Radar device and method for detecting an object using a radar device

Abstract

Radar device (1) comprising: a first measuring unit (4: S120) configured to calculate positions and relative velocities of reflection points of the frequency-modulated continuous waves from a beat signal obtained by transmitting and receiving frequency-modulated continuous waves; a first extraction unit (4: S130) configured to extract stationary reflection points from a measurement result obtained by the first measurement unit (4: S120), wherein the stationary reflection points are each a reflection point with a relative velocity of zero; a region setting unit (4: S150) configured to set, for each of the stationary reflection points extracted by the first extraction unit (4: S130), an object region estimated as an object including the stationary reflection point; a removing unit (4: S170) configured to remove a DC component from a beat signal obtained by transmitting and receiving multi-frequency continuous waves; a second measuring unit (4: S190) configured to calculate positions and relative velocities of reflection points of the multi-frequency continuous waves from the beat signal from which the DC component has been removed by the removal unit (4: S170); a second extraction unit (4: S190) configured to extract in-area reflection points from a measurement result obtained by the second measurement unit (4: S190), wherein each of the in-area reflection points is a reflection point included in the object area set by the area setting unit (4: S150); a calculation unit (4: S200) configured to calculate, for each of the in-area reflection points extracted by the second extraction unit (4: S190), a lateral velocity that is a velocity of the in-area reflection point in a transverse direction of the area; and an estimation unit (4: S210) configured to statistically process lateral velocities calculated by the calculation unit (4: S200) for each of the object areas to calculate an estimated value of a lateral velocity of an object located in the object area.

Description

[Technical field]

[0001] The invention relates to a radar device that combines different measurement methods. [State of the art]

[0002] For a short-range measurement, a method with high range resolution is used. For a long-range measurement, a method with low processing load is used. Some radars are known to combine such different measurement methods. For example, JP 2008 - 249 399 A proposes the following measurement method: A broadband radar (e.g., a UWB method) is used for the short-range measurement, and a narrowband radar (e.g., an FMCW method) is used for the long-range measurement. Note that UWB stands for ultra-wideband and FMCW stands for frequency-modulated continuous wave.

[0003] DE 10 2012 024 998 A1 discloses a method for determining the lateral velocity of a target object relative to a motor vehicle. In particular, in the near range, the lateral object velocity is determined as a function of at least two values ​​of the radial velocity and the two corresponding values ​​of the target angle.

[0004] US 8 334 802 B2 discloses a radar system having a first and a second radar, each of which can be operated in two different modes, an FMCW mode and a CW mode, wherein the control is carried out in such a way that mutual interference is suppressed. [Brief description of the invention][Problem to be solved by the invention]

[0005] However, conventional UWB-based technology (the technology described in JP 2008-249399 A) can interfere with other wireless systems. Conventional UWB-based technology also requires hardware that supports high-speed operation for transmitting and receiving extremely short pulses.

[0006] Instead of the UWB method, a multi-frequency CW method can be used for near-field measurement. However, in the multi-frequency CW method, in a case where there is a reflection point of waves reflected toward the incident direction (hereinafter referred to as a "stationary" reflection point), the signal intensity of the reflected waves from the stationary reflection point is very high. As a result, reflected waves from other reflection points located near the stationary reflection point (hereinafter referred to as "nearby reflection points") are buried. Therefore, information about nearby reflection points cannot be used as a measurement result. In addition, the relative velocity detected at the stationary reflection point is zero. Therefore, the movement of the object with the stationary reflection point cannot be accurately detected.

