing cycle multiple times, each processing cycle acquiring a per-distance bin signal; a first speed process of acquiring a per-distance bin first per-speed bin signal; a store process of storing at least some of the per-distance bin first per-speed bin signals as store information on the basis of a prescribed condition; a second speed process of acquiring a per-distance bin second per-speed bin signal; an angle process of acquiring a per-distance bin first per-speed bin angle signal and a per-distance bin second per-speed bin angle signal; and a threshold process of assessing each signal strength based on a prescribed threshold either before or after the angle process.SELECTED DRAWING: Figure 2

Description

[Technical field]

[0001] The present invention relates to a radar device, a target detection method, and a target detection program. [Background technology]

[0002] Generally, a radar device for the purpose of perimeter monitoring, ADAS (advanced driver assistance system), etc., may be mounted on a moving object such as a vehicle. In such a radar device, the longer the measurement time, the easier it is to obtain high-resolution information, but the calculation load is likely to increase. In this regard, a device (see, for example, Patent Document 1) has been proposed that includes a first signal processing unit that calculates a first velocity by observation in a first observation time, and a second signal processing unit that calculates a second velocity by observation in a second observation time longer than the first observation time. In the radar device described in Patent Document 1, the first velocity is calculated from a beat signal observed in a relatively short observation time at a relatively short interval between observation points, and the second velocity is calculated from a beat signal observed in a relatively long observation time at a relatively long interval between observation points, thereby suppressing an increase in the calculation load and realizing a relatively high velocity resolution. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] JP 2019-15625 A Summary of the Invention [Problem to be solved by the invention]

[0004] However, in the radar device described in Patent Document 1, the result of the first signal processing is utilized as a speed aliasing countermeasure, and the result of the second signal processing is used as the true speed to determine the speed, so that the second signal processing, which has a long observation time, must be completed, and there is an inconvenience that the latency (delay time) is likely to be large. In other words, it is difficult to suppress the latency while obtaining high-definition information. In addition, the second signal processing is premised on performing speed processing after thinning out the time series data, and there is a problem that the SN is inferior to the case where all the measured signals are used for speed processing because there are parts that are not used for speed processing even though they are measured during the first observation time. In addition, since the second signal processing is premised on sampling at equal intervals, in this method, the intervals between each transmission signal, that is, each chirp period in the FCM method, must also be equal across processing cycles. On the other hand, as another requirement for radar, there are cases where a period during which transmission or reception is suspended is required for a certain period within a processing cycle. This method is difficult to apply when the idle period is sufficiently longer than the thinning interval, or when the processing cycle is not a multiple of each chirp period.

[0005] The present invention has been made in consideration of the above-mentioned problems, and has an object to provide a radar device, a target detection method, and a target detection program that can suppress latency while obtaining high-resolution information. [Means for solving the problem]

[0006] In order to achieve the above-mentioned object, a radar device according to the present invention is a radar device that detects targets by transmitting and receiving signals and processing received signals, and is characterized in that it executes a processing cycle multiple times, and each of the processing cycles includes a distance process for acquiring a range bin signal, a first speed process for acquiring a range bin first speed bin signal based on the multiple range bin signals repeatedly acquired within the processing cycle, a storage process for storing at least a portion of the range bin first speed bin signal as storage information based on a predetermined condition, a second speed process for acquiring a range bin second speed bin signal based on the storage information stored in multiple processing cycles prior to the processing cycle, an angle process for acquiring a range bin first speed bin angle signal and a range bin second speed bin angle signal based on the range bin first speed bin signal and the range bin second speed bin signal acquired in multiple receiving channels, respectively, and a threshold process for determining each signal strength based on a predetermined threshold at least either before or after the angle process.

[0007] In a radar device according to one embodiment of the present invention, in the second speed processing, the second speed processing is performed on a number of bins that is less than the number of bins of all of the first speed bin signals for each distance bin acquired in the first speed processing.

[0008] In a radar device according to one embodiment of the present invention, in the storage process, the signal for each distance bin and each first velocity bin to be stored is selected based on at least one of predetermined bin components, information about the moving body on which the radar device is mounted, and the processing result of a processing cycle prior to the processing cycle in question.

[0009] In the radar device according to one aspect of the present invention, the storing process selects and stores the first velocity bin signal for each distance bin in a bin component in which the relative velocity corresponds to 0.

[0010] In a radar device according to an aspect of the present invention, the storing process selects and stores the first speed bin signal for each distance bin in a bin component that includes the speed of a moving object on which the radar device is mounted.

[0011] In a radar device according to one embodiment of the present invention, the processing cycle includes a mode determination process for determining a driving mode by acquiring information about the moving body before the storage process, and in the storage process, a signal for each distance bin and each first speed bin to be stored is selected based on the driving mode.

[0012] In a radar device according to one embodiment of the present invention, in the storage process, the signal to be stored for each distance bin and each first velocity bin is selected based on information obtained by performing the angle processing and the threshold processing in the processing cycle prior to the current processing cycle.

[0013] In a radar device according to one embodiment of the present invention, in the threshold processing, different thresholds are set for the first speed bin signal for each range bin and the second speed bin signal for each range bin, or the first speed bin angle signal for each range bin and the second speed bin angle signal for each range bin, or a common threshold is set after multiplying the signals by different coefficients.

[0014] In a radar device according to one embodiment of the present invention, the processing cycle includes, after the angle processing, a correction process for correcting distance information of the angle signal for each distance bin and each second speed bin based on the signal for each distance bin and each first speed bin acquired in a processing cycle prior to the processing cycle in question.

