JP2020060530A - Radar device and signal processing method - Google Patents

Abstract

To provide a radar device that can upgrade accuracy of pattern matching.SOLUTION: A matching determination unit of a radar device is configured to, when a matching rate of first pattern matching based on a detection point serving as a reflection point of an object and a template of a prescribed shape is greater than a first threshold, determine that the first pattern matching is established. When the matching rate of the first pattern matching is greater than a second threshold and is equal to or less than the first threshold, a virtual detection point generation unit of the radar device is configured to generate a virtual detection point relevant to second pattern matching. When a matching rate of third pattern matching based on at least one part of the virtual detection point and the detection point and the template of the prescribed shape is greater than a third threshold, the matching determination unit is configured to determine that the third pattern matching is established.SELECTED DRAWING: Figure 1

Description

  The present invention relates to a technique for recognizing the shape of an object detected by a radar device.

  Various radar devices that perform pattern matching to recognize the shape of an object have been developed (see, for example, Patent Document 1). A radar device that recognizes the shape of an object by performing pattern matching detects a detection point that is a reflection point of the object based on a reception signal output from a reception antenna that receives a reflected wave from the object, and the detection point Pattern matching is performed based on the time-series information and the template of a predetermined shape.

JP, 2010-185769, A

  For example, as shown in FIG. 6, in a scene in which a vehicle V11 equipped with a front radar device 11 searches for a parking space while moving forward, detection points (black in FIG. 6) at the front face F11 of the parked vehicle V12 and the front side portion of the left side face L11 are detected. (Triangle) is easily detected, and detection points are difficult to be detected on the rear side portion of the left side surface L11, the right side surface R11, and the back surface B11 of the parked vehicle V12. The front radar device 11 is a radar device that detects an object existing in front of the vehicle V11.

There are three reasons why it is difficult to detect the detection points on the rear side portion of the left side surface L11, the right side surface R11, and the back surface B11 of the parked vehicle V12.
(1) The rear portion of the left side surface L11 of the parked vehicle V12 is close to the boundary of the FOV (Field of View) 12 of the front radar device 11. (2) The rear side portion of the left side surface L11 of the parked vehicle V12 is the front radar. The incident angle of the transmission wave of the device 11 becomes large, and the reflected wave is difficult to return to the front radar device 11 (3) The transmission wave of the front radar device 11 does not reach the right side surface R11 and the back surface B11 of the parked vehicle V12 in the first place. thing

  As described above, since the distribution of the detection points in the parked vehicle V12 can reproduce only about half the shape of the parked vehicle V12, the accuracy of the pattern matching becomes poor, and the front radar device 11 may not be able to correctly recognize the shape of the parked vehicle V12. There is.

  In view of the above problems, it is an object of the present invention to provide a radar device and a signal processing method that can improve the accuracy of pattern matching.

  The radar device according to the present invention includes a detection unit that detects a detection point that is a reflection point of the object based on a reception signal that is output from a reception antenna that receives a reflected wave from the object, the detection point, and a template of a predetermined shape. A first matching unit that calculates a matching rate of the first pattern matching by performing a first pattern matching based on, and a matching when the second pattern matching is performed based on a virtual detection point and the detection point and the template. Based on a virtual detection point generation unit that generates the virtual detection point so that the ratio becomes a predetermined value, at least a part of the virtual detection point generated by the virtual detection point generation unit, and the detection point and the template. A second matching unit that performs a third pattern matching to calculate a matching rate of the third pattern matching; A matching determining unit, the matching determining unit determines that the first pattern matching is established when the matching ratio of the first pattern matching is larger than a first threshold, and the matching ratio of the first pattern matching is The virtual detection point generation unit generates the virtual detection point when the value is greater than a second threshold value and equal to or less than the first threshold value, and the matching determination is performed when a matching rate of the third pattern matching is greater than a third threshold value. The unit determines that the third pattern matching is established, the predetermined value is greater than the first threshold, and the third threshold is equal to or more than the first threshold and less than the predetermined value (first configuration).

  The radar device of the first configuration further includes a validity determination unit, and when at least one of the first condition and the second condition is satisfied, the validity determination unit determines that the virtual detection point is valid, The matching unit performs the third pattern matching based on one of the virtual detection points generated by the virtual detection point generation unit that is determined to be valid by the validity determination unit, and the first condition is the radar device. Is a condition that the direction of the virtual detection point is outside the first predetermined range, and the second condition is that the transmission wave of the radar device is incident on the surface of the template where the virtual detection point exists. A configuration (second configuration) may be adopted in which the incident angle of the transmission wave in the hypothetical case is outside the second predetermined range.