[0007] The invention provides a technique for accurately detecting the movement of an object located within a short distance without using a broadband radar. [Solution to the task]

[0008] A radar device according to one aspect of the inventive technique includes a first measuring unit (4: S120), a first extraction unit (4: S130), a range setting unit (4: S150), a distance unit (4: S170), a second measuring unit (4: S190), a second extraction unit (4: S190), a calculation unit (4: S200), and an estimation unit (4: S210). The first measuring unit calculates positions and relative velocities of reflection points of the FMCW from a beat signal obtained by transmitting and receiving frequency-modulated continuous waves (hereinafter referred to as "FMCW"). The first extraction unit extracts stationary reflection points from a measurement result obtained by the first measuring unit, each of the stationary reflection points being a reflection point with a relative velocity of zero.The range setting unit sets an object range for each of the stationary reflection points extracted by the first extraction unit, which is estimated to include the stationary reflection point. The ranging unit removes a DC component from a beat signal generated by transmitting and receiving multi-frequency continuous waves (hereinafter referred to as "multi-frequency CW"). The second measuring unit calculates positions and relative velocities of reflection points of the multi-frequency CW from the beat signal from which the DC component has been removed by the ranging unit. The second extraction unit extracts in-range reflection points from a measurement result obtained by the second measuring unit, each of the in-range reflection points being a reflection point included in the object range set by the range setting unit.The calculation unit calculates a lateral velocity for each of the in-area reflection points extracted by the second extraction unit, which is a velocity of the in-area reflection point in a transverse direction of the area. The estimation unit statistically processes lateral velocities calculated by the calculation unit for each of the object areas to calculate an estimated value of a lateral velocity of an object located in the object area. Accordingly, the estimation unit calculates an estimated value of a lateral velocity of an object located in the object area.

[0009] According to such a configuration, the radar device according to the invention can accurately detect the movement of a short-range object in the range direction and the lateral direction based on the results of FMCW-based and multi-frequency CW-based measurements. In short, the radar device according to the invention can accurately detect the movement of an object located within a short distance without using a broadband radar.

[0010] Note that reference numerals in parentheses described in this section and in the claims indicate exemplary correspondence relationships with components described in the embodiment discussed later as an aspect of the invention. Therefore, these reference numerals do not limit the technical scope of the invention. [Brief description of the drawings] Fig. 1 is a block diagram illustrating a configuration of a radar device. Fig. 2 is a flowchart for a near detection process. Fig. 3 is an explanatory diagram concerning a stationary reflection point and neighboring reflection points. Fig. Figure 4 is a diagram illustrating the influence of a stationary reflection point on neighboring reflection points in a 2CW-based beat signal. Fig. Figure 5 is a diagram illustrating an example of filtering. Fig. Figure 6 is a diagram illustrating a situation where the near detection process is used. [Description of the embodiment]

[0011] An embodiment of the invention will be described below with reference to the drawings. [1. Configuration]

[0012] Radar devices 1 are installed at the two widthwise ends of a bumper provided at each of the front and rear ends of a vehicle. Each of the two radar devices 1 on the front bumper is arranged in a direction that allows the area in front of and on the right or left side of the vehicle to be included in the probe area. Each of the two radar devices 1 on the rear bumper is arranged in a direction that allows the area behind and on the right or left side of the vehicle to be included in the probe area. These four radar devices 1 have similar configurations. Fig. For example, Fig. 3 illustrates an example of the radar device 1 installed at the right rear end of the vehicle.

[0013] As in Fig. 1, the radar device 1 includes an antenna unit 2, a transceiver or transmitting-receiving unit 3 and a signal processing unit 4. The radar device 1 is communicatively connected to other vehicle-internal devices, for example via an in-vehicle local area network (not shown).

[0014] The antenna unit 2 includes, for example, a plurality of antennas aligned horizontally. The antenna unit 2 transmits and receives electromagnetic millimeter waves as radar waves.