[0015] In a radar device according to one embodiment of the present invention, in the processing cycle, after the threshold processing, a signal derived from the first angle signal for each distance and speed and a signal derived from the angle signal for each distance bin and each second speed bin are mixed, and at least one of a ground fixed object determination process, a clustering process, a tracking process, and a targeting process is performed.

[0016] The target detection method of the present invention is a target detection method for detecting a target by transmitting and receiving signals and processing received signals, and is characterized in that a processing cycle is executed multiple times, and each of the processing cycles includes a distance processing for acquiring a signal for each distance bin, a first speed processing for acquiring a signal for each distance bin based on the multiple distance bin signals repeatedly acquired within the processing cycle, a storage processing for storing at least a portion of the signal for each distance bin and the first speed bin as stored information based on a predetermined condition, a second speed processing for acquiring a signal for each distance bin and the second speed bin based on the stored information stored in multiple processing cycles prior to the processing cycle, an angle processing for acquiring an angle signal for each distance bin and an angle signal for each distance bin based on the first speed bin signal for each distance bin and the second speed bin signal for each distance bin acquired in multiple receiving channels, respectively, and a threshold processing for judging each signal strength based on a predetermined threshold at least either before or after the angle processing.

[0017] The target detection program of the present invention is a target detection program for detecting targets by transmitting and receiving signals and processing received signals, and causes a computer to execute a processing cycle multiple times, each of which includes: distance processing for acquiring a range bin signal; first speed processing for acquiring a range bin first speed bin signal based on the multiple range bin signals repeatedly acquired within the processing cycle; storage processing for storing at least a portion of the range bin first speed bin signal as stored information based on a predetermined condition; second speed processing for acquiring a range bin second speed bin signal based on the stored information stored in multiple processing cycles prior to the current processing cycle; angle processing for acquiring a range bin first speed bin angle signal and a range bin second speed bin angle signal based on the range bin first speed bin signal and the range bin second speed bin signal acquired in multiple receiving channels, respectively; and threshold processing for judging each signal strength based on a predetermined threshold at least either before or after the angle processing. Effect of the Invention

[0018] According to the radar device of the present invention, it is possible to suppress latency while obtaining high-definition information. [Brief description of the drawings]

[0019] [Figure 1] 1 is a block diagram showing an example of a radar device according to an embodiment of the present invention; [Diagram 2] 4 is a flowchart illustrating an example of a process performed by a radar device according to an embodiment of the present invention. [Diagram 3] FIG. 4 is a schematic diagram showing measurement times of a radar device according to an embodiment of the present invention and a radar device of a comparative example. [Figure 4] 10 is a graph showing an example of a signal acquired by a radar device of a comparative example. [Diagram 5] 5 is a graph showing an example of a signal acquired in only one processing cycle in the radar device according to the embodiment of the present invention. [Figure 6] 5 is a graph showing an example of a signal acquired in each processing cycle in the radar device according to the embodiment of the present invention. [Figure 7] 5 is a graph showing an example of a signal acquired through a plurality of processing cycles in the radar device according to the embodiment of the present invention. [Figure 8] FIG. 2 is a schematic diagram showing an example of information acquired by a radar device according to an embodiment of the present invention. [Figure 9] 5 is a graph showing an example of information acquired by a radar device according to an embodiment of the present invention. [Figure 10] 10 is a flowchart showing another example of the process performed by the radar device according to the embodiment of the present invention. [Figure 11] 5 is a flowchart showing the above-mentioned process performed by the radar device according to the embodiment of the present invention. [Figure 12] FIG. 2 is a schematic diagram showing an example of information acquired by a radar device according to an embodiment of the present invention. [Figure 13] 10 is a flowchart showing another example of the process performed by the radar device according to the modified example of the present invention. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0020] Hereinafter, an embodiment of the present invention will be described with reference to the drawings. A radar device according to an embodiment of the present invention is a radar device that detects a target by transmitting and receiving signals and processing received signals, and executes a processing cycle multiple times, and each processing cycle includes a distance process for acquiring a range bin signal, a first speed process for acquiring a range bin first speed bin signal based on a plurality of range bin signals repeatedly acquired in the processing cycle, a storage process for storing at least a part of the range bin first speed bin signal as stored information based on a predetermined condition, a second speed process for acquiring a range bin second speed bin signal based on the stored information stored in a plurality of processing cycles prior to the processing cycle, an angle process for acquiring a range bin first speed bin angle signal and a range bin second speed bin angle signal based on the range bin first speed bin signal and the range bin second speed bin signal acquired in a plurality of receiving channels, and a threshold process for determining each signal strength based on a predetermined threshold at least either before or after the angle process. Hereinafter, the radar device will be specifically described.

[0021] An example of a radar device 1 is shown in Fig. 1. The radar device 1 has a transmitter 2 and a receiver 3. The transmitter 2 has a transmission antenna 11, a signal generator 14 that generates a signal, a transmission controller 13 that controls the signal generator 13, and an oscillator 15 that outputs the signal generated by the signal generator 14. The receiver 3 has a reception antenna 12, a demodulator 16 that demodulates the received signal, an A / D converter (ADC) 17 that digitally converts the received signal, and a signal processor 18 that processes the converted signal. The signal format in the radar device 1 may be an FCM format or a pulse format.