  In the radar device having the second configuration, even if neither the first condition nor the second condition is satisfied, if the third condition is satisfied, the validity determination unit determines that the virtual detection point is valid. The third condition is a condition that at least one of the detection point and the effective virtual detection point exists between the radar device and the virtual detection point to be determined (third configuration). May be

  In the radar device having the second configuration, even if neither the first condition nor the second condition is satisfied, if the third condition is satisfied, the validity determination unit determines that the virtual detection point is valid. The third condition is a condition that a line segment connecting the radar device and the virtual detection point that is the determination target passes through a surface on which the virtual detection point that is the determination target of the template does not exist (the first condition). 4 configuration).

  In the radar device having any one of the second to fourth configurations, the validity determining unit determines whether or not the virtual detection point is valid based on time series information of the virtual detection point (fifth configuration). Configuration).

  The radar device of the fifth configuration is a configuration (sixth configuration) for determining whether or not the virtual detection point is effective based on each position of the radar device corresponding to each of the time series information. May be.

  In the radar device having any one of the first to sixth configurations, when there are a plurality of types of templates in which the matching rate of the first pattern matching is greater than a second threshold value and equal to or less than the first threshold value, the virtual detection point generation The unit may have a configuration (seventh configuration) that generates the virtual detection points for each of the plurality of types of templates.

  A signal processing method according to the present invention is a signal processing method for a radar device, which detects a detection point which is a reflection point of the object based on a reception signal output from a reception antenna which receives a reflected wave from the object. A first matching step of performing a first pattern matching based on the detection point and a template of a predetermined shape, and calculating a matching rate of the first pattern matching; a virtual detection point, the detection point and the template. At least one of the virtual detection point generation step of generating the virtual detection point so that the matching rate when performing the second pattern matching based on the virtual detection point generation step, and the virtual detection point generated by the virtual detection point generation step The third pattern matching is performed based on the part, the detection point, and the template, and the third pattern matching map is performed. A second matching step of calculating a ringing rate, and a matching determination step, and when the matching rate of the first pattern matching is larger than a first threshold value, it is determined that the first pattern matching is established in the matching determination step. When the matching rate of the first pattern matching is greater than the second threshold value and is equal to or less than the first threshold value, the virtual detection point is generated in the virtual detection point generation step, and the matching rate of the third pattern matching is When it is larger than a third threshold value, it is determined that the third pattern matching is established in the matching determination step, the predetermined value is larger than the first threshold value, and the third threshold value is equal to or more than the first threshold value and less than the predetermined value. It is a certain configuration (eighth configuration).

  According to the present invention, it is possible to provide a radar device and a signal processing method capable of improving the accuracy of pattern matching.

Diagram showing an example of the configuration of a radar device Flowchart showing an operation example of the shape recognition unit A bird's eye view showing a scene where a vehicle equipped with a front radar device searches for a parking space while moving forward Flowchart showing an example of validity determination processing Flowchart showing another example of validity determination processing A bird's-eye view showing a scene in which a vehicle equipped with a front radar device searches for a parking space while moving forward in the prior art

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings.

<1. Structure of radar device>
FIG. 1 is a diagram showing a configuration example of a radar device. The radar device 1 is mounted on, for example, a vehicle. Hereinafter, the vehicle in which the radar device 1 is mounted is referred to as “own vehicle”.

  The radar device 1 is attached to the center of the front end of the host vehicle, and acquires target data related to the target existing in front of the host vehicle by using a frequency-modulated continuous wave (FMCW).

  As shown in FIG. 1, the radar device 1 mainly includes a transmission unit 2, a reception unit 3, and a signal processing device 4.

  The transmitter 2 includes a signal generator 21 and a transmitter 22. The signal generation unit 21 generates a modulation signal whose voltage changes in a triangular wave shape and supplies it to the oscillator 22. The oscillator 22 frequency-modulates the continuous wave signal based on the modulation signal generated by the signal generation unit 21, generates a transmission signal whose frequency changes with the passage of time, and outputs the transmission signal to the transmission antenna 23.