[0015] The transmitting / receiving unit 3 alternately transmits and receives radar waves via the antenna unit 2 at predetermined time intervals. At this time, the transmitting / receiving unit 3 transmits and receives frequency-modulated continuous waves (FMCW) or dual-frequency continuous waves (hereinafter referred to as "2CW") as radar waves. In other words, the transmitting / receiving unit 3 uses a narrow-band radar with bandwidths of hundreds of megahertz or less, rather than a wide-band radar such as UWB. The transmitting / receiving unit 3 further generates a beat signal for each reception signal received by the plurality of antennas provided in the antenna unit 2. Note that a beat signal as used herein includes the frequency component of the difference between a reception signal and a transmission signal.The transmitting-receiving unit 3 performs an A / D conversion of a beat signal into received data and outputs the received data to the signal processing unit 4.

[0016] The signal processing unit 4 essentially includes a known microcomputer having a semiconductor memory (hereinafter referred to as the "memory 42") and a CPU 41. Examples of the memory 42 include a RAM, a ROM, and a flash memory. The CPU 41 reads and executes a program stored on a non-volatile, tangible, computer-readable storage medium, thereby implementing various functions of the signal processing unit 4. In the present embodiment, the memory 42 corresponds to a non-volatile, tangible, computer-readable storage medium with a program stored therein. In the present embodiment, after executing a program, the process for implementing various functions (the method corresponding to the program) is executed. Note that one or more microcomputers may constitute the signal processing unit 4.

[0017] The CPU 41 of the signal processing unit 4 executes a program. Consequently, the signal processing unit 4 executes at least one proximity detection method for detecting an object located within a short distance based on beat signals received from the transmitting / receiving unit 3. Note that the manner in which the proximity detection function of the signal processing unit 4 is implemented is not limited to software such as a program. For example, some or all of the proximity detection elements may be implemented using one or more hardware components. In a case where the proximity detection function is implemented using hardware such as electronic circuits, digital circuits including a plurality of logic circuits or analog circuits may be used. Alternatively, these circuits may be combined to implement the proximity detection function. [2nd trial]

[0018] Next, the proximity detection process performed by the signal processing unit 4 is explained using the flowchart in Fig. 2. It is noted that this process is initiated and executed repeatedly in a given measurement cycle (e.g., 80 ms).

[0019] Once this process is initiated, the signal processing unit 4 detects a beat signal generated by transmitting and receiving FMCW from the transmitting-receiving unit 3 (S110).

[0020] The signal processing unit 4 performs frequency analysis of the beat signal acquired in step S110 to extract reflection points of FMCW (S120). Note that any known FMCW radar method can be used to extract reflection points. At this time, the signal processing unit 4 obtains the following information about reflection points. More specifically, the signal processing unit 4 calculates the relative speed Vr, the distance R, and the direction φ of a reflection point with respect to the vehicle equipped with the radar device 1 (hereinafter referred to as the "host vehicle"). Note that the relative speed Vr as used herein is the relative speed in a range direction, that is, the direction along the straight line connecting the host vehicle and a reflection point.In the following, the relative velocity in the range direction is referred to as range velocity dR. Fig. Figure 3 illustrates the ranging velocities dR of the respective reflection points on the same object moving in the width direction of the host vehicle. As shown in the figure, the ranging velocities dR vary in accordance with the positional relationship between the radar device 1 and the reflection points, even though the reflection points belong to the same object moving in the width direction of the host vehicle. More specifically, the ranging velocity dR is zero (dR = 0) at a position A where the straight line indicating the ranging direction is orthogonal to the reflection surface. The ranging velocity dR has a positive value (dR > 0) at any position between a position B and a position A. The ranging velocity dR has a negative value (dR < 0) at any position between a position C and a position A.

[0021] The signal processing unit 4 extracts, as a stationary reflection point, the reflection point with a range velocity dR of zero from the reflection points extracted in step S120 (S130). Note that a counter is prepared for counting the number of continuously measured stationary reflection points. The count value of the counter is set as follows. For example, assume that the stationary reflection points measured in the previous measurement cycle and those in the current measurement cycle are at the same position (in the same direction and within the same distance). In this case, the count value for the stationary reflection point measured in the previous measurement cycle is incremented (+1). Conversely, assume that the stationary reflection points measured in the previous measurement cycle and the current measurement cycle are not at the same location.In this case, the count value for the stationary reflection point measured in the previous measurement cycle is set to the initial value (1).