[0022] That is, the radar device 1 transmits a signal R1 from the transmitting antenna 11, the signal R1 is reflected by the target T1, and the reflected signal R1 is received by the receiving antenna 12. The signal R1 received in this manner is appropriately demodulated, converted, and the like in the receiving unit 3, and then processed by the signal processing unit 18. The signal processing unit 18 has a distance processing unit 181, a first velocity processing unit 182, a storage unit 183, a second velocity processing unit 183, an angle processing unit 184, and a threshold processing unit 186. In this case, each unit of the signal processing unit 18 may actually be one device, or may be composed of multiple devices.

[0023] The radar device 1 is mounted on a moving body such as an automobile, and is used to detect the relative position of the target and ground fixed objects in the vicinity by processing the signal reflected from the unknown object. The target detected by the radar device 1 of this embodiment refers to dynamic objects such as surrounding automobiles and pedestrians, and the ground fixed objects detected by the radar device 1 refer to static objects such as guard rails, walls, stopped vehicles, and other structures, and both are detected. The radar device 1 is mounted on each corner of the moving body (front right, rear right, front left, rear left) and the like and used for periphery monitoring. Examples of applications for periphery monitoring include blind spot detection on the rear side, collision detection on the front side, free space application for grasping the roadside shape on the front side and recognizing the drivable area of ​​the detected moving body, and avoidance of collisions when parking. In addition to the above, it may be collision prevention at intersections and separation of ground fixed objects and targets. That is, in each scene, the radar device 1 detects targets and fixed objects on the ground, and the surroundings monitoring application issues a warning to the passengers and controls the brakes, steering, and the like.

[0024] In order to make the above application viable, it is necessary to increase the resolution of the point cloud (sometimes referred to as response in this embodiment), which is raw data that is a candidate for a target or a fixed object on the ground. In order to detect a point cloud corresponding to an object or a contour that spreads in the angular direction, a high angular resolution is required. On the other hand, in a radar device, it is possible to grasp the Doppler speed of an object by signal processing, and it is possible to separate the reflection points of a spread object into different speed components without directly having a high angular resolution. Compared to angle information mixed as the same speed component, it is possible to achieve high precision by separating each speed component and making each independent angle information. Therefore, speed separation performance is important in making the application viable.

[0025] For example, when it is necessary to detect a fixed object on the ground, such as in a free space application, the magnitude of the relative speed of the fixed object on the ground depends on the angle of view of the object with respect to the traveling direction. The relationship is expressed by a Cos function, so it is not linear, and there are areas where the change in relative speed, i.e., the speed resolution is high and areas where it is low. In particular, the speed resolution is high in the direction perpendicular to the traveling direction, but the speed resolution is low in the direction close to the traveling direction. Also, for example, in a blind spot detection application or a parking assistance application, the relative speed between the vehicle and the object varies depending on the scene. It is difficult to separate the reflection points of each of the spread objects by speed in a scene where the relative speed of the object is close to 0, compared to when the relative speed of the object is relatively high. According to the radar device 1 of this embodiment, it is possible to deal with these situations, as will be described below.

[0026] The radar device 1 may be mounted on construction machinery, transportation machinery, agricultural machinery, etc., in addition to ordinary passenger vehicles. In the following description, the radar device 1 detects targets and the like in a horizontal plane, but it may also detect targets and the shape of the ground surface in an elevation plane. The radar device 1 may also be mounted on a fixed structure instead of a moving object.

[0027] The signal processing unit 18 of the radar device 1 performs information processing (target detection method) as described below. First, a basic method will be described, and in the description of other methods, the description of parts common to the basic method will be omitted. In addition, a program for causing a computer to execute such a target detection method may be stored in a storage medium.

[0028] [Basic method] The signal processing unit 18 executes a processing cycle multiple times, as shown in FIG. 2. As shown in the lower part of FIG. 3, the measurement time in one processing cycle is, for example, 25.6 msec, and the time from the start of one processing cycle to the start of the next processing cycle is set to, for example, 50 msec. Each processing cycle has a distance processing (step S1), a first speed processing (step S2, simply "speed processing" in the figure), a store processing (step S3, simply "store" in the figure), a second speed processing (step S4, simply "speed processing" in the figure), an angle processing (step S5), and a threshold processing (step S6). Note that, although S1 to S6 are shown in the N-3 cycle in FIG. 2, the same is true for the other cycles. Also, in other drawings, when the steps of each processing cycle are the same, only one processing cycle is given a reference symbol.

[0029] When a certain processing cycle (e.g., N-3) is completed, the next processing cycle (e.g., N-2) is executed. During these processing cycles, appropriate steps may be performed, such as clustering or tracking to detect targets and fixed objects on the ground based on the point cloud data obtained by the above processing cycles.

[0030] In distance processing, a signal for each distance bin is acquired. For example, in the FCM method, a part of the transmitted signal is mixed with the received signal, and a beat signal is generated based on the difference between them. This beat signal is then subjected to FFT processing to acquire a spectrum corresponding to the distance to the target across distance bins (signal for each distance bin). Based on the bin information of the peak, it is possible to acquire information on the distance to the target. Note that the pulse method differs from the FCM method in that a transmitted signal modulated into a pulse is continuously received and a signal for each distance bin is acquired without performing FFT processing, but the subsequent processing is similar.

[0031] In the first velocity processing, a first velocity bin signal for each distance bin is acquired based on a plurality of distance bin signals repeatedly acquired within a processing cycle. Here, by performing FFT processing on the many repeatedly acquired signals, velocity components are acquired from the time changes and phase changes of the signals, and the results are stored in the velocity bins. In addition, by performing this processing for each distance bin, a velocity bin signal for each distance bin is acquired. At this stage, the first velocity bin signal for each distance bin is stored in a two-dimensional distance bin velocity bin, and so-called range-Doppler map information as shown in FIG. 8 is acquired.