  The transmission antenna 23 outputs a transmission wave TW based on the transmission signal from the oscillator 22. The transmission wave TW output by the transmission antenna 23 becomes an FMCW whose frequency fluctuates in a predetermined cycle. The transmission wave TW transmitted from the transmission antenna 23 to the left front side of the host vehicle becomes a reflected wave RW when reflected by an object such as a person or another vehicle.

  The receiving unit 3 includes a plurality of receiving antennas 31 forming an array antenna and a plurality of individual receiving units 32 connected to the plurality of receiving antennas 31. In the present embodiment, the receiving unit 3 includes, for example, four receiving antennas 31 and four individual receiving units 32. The four individual receiving units 32 correspond to the four receiving antennas 31, respectively. Each reception antenna 31 receives the reflected wave RW from the object to acquire a reception signal, and each individual reception unit 32 processes the reception signal obtained by the corresponding reception antenna 31.

  Each individual reception unit 32 includes a mixer 33 and an A / D converter 34. The reception signal obtained by the reception antenna 31 is amplified by a low noise amplifier (not shown) and then sent to the mixer 33. The transmission signal from the transmitter 22 of the transmission unit 2 is input to the mixer 33, and the transmission signal and the reception signal are mixed in the mixer 33. As a result, a beat signal having a beat frequency that is the difference between the frequency of the transmission signal and the frequency of the reception signal is generated. The beat signal generated by the mixer 33 is output to the signal processing device 4 after being converted into a digital signal by the A / D converter 34 after adjusting the timing between the receiving antennas 31 by a synchronizing circuit (not shown). To be done.

  The signal processing device 4 includes a microcomputer including a CPU (Central Processing Unit) and a memory. The signal processing device 4 can generate the target data based on the signal supplied from each A / D converter 34. Any known method can be used as a method for generating the target data based on the signal supplied from each A / D converter 34.

  The signal processing device 4 includes a transmission control unit 41, a detection unit 42, and a shape recognition unit 43 as functions implemented by software in a microcomputer.

  The transmission controller 41 controls the signal generator 21 of the transmitter 2.

  The detection unit 42 detects a detection point, which is a reflection point of the object, based on the reception signal output from the reception antenna 31 that receives the reflected wave RW from the object. More specifically, the detection unit 42 detects a detection point, which is a reflection point of an object, based on the beat signal represented by the digital signal format supplied from each A / D converter 34 of the reception unit 3, and detects the detection point. Generate detection point data related to points.

  The detection unit 42 detects only the reflection point that is the source of the reflected wave RW received by the reception antenna 31. In other words, the detection unit 42 does not detect (cannot detect) the reflection point that is the source of the reflected wave RW that is not received by the receiving antenna 31 (the reflected wave RW that does not reach the receiving antenna 31).

  The detection point data includes information indicating the azimuth of the detection point with respect to the radar device 1, the position of the detection point with respect to the radar device 1, and the position of the detection point with respect to the fixed reference point. The detection point data is a kind of target data. The detection unit 42 provides information indicating the position of the detection point with respect to the fixed reference point on the basis of the information on the amount of movement of the vehicle supplied from the vehicle speed sensor 5, the steering angle sensor 6 and the like and the position of the detection point with respect to the radar device 1. To generate.

  The shape recognition unit 43 includes a first matching unit 43a, a virtual detection point generation unit 43b, a second matching unit 43c, a validity determination unit 43d, and a matching determination unit 43e.

  The first matching unit 43a performs the first pattern matching based on the detection points detected by the detection unit 42 and the template having a predetermined shape (for example, a quadrangle), and calculates the matching rate MR1 of the first pattern matching. In the first pattern matching, the detection point and the template are matched. It is desirable to prepare multiple types of templates. For example, a template showing the plane shape of a private car (the shape seen from directly above), a template showing the plane shape of a bus, a template showing the plane shape of a bicycle, and the like are conceivable. The data related to the template may be stored in the memory of the signal processing device 4 in a non-volatile manner. In pattern matching, for example, the template is reduced or enlarged at a magnification based on the distance between the radar device 1 and the detection point.

  The virtual detection point generation unit 43b generates the virtual detection points so that the matching rate MR2 when performing the second pattern matching based on the virtual detection points and the detection points and the template becomes a predetermined value. In the second pattern matching, all the virtual detection points generated by the virtual detection point generation unit 43b and the detection points are matched with the template.