[0022] The signal processing unit 4 determines whether any of the extracted stationary reflection points satisfies a predetermined detection condition (S140). Specifically, the present embodiment uses the following detection condition: the distance from the host vehicle to a reflection point is equal to or less than a predetermined distance (upper limit distance), and the count value indicating the number of times the reflection point has been measured is equal to or greater than a predetermined value N (for example, N = 4). Consequently, if the signal processing unit 4 determines that none of the stationary reflection points satisfies the detection condition (S140: NO), this process is temporarily terminated. On the other hand, if the signal processing unit 4 determines that at least one of the stationary reflection points satisfies the detection condition (S140: YES), step S150 is executed.

[0023] The signal processing unit 4 sets an object range for each of the stationary reflection points that satisfied the detection condition in step S140 (S150). An object range with a stationary reflection point is set taking into account the length of a typical vehicle. Note that an object range is a range of directional angles estimated to include an object with a stationary reflection point.

[0024] The signal processing unit 4 acquires a beat signal obtained by transmitting and receiving 2CW from the transmitting / receiving unit 3 (S160). The beat signal acquired in this step is a signal obtained at the time considered to be the same as the time of acquisition of the FMCW beat signal in step S110.

[0025] The signal processing unit 4 removes a DC signal component from the beat signal acquired in step S160 (S170). Specifically, the signal processing unit 4 removes the signal component by which the stationary reflection point acquired in step S130 influences the measurement result of nearby reflection points. The signal intensity of reflected waves from the stationary reflection point is very high. Therefore, signal components based on reflected waves from nearby reflection points near the stationary reflection point are buried in the signal component based on reflected waves from the stationary reflection point.In this context, the signal processing unit 4 removes the signal component that causes signal components based on reflected waves from nearby reflection points to be buried in the signal component based on reflected waves from the stationary reflection point. Accordingly, as shown in FIG. Fig. 4, the radar device 1 according to the present embodiment can extract signal components based on reflected waves from nearby reflection points in the vicinity of the stationary reflection point.

[0026] The signal processing unit 4 selects one of the object areas determined in step S150 as a target area (S180).

[0027] The signal processing unit 4 performs frequency analysis of the beat signal from which the DC component was removed in step S170 to extract reflection points of 2CW. Consequently, the signal processing unit 4 extracts in-range reflection points, i.e., reflection points located within the target range selected in step S180 (S190). Note that any known 2CW radar method can be used to extract reflection points. At this time, the signal processing unit 4 obtains the following information about in-range reflection points. More specifically, the signal processing unit 4 calculates the range velocity dR (relative velocity in the range direction) and a direction θ of a reflection point with respect to the host vehicle.It is noted that the direction θ as used herein is the angle formed with the direction to the detected stationary reflection point in the selected target area (see . Fig. 3).

[0028] The signal processing unit 4 calculates the velocity Vc (hereinafter referred to as the “lateral velocity Vc”) of an in-area reflection point in a transverse direction based on the result of the calculation in step S190. The transverse area direction as used herein is the direction orthogonal to the straight line indicating the range direction of the stationary reflection point in the target area (see Fig. 3). The transverse range direction is the direction along the reflection surface of the FMCW or 2CW. The signal processing unit 4 calculates the transverse velocity Vc of each of the in-range reflection points using a formula (1) based on the range velocity dR and the direction θ. Vc=dR / cosθ

[0029] The signal processing unit 4 filters the lateral velocities Vc of the respective in-area reflection points calculated in step S200. Accordingly, the signal processing unit 4 estimates the lateral velocity Vo (hereinafter referred to as the “estimated lateral velocity Vo”) (S210). More specifically, the signal processing unit 4 calculates the mean or median of the lateral velocities Vc of all the in-area reflection points by filtering (i.e., statistically processes the lateral velocities Vc of all the in-area reflection points). The signal processing unit 4 sets the calculation result as the estimated lateral velocity Vo. It is noted that the method for calculating the estimated lateral velocity Vo is not limited to this method. As shown in Fig. For example, as shown in Figure 5, the signal processing unit 4 calculates the probability density function of the distribution of lateral velocities Vc using a kernel function. The signal processing unit 4 then sets the mode of the probability density function as the estimated lateral velocity Vo.