[0032] In the storage process, at least a part of the signals for each distance bin and each first speed bin is stored as storage information based on a predetermined condition. The predetermined condition will be described in another method. The information to be stored may be selected in the storage process, or all the signals for each distance bin and each first speed bin may be stored in the storage process, and the information to be used may be selected in the second speed process. In FIG. 8, the shaded squares correspond to the bins to be stored, and the bin components for which high resolution is required are hereinafter referred to as "bins of interest."

[0033] In the second speed processing, a second speed bin signal for each distance bin is acquired based on the stored information stored in a plurality of processing cycles prior to the processing cycle. For example, in the second speed processing in the Nth cycle, a second speed bin signal for each distance bin may be acquired based on the stored information stored by the storage processing in the N-3th to N-1th cycles. In addition, in the second speed processing, a distance and speed signal may be acquired by FFT processing, similar to the first speed processing. In the present embodiment, in the second speed processing, a number of first speed bin signals for each distance bin that is smaller than all the first speed bin signals for each distance bin acquired in the first speed processing is processed.

[0034] In angle processing, a first speed bin signal for each distance bin and a second speed bin signal for each distance bin are processed in a plurality of receiving channels (antenna channels) based on each of the first speed bin signal for each distance bin and the second speed bin signal for each distance bin, which include distance bin information, speed bin information, and angle information. That is, an angle peak (see FIG. 9) can be obtained by performing angle processing (FFT processing) using signals with the same distance bin and speed bin in each antenna channel.

[0035] The threshold processing is performed after the angle processing, and a judgment is made based on the threshold. Here, different thresholds are set for the angle signal for each first speed bin per range bin and the angle signal for each second speed bin per range bin, and it is judged whether the strength of each signal is equal to or greater than the threshold. The angle signal for each first speed bin per range bin and the angle signal for each second speed bin per range bin may be multiplied by different coefficients, and then it may be judged whether each signal is equal to or greater than a common threshold. The signal strength (amplitude) of the signal for each first speed bin per range bin and the signal for each second speed bin per range bin differs even if the signals are obtained from the same target due to the difference in the number of cycles stored. Therefore, by setting different thresholds or multiplying by different coefficients as described above, it is possible to set an appropriate weight when merging the angle signal for each first speed bin per range bin and the angle signal for each second speed bin per range bin.

[0036] The threshold processing may be performed only before the angle processing, or may be performed both before and after the angle processing. When the threshold processing is performed before the angle processing, the target signals are the signals for each distance bin and each first speed bin, and the signals for each distance bin and each second speed bin. In this case, similar to the above, different thresholds may be set in the threshold value, and it may be determined whether the intensity of each signal is equal to or greater than the threshold, or it may be determined whether each signal is equal to or greater than a common threshold after multiplying the signals by different coefficients. In addition, the method of threshold determination is not particularly limited, and various CFARs, fixed values, values ​​at other reference points, values ​​considered in time series, etc. may be used.

[0037] By acquiring the signal angle signal for each distance bin and each first speed bin as described above, the distance and angle as seen from the radar device 1 can be grasped and can be detected as a point cloud in a two-dimensional plane. In addition, for the sake of processing, this point cloud also has speed bin information as an attribute. The signal for each distance bin and each first speed bin is premised on being acquired within a processing cycle, which shortens the time from measurement to output of the processing result and ensures responsiveness.

[0038] The second speed bin signal angle signal for each distance bin can be used to detect points in a two-dimensional plane from distance and angle information, similarly to the first speed bin signal angle signal for each distance bin, but the second speed processing result, which has higher resolution than the first speed processing, allows for higher resolution of the angles of the point cloud. In this regard, the difference from normal speed processing and the higher resolution of angles due to speed separation will be described with reference to Figs. 3 to 7 and 9. First, Fig. 3 shows the horizontal axis as time, with the lower part showing a state in which a processing cycle is repeated four times as an example of this embodiment, and the upper part showing a state in which one processing cycle is performed with a longer measurement time as a comparative example, and the total times of these are approximately equal to each other.

[0039] For example, consider a case where a mobile body equipped with the radar device 1 receives a reflected signal from a stopped vehicle as a ground fixed object spreading in an angular direction in a direction close to the direction of travel while traveling. In this case, the signal R1 transmitted by the radar device 1 is a composite signal reflected at the left end, center, and right end of the stopped vehicle, and the relative speeds of the stopped vehicle are 45.17 km / h, 45.28 km / h, and 45.39 km / h, respectively. When the measurement time is relatively long as in the comparative example, the speeds can be separated by performing FFT processing in the speed processing as shown in FIG. 4. Specifically, the dashed line, the one-dot chain line, and the two-dot chain line show the results of the speed FFT when the signals of each reflection point exist independently, and each signal has a peak at a different speed bin. The solid line shows the results in the case of a composite signal of each reflection point, but the signals can be separated into three speed bins in the same way, and the left end, center, and right end of the stopped vehicle can be stored in independent speed bins.