  The second matching unit 43c performs the third pattern matching based on at least a part of the virtual detection points generated by the virtual detection point generation unit 43b and the detection points and the template to calculate the matching rate MR3 of the third pattern matching. . In the third pattern matching, at least a part of the virtual detection points generated by the virtual detection point generation unit 43b and the detection points are matched with the template.

  The validity determination unit 43d determines whether the virtual detection point generated by the virtual detection point generation unit 43b is valid.

  The matching determination unit 43e determines whether the first pattern matching is established, and also determines whether the third pattern matching is established.

<2. Operation of shape recognition part>
Next, the operation of the shape recognition unit 43 will be described. FIG. 2 is a flowchart showing an operation example of the shape recognition unit 43. The detection unit 42 periodically repeats detection of detection points at regular intervals. Therefore, the detection timing of the detection point arrives at regular intervals.

  For example, when the vehicle speed of the host vehicle becomes equal to or lower than a predetermined speed, the shape recognition unit 43 starts the flow operation shown in FIG. 2, and when the vehicle speed of the host vehicle exceeds the predetermined speed, the shape recognition unit 43 performs the flow shown in FIG. The operation may be ended. Accordingly, for example, the shape of the parked vehicle can be recognized in a scene where the own vehicle is moving forward at a low speed to search for the parking space, and it is possible to assist the search for the parking space.

  First, the shape recognition unit 43 determines whether or not N times of detection timings have arrived (step S1), and when N times of detection timings have arrived, the shape recognition unit 43 proceeds to step S2. In the following description, the case where N = 4 will be described as an example.

  At the time of shifting from step S1 to step S2, the first matching unit 43a acquires the information on the detection points related to the parked vehicle V2 detected at the first detection timing to the fourth detection timing. Note that FIG. 3 shows the position of the host vehicle V1 at each of the a-th detection timing ta to the d-th detection timing td, and indicates the position of the parking vehicle V2 detected at the a-th detection timing to the d-th detection timing. The detection points involved are indicated by black triangles. In addition, a, b, c, and d in FIG. 3 mean four continuous natural numbers.

  In step S2, the first matching unit 43a performs the first pattern matching based on the detection points detected by the detection unit 42 at the latest four detection timings and the template of the predetermined shape, and the matching rate MR1 of the first pattern matching. To calculate.

  In step S3 following step S2, the matching determination unit 43e determines whether the matching rate MR1 of the first pattern matching is larger than the first threshold value TH1. If the matching rate MR1 of the first pattern matching is larger than the first threshold value TH1, the matching determination unit 43e determines that the first pattern matching is established (step S4). Thereby, the shape recognition unit 43 recognizes that the shape of the object related to the detection point detected by the detection unit 42 is the shape of the template. When the process of step S4 ends, the process proceeds to step S11 described below.

  If the matching rate MR1 of the first pattern matching is not greater than the first threshold TH1, the matching determination unit 43e determines whether the matching rate MR1 of the first pattern matching is greater than the second threshold TH2 (step S5). The second threshold TH2 is a value smaller than the first threshold TH1.

  If the matching rate MR1 of the first pattern matching is not larger than the second threshold value TH2, the shape recognition unit 43 cannot recognize the shape of the object related to the detection point detected by the detection unit 42, and therefore, it will be immediately described in step S11. Move to.

  On the other hand, if the matching rate MR1 of the first pattern matching is larger than the second threshold value TH2, the shape recognition unit 43 may be able to recognize the shape of the object related to the detection point detected by the detection unit 42, and thus the process proceeds to step S6. To do.

  In step S6, the virtual detection point generation unit 43b determines that the matching rate MR2 when the second pattern matching is performed based on the virtual detection point and the detection point detected by the detection unit 42 at the latest four detection timings and the template. A virtual detection point is generated so as to have a predetermined value PD1. In the second pattern matching, all the virtual detection points generated by the virtual detection point generation unit 43b and the detection points detected by the detection unit 42 are matched with the template at the latest four detection timings. The predetermined value PD1 is a value larger than the first threshold TH1. The white triangles in FIG. 3 indicate virtual detection points.

  In step S7 following step S6, the validity determination unit 43d determines whether the virtual detection point generated by the virtual detection point generation unit 43b is valid. Details of the validity determination will be described later.