[0030] The signal processing unit 4 determines whether all the object regions specified in step S150 have been selected as target regions (S220). Consequently, if the signal processing unit 4 determines that not all the object regions have been selected as target regions (S220: NO), the process returns to step S180. Subsequently, steps S180 to S210 are repeated. In other words, the signal processing unit 4 calculates the estimated lateral velocity Vo for each of the object regions. Conversely, if the signal processing unit 4 determines that all the object regions have been selected as target regions (S220: YES), step S230 is executed.

[0031] The signal processing unit 4 outputs the estimated lateral velocity Vo calculated for each of the object regions as the estimated velocity in the region transverse direction of the object located within the object region (S230). Then, this process is terminated. [3. Effects]

[0032] According to the first embodiment described in detail above, the following effects can be obtained. (3a) The radar device 1 processes a beat signal obtained using the FMCW method. Accordingly, the existence direction of the stationary reflection point is determined. The radar device 1 further removes a DC component from a beat signal obtained using the 2CW method with respect to the specified direction. More specifically, the radar device 1 removes the signal component based on reflected waves from the stationary reflection point. Accordingly, nearby reflection points (a plurality of reflection points) located near the stationary reflection point are extracted. The range velocity dR of a nearby reflection point is not zero. Therefore, the lateral velocity Vc can be calculated based on the range velocity dR.In this way, the radar device 1 can accurately detect the movement of an object located within a short distance in the cross-range direction without using a broadband radar. (3b) The radar device 1 statistically processes the lateral velocities Vc calculated from a plurality of nearby reflection points. Accordingly, the lateral velocity Vc of the object is estimated using the nearby or adjacent reflection points. In this way, the radar device 1 can perform the near-object detection process using signal components (low-intensity signal components) based on reflected waves from nearby reflection points. Furthermore, the radar device 1 can prevent misdetection of the movement of an object even when the detection process is performed using low-intensity signal components. (3c) For example, the radar device 1 can detect the movement of the vehicle in the Fig. 6 accurately capture the situation presented. Fig. Figure 6 illustrates a situation in which the host vehicle is about to reverse and leave the parking space in a parking lot. In this situation, the other vehicle is about to cross behind the host vehicle. In such a situation, the radar device 1 can accurately detect the speed of the other vehicle in its direction of travel. [4. Other embodiments]

[0033] Although an embodiment of the invention has been described so far, the inventive technique is not limited to the above embodiment, but can be changed in various ways for implementation. (4a) According to the inventive technique, a plurality of functions of a single component in the above embodiment can be implemented by a plurality of components. A single function of a single component can be implemented by a plurality of components. A plurality of functions of a plurality of components can be implemented by a single component. A single function implemented by a plurality of components can be implemented by a single component. A part of the configuration of the above embodiment can be omitted. At least a part of a configuration of the above embodiment can be added to or replaced by another configuration of the above embodiment.It is noted that the embodiment of the invention includes every aspect contained in the technical idea specified by the terms described in the claims. (4b) In addition to the radar device 1 mentioned above, the technique of the invention can be provided in various forms including the following examples. More specifically, the technique of the invention can be provided by a system including the radar device 1 as a component. The technique of the invention can also be provided by a program for causing a computer to operate as the radar device 1. The technique of the invention can also be provided by a non-transitory tangible computer-readable storage medium having the program recorded therein, such as a semiconductor memory. The technique of the invention can further be provided by a method for detecting the moving speed in the lateral direction. [Reference character list] 1 radar device (in-vehicle radar device) 2 antenna unit 3 Transmitter-receiver unit 4 Signal processing unit 41 CPU 42 storage