[0040] On the other hand, if the velocity processing is performed in only one processing cycle, which has a relatively short measurement time, the velocity resolution is low, and as shown in Figure 5, peaks corresponding to each part of the reflection point overlap in the same velocity bin, making signal separation difficult. Therefore, in the second speed processing, FFT processing is performed based on information acquired in four processing cycles by using the velocity information stored after the past first speed processing. In Figure 6, the stored information after the first speed processing obtained by each processing cycle is shown from the top, and the change in the velocity signal itself can be confirmed over the course of four processing cycles. Note that the signal real part (I channel) is shown by a solid line, and the signal imaginary part (Q channel) is shown by a dashed line. Here, an example of complex signal processing by IQ reception (2 channels) of the received signal is shown, and it is possible to distinguish between ± components when performing FFT. Note that the reception method is not limited to complex reception, and real signal (1 channel) processing may also be used, and if real signal processing is used, the number of channels can be reduced. By performing FFT processing again using the above stored information, it is possible to separate signals corresponding to each reflection point, as shown in Figure 7. The line types in Figures 5 and 7 are the same as those in Figure 4, and the solid lines in Figure 7 confirm that each signal can be stored in an independent speed bin for each of the three speed bins.

[0041] In the comparative example in which the measurement time is lengthened, in comparison with the case where 2048 points of time series data are held and 2048-point FFT is performed, if 512 points of time series data are held in one processing cycle, if all measurement data from four processing cycles are stored, 512 x 4 points of time series data must be held. Although the total number of data is the same, the processing load is improved by dividing the multi-point FFT processing as in this embodiment. Furthermore, if the second speed processing is performed only for the speed of interest, only the number of interesting bins N x 4 points of data need to be stored and held, and the memory load can be reduced by deleting data related to unnecessary speeds that are not of interest. In addition, in this embodiment, one processing cycle has a transmission pause period, and only during the transmission period, about 256 points of received time series data are held and 256-point FFT is performed, but since the speed processing is closed within one cycle, it is also applicable to radars that include such pause periods. Note that the second speed processing is not limited to a method using information stored by four processing cycles. In addition, from the viewpoint of reducing the processing load, it is preferable to perform processing without 0 padding.

[0042] The above-mentioned processing is performed simultaneously or equivalently on the received signals for each of the multiple antenna channels. As a result, the second speed bin signal for each distance bin, in which information corresponding to each part of the reflection point is stored in an independent speed bin, is held for each antenna channel. Angle information is obtained by performing angle processing such as FFT on the signal for each antenna channel. In a normal radar device, the speed processing is not repeated and the above angle processing is performed using a single speed bin signal for each distance bin. However, in this embodiment, in addition to the first speed bin signal for each distance bin after the first speed processing, the second speed bin signal for each distance bin, which is the signal after the second speed processing, is utilized.

[0043] As a result of the second speed processing, the signals of each reflection point stored in an independent speed bin can be obtained as angle information corresponding to each reflection point by angle processing performed independently for each speed bin, for example, as shown by the dashed line, the dashed line, and the broken line in FIG. 9. On the other hand, in the case of only the first speed processing, which has a short measurement time, the signals of the reflection points overlap in the same speed bin, so that only one peak (angle value) is obtained as angle information, as shown by the solid line, and only one detection point is obtained for the stopped vehicle, which is a fixed object on the ground. On the other hand, by additionally separating the speeds by the second speed processing and then performing angle processing, in this example, three detection points are obtained at the three reflection points at the left end, center, and right end of the stopped vehicle, which is a fixed object on the ground, and the target shape spread in the angle direction is detected with higher resolution. Note that the second speed processing is an additional separation in a process before the angle processing, and the speed bins after the processing are for convenience, and the information of the first speed processing, in which the speed itself is determined, may be used.

[0044] The above explanation was about the high-resolution detection of fixed objects on the ground in a direction close to the traveling direction while the mobile body is moving, that is, an example close to a free space application, but it is also useful in other applications. For example, in a blind spot detection application, other mobile bodies may run side by side with the mobile body on which the sensor is mounted, and it is difficult to separate the mobile bodies in the angular direction in terms of speed, especially when the relative speed is 0 or nearly equal to 0. Even in such a case, by performing the second speed processing on the signal of the speed bin with a relative speed of 0 or close to 0, it is possible to perform angle processing after additionally separating the speed components, and the target shape spread in the angular direction can be detected with higher resolution. Note that this effect is not limited to these applications.

[0045] [Specific method examples] In the above basic method, the conditions for storing the first speed bin signal for each distance bin in the storage process are omitted, but in this method example, the first speed bin signal for each distance bin to be stored is determined using all of the predetermined bin components, the information of the moving object on which the radar device 1 is mounted, and the processing result of the processing cycle prior to that processing cycle, as will be described below with reference to Figures 10 and 11. This process is the selection process (step S11).

[0046] In the selection process, a determination is made based on the speed (vehicle speed) of the moving body as vehicle information. That is, when the relative speed between the moving body and the target is close to the vehicle speed, the target is located in the traveling direction of the moving body, and the smaller the angle between the direction of the target as seen from the moving body and the traveling direction of the moving body (the target has an angle of view closer to the traveling direction), the smaller the difference in the relative speed between the moving body and each part of the target is, and separation by speed components is difficult. Therefore, a bin component whose relative speed is close to the vehicle speed becomes an interesting bin. That is, in a free space application for recognizing the travelable area of ​​a moving body, it is effective to set such an interest bin when detecting ground fixed objects such as guard rails and walls on the front side and roadside shape detection. Therefore, in the selection process, the vehicle speed in the previous processing cycle is acquired (step S21), and bin components within a range obtained by adding or subtracting a predetermined number from the vehicle speed are selected as interest bins (step S22), and it is determined whether each bin component corresponds to a bin to be selected (step S23).

[0047] In the illustrated example, the vehicle information in the previous processing cycle is acquired, but it is sufficient to acquire the latest vehicle information at an appropriate timing according to the timing of communication with the moving object, etc. For example, the vehicle information may be acquired in the current processing cycle.