  In step S8 subsequent to step S7, the second matching unit 43c performs the third pattern matching based on the effective virtual detection point and the detection point detected by the detection unit 42 at the latest four detection timings, and the template. A matching rate MR3 of three pattern matching is calculated. In the third pattern matching, the detection point detected by the detection unit 42 and the template are matched at the effective virtual detection point and the latest four detection timings.

  In step S9 following step S8, the matching determination unit 43e determines whether the matching rate MR3 of the third pattern matching is larger than the third threshold TH3. The third threshold TH3 is a value equal to or larger than the first threshold TH1 and smaller than the predetermined value PD1.

  If the matching rate MR3 of the third pattern matching is larger than the third threshold value TH3, the matching determination unit 43e determines that the third pattern matching is established (step S10). Thereby, the shape recognition unit 43 recognizes that the shape of the object related to the detection point detected by the detection unit 42 is the shape of the template. When the process of step S10 ends, the process proceeds to step S11 described below.

  If the matching rate MR3 of the third pattern matching is not larger than the third threshold value TH3, the shape recognition unit 43 can recognize the shape of the object related to the detection point detected by the detection unit 42 even by using the virtual detection point. Since it has not been found, the process immediately proceeds to step S11 described later.

  In step S11, the shape recognition unit 43 determines whether the next detection timing has arrived, and when the next detection timing has arrived, the shape recognition unit 43 returns to step S2. As a result, after the fourth detection timing, the result of shape recognition by the shape recognition unit 43 is updated at each detection timing.

  By performing the processing of steps S5 to S10 described above, virtual detection points are added, and the recall ratio of the shape of the object whose shape is to be recognized is improved, so that the accuracy of pattern matching is improved.

  By performing the processing of step S7 described above, it is possible to validate only the virtual detection point at a position where it is appropriate that the detection point is not considered when the physical characteristics of the radar device 1 are taken into consideration. As a result, it is possible to prevent the incorrect shape from being recognized due to the addition of the virtual detection points.

<3. Details of validity judgment>
FIG. 4 is a flowchart showing an example of the validity determination process (step S7 in FIG. 2). The validity determining unit 43d executes the flow operation shown in FIG. 4 for each virtual detection point when executing step S7 in FIG. 2 once.

  In step S71, the validity determination unit 43d determines whether the first condition, that is, the absolute value of the azimuth angle of the virtual detection point is equal to or greater than the first predetermined angle θ1. In addition, the validity determination unit 43d determines whether or not the ratio of detection timings satisfying the first condition (hereinafter referred to as ratio RT1) is equal to or greater than a predetermined value α. Θa in FIG. 3 is the azimuth angle of the virtual detection point VDP1 at the detection timing ta. The azimuth angle of the virtual detection point is an angle formed by the line segment connecting the radar device 1 and the virtual detection point and the front direction of the host vehicle V1. For example, when the virtual detection point is on the right side of the host vehicle V1, the azimuth is positive. When the virtual detection point exists on the left side of the host vehicle V1, it becomes a negative value.

  In the present embodiment, the validity determination unit 43d acquires the absolute values | θa | to | θd | of the azimuth angles of the virtual detection points at the latest four detection timings ta to td, respectively, and calculates the ratio RT1. For example, only the absolute value | θa | of the azimuth angle of the virtual detection point at the detection timing ta is equal to or larger than the first predetermined angle θ1, and the absolute values | θb | to | of the azimuth angles of the virtual detection points at the detection timings tb to td, respectively. When θd | is less than the first predetermined angle θ1, the ratio RT1 becomes 0.25.

  If the ratio RT1 is equal to or greater than the predetermined value α, the validity determination unit 43d determines that the virtual detection point is valid (step S73), and then proceeds to step S74.

  On the other hand, if the ratio RT1 is not equal to or greater than the predetermined value α, the process proceeds to step S72.