Claims

[1] Radar device (1) comprising: a first measuring unit (4: S120) configured to calculate positions and relative velocities of reflection points of the frequency-modulated continuous waves from a beat signal obtained by transmitting and receiving frequency-modulated continuous waves; a first extraction unit (4: S130) configured to extract stationary reflection points from a measurement result obtained by the first measurement unit (4: S120), wherein the stationary reflection points are each a reflection point with a relative velocity of zero; a region setting unit (4: S150) configured to set, for each of the stationary reflection points extracted by the first extraction unit (4: S130), an object region estimated as an object including the stationary reflection point; a removing unit (4: S170) configured to remove a DC component from a beat signal obtained by transmitting and receiving multi-frequency continuous waves; a second measuring unit (4: S190) configured to calculate positions and relative velocities of reflection points of the multi-frequency continuous waves from the beat signal from which the DC component has been removed by the removal unit (4: S170); a second extraction unit (4: S190) configured to extract in-area reflection points from a measurement result obtained by the second measurement unit (4: S190), wherein each of the in-area reflection points is a reflection point included in the object area set by the area setting unit (4: S150); a calculation unit (4: S200) configured to calculate, for each of the in-area reflection points extracted by the second extraction unit (4: S190), a lateral velocity that is a velocity of the in-area reflection point in a transverse direction of the area; and an estimation unit (4: S210) configured to statistically process lateral velocities calculated by the calculation unit (4: S200) for each of the object areas to calculate an estimated value of a lateral velocity of an object located in the object area. [2] The radar device (1) according to claim 1, wherein the area setting unit (4: S150) sets the object area for each of the stationary reflection points extracted by the first extraction unit (4: S130) and continuously measured at the same position at a predetermined number of times or more. [3] The radar device (1) according to claim 1 or 2, wherein the area setting unit (4: S150) sets the object area for each of the stationary reflection points extracted by the first extraction unit (4: S130) and located within a predetermined upper limit distance. [4] The radar device (1) according to any one of claims 1 to 3, wherein the estimation unit calculates a probability density function using a kernel function and sets a mode of the probability density function as the estimated value. [5] A method for detecting an object using a radar device (1), the method comprising: a first measuring step (4: S120) for calculating, from a beat signal obtained by transmitting and receiving frequency-modulated continuous waves, positions and relative velocities of reflection points of the frequency-modulated continuous waves; a first extraction step (4: S130) for extracting stationary reflection points from a measurement result obtained in the first measurement step (4: S120), wherein the stationary reflection points are each a reflection point with a relative velocity of zero; a region setting step (4: S150) for setting, for each of the stationary reflection points extracted in the first extraction step (4: S130), an object region estimated as an object including the stationary reflection point; a removing step (4: S170) for removing a DC component from a beat signal obtained by transmitting and receiving multi-frequency continuous waves; a second measuring step (4: S190) for calculating, from the beat signal from which the DC component was removed in the removal step (4: S170), positions and relative velocities of reflection points of the multi-frequency continuous waves; a second extraction step (4: S190) for extracting in-area reflection points from a measurement result obtained in the second measurement step (4: S190), each of the reflection points being an in-area reflection point included in the object area specified in the area setting step (4: S150); a calculation step (4: S200) for calculating, for each of the in-area reflection points extracted in the second extraction step (4: S190), a lateral velocity which is a velocity of the in-area reflection point in a lateral range direction; and an estimation step (4: S210) for statistically processing lateral velocities calculated in the calculation step (4: S200) for each of the object areas to calculate an estimated value for a lateral velocity of an object located in the object area.

Images