[0048] In addition, in the selection process, a determination is also made based on the response information in the previous processing cycle. That is, in each processing cycle, a response (i.e., the presence of a target candidate or a ground fixed object candidate) is determined based on the result of threshold processing. In order to increase the resolution of targets and ground fixed objects generated using the response, it is desirable to increase the resolution of the response itself, and such bin components become bins of interest. Therefore, response determination information is obtained from the previous processing cycle (step S31), a bin with a response is selected as an interesting bin (step S32), and it is determined whether each bin component corresponds to a bin to be selected (step S33).

[0049] In addition, in the selection process, a judgment is also made based on predetermined bin components. For targets with a relative speed close to 0, it is difficult to separate the speeds, and detailed information is difficult to obtain, so high definition is desired, and such bin components become interest bins. That is, in a blind spot detection application that detects a predetermined target of another moving body (such as a vehicle or a motorcycle) in a blind spot behind and to the side of a moving body and warns the driver when changing lanes, it is effective to set such interest bins. Therefore, for the first speed bin signal for each distance bin acquired by the first speed process, it is judged whether each bin component corresponds to a bin to be selected based on whether the relative speed is equal to or less than a predetermined value (for example, 5 km / h or less) (step S41).

[0050] The above steps S21 to S23, S31 to S33, and S41 correspond to the selection process (step S11). If it is determined that the bin corresponds to a bin to be selected (Yes in steps S23, S33, and S41), the first speed bin signal for each distance bin is stored (step S12). Furthermore, for each first speed bin signal for each distance bin, it is determined whether or not a predetermined number of cycles have been stored (step S13), and if the predetermined number of cycles have been stored (Yes in step S13), the second speed process is performed on the first speed bin signal for each distance bin of the bin component.

[0051] When the above-mentioned selection process is performed, which speed bin is stored and the second speed process is performed will be described with reference to FIG. 12. In FIG. 12, the vertical direction indicates the speed bins, and the processing cycle progresses from left to right. The black circles and the graph connecting them indicate the speed of the moving object. Also, the star marks indicate the speed bins where the response is confirmed. The hatched squares with diagonal lines going from the upper left to the lower right (hereinafter referred to as the execution bins) indicate the speed bins where the information of the number of cycles determined in step S13 is stored and where the second speed process is actually performed. The hatched squares with diagonal lines going from the upper right to the lower left (hereinafter referred to as the planned execution bins) indicate the speed bins where the information is stored but the number of cycles determined in step S13 is not reached, so the second speed process is not performed (it may be performed later).

[0052] The first and second speed bins from the bottom are all executed bins. This is because the relative speed with respect to the target is equal to or less than a predetermined value, and is a speed bin that is determined to be applicable in step S41. Also, the response is determined in the third processing cycle from the left, but in this speed bin, the number of cycles determined in step S13 is not met in the fourth to sixth processing cycles, so the second speed process is not performed and the bin is scheduled to be executed. After that, the number of cycles is met in the seventh processing cycle, so the second speed process is performed and the bin becomes an executed bin.

[0053] Furthermore, the speed bin corresponding to the speed of the moving object and the speed bins before and after it are determined as bins of interest in step S22. In the fourth processing cycle from the left, the speed bin one step below the speed bin corresponding to the speed of the moving object does not have the number of cycles determined in step S13, so the second speed process is not performed on the speed bin below, but information is stored as the bin of interest. As a result, in the seventh processing cycle from the left, even when the speed of the moving object changes, the second speed process can be performed immediately using the information stored in advance. It is also possible to set the bin of interest with a certain degree of spread in this manner as appropriate.

[0054] According to the radar device 1 according to the embodiment of the present invention as described above, in the angle processing, the first speed bin angle signal per distance bin and the second speed bin angle signal per distance bin are acquired based on the unstored first speed bin signal per distance bin and the stored second speed bin signal per distance bin, respectively, so that the first speed bin angle signal per distance bin can suppress latency while the second speed bin angle signal per distance bin can obtain high-definition information. That is, high-definition information can be obtained while suppressing latency. Furthermore, memory load and processing load can be reduced compared to a method of simply lengthening the measurement time and performing speed processing.

[0055] In addition, the data after the first speed processing is used to perform FFT processing in the second speed processing, i.e., the speed processing is repeated twice, thereby achieving high resolution. The input signal to the second speed processing is the output result of the first speed processing, and is not thinned out with respect to the measurement time of the first speed processing. In other words, by using the integrated result without loss, the result is equivalent to using all the measured signal in the speed processing, making it possible to maximize the SN ratio. In addition, by using the result of first-speed processing closed within one transmission period of cycle as the input signal for second-speed processing, the equal interval required for second-speed processing can be ensured as long as the equal interval of each processing cycle can be ensured. In other words, it is possible to configure a radar device with a high degree of freedom, such as a radar device with a long pause period within a processing cycle, or a radar device in which the processing cycle is set regardless of the chirp period or a multiple of the transmission signal interval, as shown in Figure 3.

[0056] In addition, in the storage process, a part of the first speed bin signals for each distance bin is stored as storage information, thereby reducing the memory load. Here, since the information after the first speed process includes not only distance but also speed information, the memory load can be efficiently reduced by storing only the distance and speed components of interest and not storing unnecessary distance and speed components at this stage. In addition, in the second speed process, a smaller number of first speed bin signals for each distance bin than all the first speed bin signals for each distance bin acquired in the first speed process are processed, thereby further reducing the processing load.