  In step S72, the validity determination unit 43d determines that the incident angle of the transmission wave is the second predetermined value when it is assumed that the transmission wave of the radar device 1 is incident on the second condition, that is, the position where the virtual detection point on the surface of the template exists. It is determined whether the angle is θ2 or more. Further, the validity determination unit 43d determines whether or not the ratio of the detection timings satisfying the second condition (hereinafter referred to as ratio RT2) is equal to or greater than the predetermined value β. Θa ′ in FIG. 3 is the incident angle of the transmission wave when it is assumed that the transmission wave of the radar device 1 is incident at the detection timing ta at the position where the virtual detection point VDP1 on the surface F1 of the template TP is present. Under the second condition, the virtual detection point does not necessarily have to exist strictly on the surface of the template. In terms of design, it is sufficient that they are close to each other within a range that can be determined to exist on the surface. For example, a method may be used in which virtual detection points are generated by varying the measurement error of the radar device 1 to some extent. When the virtual detection point does not exist on the surface of the template, the surface close to the virtual detection point may be selected to evaluate the second condition. Further, when the virtual detection points are present close to a plurality of surfaces such as the corners of the template, either one surface may be evaluated, or both surfaces may be evaluated. When the template is a curved surface, the incident angle with respect to the contact surface at the virtual detection point may be used. The predetermined value β may be the same as the predetermined value α or may be different from the predetermined value α.

  In the present embodiment, the validity determination unit 43d acquires the incident angles θa ′ to θd ′ of the transmitted wave at the latest four detection timings ta to td, and calculates the ratio RT2. For example, only the incident angle θa ′ of the transmitted wave at the detection timing ta is less than the second predetermined angle θ2, and the incident angles θb ′ to θd ′ of the transmitted wave at each of the detection timings tb to td are equal to or larger than the second predetermined angle θ2. For example, the ratio RT2 becomes 0.75.

  If the ratio RT2 is equal to or greater than the predetermined value β, the validity determination unit 43d determines that the virtual detection point is valid (step S73), and then proceeds to step S74.

  On the other hand, if the ratio RT2 is not equal to or greater than the predetermined value β, the process proceeds to step S74 without proceeding to step S73.

  Step S74 is executed after the processes of steps S71 to S73 are completed for all the virtual detection points. In step S74, the validity determination unit 43d detects the third condition, that is, between the radar device 1 and the virtual detection point that is the determination target (on the line segment that connects the radar device 1 and the virtual detection point that is the determination target). It is determined whether or not at least one of the point and the effective virtual detection point exists. In addition, the validity determination unit 43d determines whether or not the ratio of the detection timings satisfying the third condition (hereinafter referred to as ratio RT3) is equal to or greater than the predetermined value γ. The predetermined value γ may be the same as the predetermined value α, may be different from the predetermined value α, may be the same as the predetermined value β, or may be different from the predetermined value β. The line segment connecting the radar device 1 and the virtual detection point to be determined may be a range on the line segment having a certain width. On the contrary, it is possible to determine whether or not to ride on the line segment by giving a spread range at the virtual detection point.

  In the present embodiment, the validity determining unit 43d has at least one of a detection point and a valid virtual detection point between the radar device 1 and the virtual detection point that is the determination target at each of the latest four detection timings ta to td. Whether or not to do so is determined, and the ratio RT3 is calculated. For example, at the detection timings ta and tb, at least one of the detection point and the effective virtual detection point exists between the radar device 1 and the virtual detection point that is the determination target, and the radar device 1 and the determination target are detected at the detection timings tc and td. If neither a detection point nor an effective virtual detection point exists between a certain virtual detection point, the ratio RT3 becomes 0.5.

  If the ratio Rt3 is equal to or greater than the predetermined value γ, the validity determination unit 43d determines that the virtual detection point that is the determination target is valid (step S75), and ends the flow operation.

  On the other hand, if the ratio RT3 is not equal to or larger than the predetermined value γ, the flow operation is ended without proceeding to step S75.

  By performing the processing of steps S71 to S73 described above, a virtual detection point at a position that is not a detection point when considering the reasons (1) and (2) described in paragraph 0005 of the present specification is determined. Can be enabled.

  By performing the processing of steps S74 to S75 described above, a virtual detection point at a position where it is appropriate that the detection point does not become a detection point when the reason (3) described in paragraph 0005 of the present specification is considered is validated. You can

  The same effect can be obtained by changing the processing content of step S74 shown in FIG. 4 described above (more specifically, the content of the third condition) so that the validity determination section 43d performs the flow operation shown in FIG. . When the validity determining unit 43d performs the flow operation shown in FIG. 5, step S74 does not have to be executed after the processes of steps S71 to S73 have been completed for all virtual detection points.

  In step S74 of the flowchart shown in FIG. 5, the validity determination unit 43d determines that the line segment connecting the third condition, that is, the radar device 1 and the virtual detection point to be determined is the other surface of the template (virtual detection to be determined). It is determined whether or not the object passes through a surface on which no point exists. Further, the validity determination unit 43d determines whether or not the ratio of the detection timings satisfying the third condition (hereinafter referred to as ratio RT4) is equal to or greater than the predetermined value γ. The above passage will be described with reference to an example. When the determination target is the virtual detection point VDP1 shown in FIG. 2, the line segment connecting the radar device 1 and the virtual detection point VDP1 at the detection timing ta is the virtual detection of the template TP. The point VDP1 does not pass through the plane (planes F2 to F4) where it does not exist.

  In the present embodiment, in the validity determination unit 43d, the line segment connecting the radar device 1 and the virtual detection point that is the determination target exists at each of the latest four detection timings ta to td that is the virtual detection point that is the determination target of the template. It is determined whether or not the surface passes, and the ratio RT4 is calculated. For example, at detection timings ta and tb, a line segment that connects the radar device 1 and the virtual detection point that is the determination target passes through a surface on which the virtual detection point that is the determination target of the template does not exist, and at the detection timings tc and td. If the line segment that connects the virtual detection point that is the determination target does not pass through the surface of the template on which the virtual detection point that is the determination target does not exist, the ratio RT4 is 0.5.

  In the present embodiment, the validity determination unit 43d determines whether the virtual detection point is valid based on the time-series information of the virtual detection point. For example, in steps S71 and S73, the virtual detection point is based on the absolute values | θa | to | θd | of the azimuth angles of the virtual detection points at the latest four detection timings ta to td corresponding to the time-series information of the virtual detection points. Is determined to be valid or not.

  By determining whether the virtual detection point is valid based on the time-series information of the virtual detection point, it is possible to suppress the momentary noise related to the information of the virtual detection point from adversely affecting the validity determination.

  In the present embodiment, the validity determination unit 43d determines whether or not the virtual detection point is valid based on each position of the radar device 1 corresponding to each time series information of the virtual detection point. For example, in step S71, the absolute value | θa | of the azimuth angle of the virtual detection point is calculated based on the position of the radar device 1 at the detection timing ta, and the absolute value | θb | of the azimuth angle of the virtual detection point is at the detection timing tb. The absolute value | θc | of the azimuth angle of the virtual detection point is calculated based on the position of the radar device 1, and the absolute value | θd of the azimuth angle of the virtual detection point is calculated based on the position of the radar device 1 at the detection timing tc. | Is calculated based on the position of the radar device 1 at the detection timing td.

  By determining whether or not the virtual detection point is valid based on each position of the radar device 1 corresponding to each time series information of the virtual detection point, the radar device corresponding to each time series information of the virtual detection point. The accuracy of validity determination can be improved as compared with the case where each position of 1 is approximated by one representative position.

  When there are a plurality of types of templates in which the matching rate MR1 of the first pattern matching is greater than the second threshold value TH2 and less than or equal to the first threshold value TH1, the virtual detection point generation unit 43 causes the virtual detection point generation unit 43 to detect the virtual detection points. Is desirable to generate. Then, the second matching unit 43c calculates the matching rate MR3 of the third pattern matching for each of the plurality of types of templates, and only for the third pattern matching of the template having the highest matching rate MR3 of the third pattern matching, steps S9 and S9. The process of S10 may be performed.

  As a result, the accuracy of validity determination can be improved as compared with the case where the virtual detection point generation unit 43 generates virtual detection points only for the template having the highest matching rate MR1 of the first pattern matching. Even if the matching rate MR1 of the first pattern matching for a certain template is lower than the matching rate MR1 of the first pattern matching for another template, the matching rate MR3 of the third pattern matching for a certain template is the third pattern for another template. This is because the matching rate MR3 may be higher than the matching rate MR3.

<4. Other>
Various technical features disclosed in this specification can be variously modified in addition to the above-described embodiment without departing from the spirit of the technical creation. In addition, a plurality of embodiments and modifications shown in the present specification may be combined and implemented within a possible range.

  For example, in the above-described embodiment, the radar device is the FMCW type radar device, but another type of radar device may be used. For example, an FCM (Fast-Chirp Modulation) type radar device may be used, for example.

  For example, although the radar device is mounted on the vehicle in the above-described embodiments, it may be mounted on a moving body other than the vehicle, such as a ship or an air vehicle.