[0057] In addition, in the storage process, by storing signals for each distance bin and each first speed bin based on bin components where the relative speed is below a predetermined value, it is possible to detect objects with low relative speeds with high precision, thereby improving the accuracy of, for example, perimeter monitoring applications.

[0058] In addition, in the storage process, the first speed bin signal for each distance bin is stored based on the bin components that make the relative speed between the moving body on which the radar device 1 is mounted and the target close to the speed of the moving body, thereby making it possible to detect objects located directly in front of the moving body in the direction of travel with high precision, thereby improving the accuracy of, for example, roadside shape detection applications far away in the direction of travel.

[0059] In addition, in the storage process, the signal for each distance bin and the first speed bin is stored based on the response information in the previous processing cycle. As a result, at the stage where distance processing, speed processing, and angle processing are completed, threshold processing is performed on the signal with the highest S / N ratio due to the integral effect, and by efficiently limiting the target candidates of interest, high-resolution detection can be achieved while reducing memory load.

[0060] In addition, in threshold processing, by setting different threshold values ​​or multiplying the angle signal for each first speed bin per distance bin and the angle signal for each second speed bin per distance bin by different coefficients, it is possible to set appropriate weights when merging the angle signal for each first speed bin per distance bin and the angle signal for each second speed bin per distance bin.

[0061] The present invention is not limited to the above-described embodiment, and includes other configurations that can achieve the object of the present invention, and the following modifications are also included in the present invention. For example, in the above-described embodiment of the present invention, in a certain processing cycle, the second speed bin signal for each distance bin acquired by the second speed processing of the processing cycle is subjected to angle processing. However, as shown in FIG. 13, the second speed bin signal for each distance bin acquired by the second speed processing of the processing cycle may not be subjected to angle processing, and the second speed bin signal for each distance bin acquired by the second speed processing of the previous processing cycle may be subjected to angle processing. According to such a processing method, it is possible to ensure a time margin between the second speed processing and the angle processing.

[0062] In the above embodiment of the present invention, the FFT process is exemplified as a specific process of the first speed process and the second speed process, but the present invention is not limited to this. That is, it is preferable to adopt the FFT process from the viewpoint of reducing the processing load, but it is sufficient to adopt an appropriate processing method according to the processing conditions, purpose, etc.

[0063] In addition, in the above embodiment of the present invention, in the second speed processing, a smaller number of first speed bin signals for each distance bin are processed than all of the first speed bin signals for each distance bin acquired in the first speed processing. However, for example, if there is room in the memory of the radar device, the second speed processing may be performed on all of the first speed bin signals for each distance bin.

[0064] In addition, in the above embodiment of the present invention, in the storage process, the first speed bin signal for each distance bin to be stored is selected based on all of the predetermined bin components, information about the moving body on which the radar device is mounted, and the processing result of the previous processing cycle. However, the first speed bin signal for each distance bin to be stored may be selected based on only one of these, or the first speed bin signal for each distance bin to be stored may be selected based on two of these, or the selection may be based on other conditions.

[0065] Furthermore, when selecting the first speed bin signal for each distance bin to be stored based on the processing result of the previous processing cycle, a mode determination process for determining the driving mode by acquiring information on the moving body may be executed before the storage process, and the first speed bin signal for each distance bin to be stored based on the driving mode may be selected in the storage process. For example, the driving mode may be determined based on information on the gear selected in the transmission, the turn indicator, the steering angle, the brake, etc., and the first speed bin signal for each distance bin to be stored may be selected. By determining the driving mode in this way and selecting the first speed bin signal for each distance bin to be stored based on the driving mode, the number of processes can be more efficiently limited.

[0066] In addition, in the above embodiment of the present invention, when selecting the signal for each distance bin and each first speed bin to be stored based on the processing result of the previous processing cycle, response determination information, which is information after threshold processing, is used, but information before threshold processing may also be used.

[0067] In addition, in the above embodiment of the present invention, in the judgment process, different threshold values ​​are set for the first speed bin signal per distance bin and the second speed bin signal per distance bin, or the first speed bin angle signal per distance bin and the second speed bin angle signal per distance bin, or the threshold values ​​are multiplied by different coefficients. However, for example, if the first speed bin angle signal per distance bin and the second speed bin angle signal per distance bin are used independently without being merged, these processes do not need to be performed.

[0068] In the above embodiment of the present invention, the distance information of the second speed bin angle signal for each distance bin is used as is, but a correction process for correcting this distance information may be performed. That is, in a certain processing cycle, after the angle process, the distance information of the second speed bin angle signal for each distance bin may be corrected based on the first speed bin signal for each distance bin acquired in a processing cycle prior to this processing cycle. When the second speed bin signal for each distance bin is processed, the first speed process has already been completed and the speed information has been acquired, so that an appropriate correction process according to the speed and the time latency can be performed on the distance direction from the radar device, i.e., the direction in which the relative speed occurs.

[0069] In the above embodiment of the present invention, the description of other processes between processing cycles has been omitted. However, for example, after the threshold processing and response determination, a ground fixed object determination process, a clustering process, a tracking process, a targeting process, etc. may be performed based on the angle signal for each first speed bin per distance bin and the angle signal for each second speed bin per distance bin. Only one of these post-processing processes may be performed, or multiple processes may be performed. In this case, the signal derived from the first angle signal for each distance and speed and the signal derived from the angle signal for each second speed bin per distance bin may be mixed and the above process may be performed. By handling the mixed signals in the same processing process, it is not necessary to add a special process, and the processing can be simplified.