  For example, step S71 and step S72 of the flowchart shown in FIG. 4 or the flowchart shown in FIG. 5 in the above-described embodiment may be replaced.

1 radar device 43 shape recognition unit 43a first matching unit 43b virtual detection point generation unit 43c second matching unit 43d validity determination unit 43e matching determination unit

Claims (8)

A detection unit that detects a detection point that is a reflection point of the object based on a reception signal output from a reception antenna that receives a reflected wave from the object,
A first matching unit that performs a first pattern matching based on the detection points and a template of a predetermined shape, and calculates a matching rate of the first pattern matching;
A virtual detection point and a virtual detection point generation unit that generates the virtual detection point such that the matching rate when performing the second pattern matching based on the detection point and the template becomes a predetermined value,
A second matching unit that performs third pattern matching based on at least a part of the virtual detection points generated by the virtual detection point generation unit and the detection points and the template to calculate a matching rate of the third pattern matching. When,
A matching determination unit,
When the matching rate of the first pattern matching is larger than a first threshold value, the matching determination unit determines that the first pattern matching is established,
When the matching rate of the first pattern matching is greater than a second threshold value and is equal to or less than the first threshold value, the virtual detection point generation unit generates the virtual detection point,
When the matching rate of the third pattern matching is larger than a third threshold value, the matching determination unit determines that the third pattern matching is established,
The predetermined value is larger than the first threshold value, and the third threshold value is equal to or more than the first threshold value and smaller than the predetermined value. Further provided with a validity determination unit,
When at least one of the first condition and the second condition is satisfied, the validity determining unit determines that the virtual detection point is valid,
The second matching unit performs the third pattern matching based on one of the virtual detection points generated by the virtual detection point generation unit that is determined to be valid by the validity determination unit,
The first condition is a condition that the azimuth of the virtual detection point with respect to the radar device is outside a first predetermined range,
The second condition is that the incident angle of the transmission wave is outside the second predetermined range when it is assumed that the transmission wave of the radar device is incident on the position of the virtual detection point on the surface of the template. The radar device according to claim 1, wherein Even if neither the first condition nor the second condition is satisfied, if the third condition is satisfied, the validity determination unit determines that the virtual detection point is valid,
The radar according to claim 2, wherein the third condition is a condition that at least one of the detection point and the effective virtual detection point exists between the radar device and the virtual detection point that is a determination target. apparatus. Even if neither the first condition nor the second condition is satisfied, if the third condition is satisfied, the validity determination unit determines that the virtual detection point is valid,
The third condition is a condition that a line segment connecting the radar device and the virtual detection point that is the determination target passes through a surface on which the virtual detection point that is the determination target of the template does not exist. The radar device according to 1.   The radar device according to claim 2, wherein the validity determination unit determines whether or not the virtual detection point is valid based on time series information of the virtual detection point.   The radar device according to claim 5, wherein it is determined whether or not the virtual detection point is valid based on each position of the radar device corresponding to each of the time series information.   When there are a plurality of types of templates in which the matching rate of the first pattern matching is greater than a second threshold and is equal to or less than the first threshold, the virtual detection point generation unit determines the virtual detection points for each of the plurality of types of templates. The radar device according to claim 1, which is generated. A signal processing method for a radar device, comprising:
A detection step of detecting a detection point that is a reflection point of the object based on a reception signal output from a reception antenna that receives a reflected wave from the object,
A first matching step of performing a first pattern matching based on the detection points and a template of a predetermined shape to calculate a matching rate of the first pattern matching;
A virtual detection point and a virtual detection point generation step of generating the virtual detection point such that the matching rate when the second pattern matching is performed based on the detection point and the template becomes a predetermined value,
A second matching step of performing a third pattern matching based on at least a part of the virtual detection points generated by the virtual detection point generation step and the detection points and the template to calculate a matching rate of the third pattern matching. When,
A matching determination step,
When the matching rate of the first pattern matching is larger than a first threshold value, it is determined that the first pattern matching is established in the matching determination step,
When the matching rate of the first pattern matching is greater than a second threshold value and is equal to or less than the first threshold value, the virtual detection point is generated in the virtual detection point generation step,
When the matching rate of the third pattern matching is larger than a third threshold value, it is determined that the third pattern matching is established in the matching determination step,
The signal processing method, wherein the predetermined value is larger than the first threshold value, and the third threshold value is greater than or equal to the first threshold value and less than the predetermined value.

Images