[0070] Although the embodiment of the present invention has been described above, the present invention is not limited to the radar device according to the embodiment of the present invention, and includes all aspects included in the concept of the present invention and the scope of the claims. In addition, each configuration may be appropriately and selectively combined so as to achieve at least a part of the above-mentioned problems and effects. For example, the shape, material, arrangement, size, etc. of each component in the above embodiment may be appropriately changed depending on the specific use mode of the present invention. [Explanation of symbols]

[0071] 1... radar device, 18... signal processing section, 181... distance processing section, 182... first speed processing section, 183... storage section, 184... second speed processing section, 185... angle processing section, 186... threshold processing section

Claims

1. A radar device that detects a target by transmitting and receiving signals and processing received signals, Executing the processing cycle multiple times, Each of said processing cycles comprises: Range processing to obtain a signal for each range bin; a first velocity process for acquiring a first velocity bin signal for each range bin based on a plurality of range bin signals repeatedly acquired within the processing cycle; a storing process for storing at least a part of the signals for each distance bin and each first velocity bin as stored information based on a predetermined condition; a second speed process for acquiring a second speed bin signal for each distance bin based on the stored information stored in a plurality of processing cycles prior to the current processing cycle; An angle process for acquiring an angle signal for each first speed bin per range bin and an angle signal for each second speed bin per range bin based on the first speed bin signal for each range bin and the second speed bin signal for each range bin acquired in a plurality of receiving channels; and threshold processing for determining each signal strength based on a predetermined threshold at least before or after the angle processing.

2. 2. The radar device according to claim 1, wherein the second speed processing is performed for a number of bins that is less than the number of bins of all of the first speed bin signals for each distance bin acquired in the first speed processing.

3. The radar device according to claim 1 or 2, characterized in that in the storage process, the signal for each distance bin and each first velocity bin to be stored is selected based on at least one of predetermined bin components, information on the moving body on which the radar device is mounted, and the processing result of a processing cycle prior to the current processing cycle.

4. 4. The radar device according to claim 3, wherein in the storing process, the signal for each distance bin and each first velocity bin in predetermined bin components including a bin component corresponding to a relative velocity of 0 is selected and stored.

5. 4. The radar device according to claim 3, wherein in the storing process, a signal for each distance bin and each first speed bin in a bin component including a speed of a moving object on which the radar device is mounted is selected and stored.

6. the processing cycle includes, before the storing process, a mode determining process of determining a traveling mode by acquiring information of the moving body, 4. The radar device according to claim 3, wherein in the storing process, the signals to be stored for each distance bin and each first speed bin are selected based on the driving mode.

7. 4. The radar device according to claim 3, wherein in the storage process, the signal to be stored for each distance bin and each first velocity bin is selected based on information obtained by performing the angle processing and the threshold processing in the processing cycle prior to the current processing cycle.

8. The radar device according to any one of claims 1 to 7, characterized in that in the threshold processing, different thresholds are set for the first speed bin signal for each range bin and the second speed bin signal for each range bin, or the first speed bin angle signal for each range bin and the second speed bin angle signal for each range bin, in terms of threshold values, or a common threshold is set after multiplying the signals by different coefficients.

9. The radar device according to any one of claims 1 to 8, characterized in that the processing cycle includes, after the angle processing, a correction process for correcting distance information of the angle signal for each distance bin and each second speed bin based on the signal for each distance bin and each first speed bin acquired in a processing cycle prior to the processing cycle.

10. The radar device according to any one of claims 1 to 9, characterized in that in the processing cycle, after the threshold processing, a signal derived from the angle signal for each first speed bin for each distance bin and a signal derived from the angle signal for each second speed bin for each distance bin are mixed, and at least one of a ground fixed object determination process, a clustering process, a tracking process, and a targeting process is performed.

11. A method for detecting a target by transmitting and receiving a signal and processing a received signal, comprising the steps of: Executing the processing cycle multiple times, Each of said processing cycles comprises: Range processing to obtain a signal for each range bin; a first velocity process for acquiring a first velocity bin signal for each range bin based on a plurality of range bin signals repeatedly acquired within the processing cycle; a storing process for storing at least a part of the signals for each distance bin and each first velocity bin as stored information based on a predetermined condition; a second speed process for acquiring a second speed bin signal for each distance bin based on the stored information stored in a plurality of processing cycles prior to the current processing cycle; An angle process for acquiring an angle signal for each first speed bin per range bin and an angle signal for each second speed bin per range bin based on the first speed bin signal for each range bin and the second speed bin signal for each range bin acquired in a plurality of receiving channels; A target detection method comprising: a threshold process for determining each signal strength based on a predetermined threshold at least before or after the angle process.

12. A target detection program for detecting a target by transmitting and receiving a signal and processing a received signal, On the computer, Performing a processing cycle multiple times; Each of the processing cycles includes: Range processing to obtain a signal for each range bin; a first velocity process for acquiring a first velocity bin signal for each range bin based on a plurality of range bin signals repeatedly acquired within the processing cycle; a storing process for storing at least a part of the signals for each distance bin and each first velocity bin as stored information based on a predetermined condition; a second speed process for acquiring a second speed bin signal for each distance bin based on the stored information stored in a plurality of processing cycles prior to the current processing cycle; An angle process for acquiring an angle signal for each first speed bin per range bin and an angle signal for each second speed bin per range bin based on the first speed bin signal for each range bin and the second speed bin signal for each range bin acquired in a plurality of receiving channels; and threshold processing for determining each signal strength based on a predetermined threshold at least either before or after the angle processing.

Images