JP7002860B2 - Directional error detection method and equipment - Google Patents

Description

The present disclosure relates to a technique for detecting an orientation error of an in-vehicle radar device.

It is known that when an in-vehicle radar device is installed in a bumper, the detection performance of the direction in which an object exists deteriorates due to the disturbance of the radiation characteristics of the antenna caused by the multiple reflection of the transmitted radar wave in the pumper. ..

On the other hand, in Patent Document 1, the relative velocity of the stationary object reflecting the radar wave and the distribution of the arrival direction of the reflected wave are obtained, and the relationship between the relative velocity and the orientation of the own vehicle and the stationary object is obtained. A technique for obtaining the orientation error by comparing with the represented theoretical curve is disclosed.

Japanese Unexamined Patent Publication No. 2016-121899

In the prior art described in Patent Document 1, the error between the collected data and the theoretical curve is obtained for each frequency bin associated with the direction. For this reason, it is necessary to collect data for all frequency bins to the extent that significant results can be obtained when statistical processing is performed. Therefore, it takes a lot of time for the required accuracy error to be obtained in all directions. On the other hand, the inventor separates the observed values into the actual azimuth and the directional error using the most probable method, and from the distribution of points representing the relationship between the separated actual azimuth and the directional error, the relationship between the two is determined. By finding the regression function that represents it, I thought about finding the directional error over the entire directional range in a short time.

However, as a result of detailed examination by the inventor, if the detection range of the radar includes a direction along the straight direction of the own vehicle, that is, a direction directly in front or a direction directly behind, it is not possible to estimate an error in these directions. I found a problem. That is, for an object located in these directions, the lower limit of the variance of the error, which is one of the parameters used in the maximum likelihood method, becomes infinite, which makes it impossible to calculate.

The present disclosure provides a technique for obtaining a directional error in the front or rear front direction along a straight direction of a vehicle equipped with an in-vehicle radar device.

The orientation error detection method according to one aspect of the present disclosure includes a first step, a second step, a third step, and a fourth step.
In the first step, the reflection points observed by the FMCW radar mounted on the vehicle are acquired so that the detection range includes the front direction of the front or the rear along the straight direction of the vehicle, and among the reflection points. Extract the stationary points that are presumed to be stationary.

In the second step, the front direction obtained from the installation state of the FMCW radar on the vehicle is set as the reference direction, and among the stationary points extracted in the first step, the position is within the front range set around the reference direction. Extract the front point to be used.

In the third step, among the front points extracted in the second step, the high points estimated to exist at positions higher than the preset high thresholds are extracted.
In the fourth step, the front direction obtained from the distribution of the high points extracted in the third step is used as the measurement direction, and the deviation of the measurement direction with respect to the reference direction is obtained as the orientation error in the front direction.

According to such a method, the reflection point on the superstructure such as the signboard installed above the road, that is, the reflection point on the object existing in the front direction of the vehicle is extracted as the high point. Therefore, the directional error in the front direction can be accurately obtained in a short time based on the extracted high-level points.

The orientation error detection device according to another aspect of the present disclosure includes a static extraction unit, a front extraction unit, a high-level extraction unit, and an error calculation unit. The static extraction unit realizes the processing of the first step described above, the front extraction unit realizes the processing of the second step described above, and the high-level extraction unit realizes the processing of the third step described above. This is realized, and the error calculation unit realizes the processing of the fourth step described above.

According to such a configuration, it is possible to obtain the same effect as the effect obtained by implementing the above-mentioned directional error detection method.
In addition, the reference numerals in parentheses described in this column and the scope of claims indicate the correspondence with the specific means described in the embodiment described later as one embodiment, and the technical scope of the present disclosure is defined. It is not limited.

It is a block diagram which shows the structure of an in-vehicle radar apparatus. It is explanatory drawing about the mounting state of an in-vehicle radar and an observation model. It is a flowchart of the main process. It is a flowchart of the 2nd learning process. It is explanatory drawing which shows the extraction method of a stationary point. It is explanatory drawing which shows the extraction method of the front point. It is a graph which shows the theoretical value and the measured value of the height pattern obtained by tracking a high point. It is a graph which shows the theoretical value and the measured value of the height pattern obtained by tracking a non-high point. It is a graph which shows the distribution map of high-ranking points, and the cumulative distribution map based on the distribution map.

Hereinafter, embodiments of the present disclosure will be described with reference to the drawings.
[1. Constitution]
The in-vehicle radar device 1 is installed and used in a bumper made of a material that transmits electromagnetic waves. Specifically, as shown in FIG. 2, the in-vehicle radar device 1 is installed on a bumper installed on the rear side of the vehicle near the right end in the traveling direction of the vehicle. Further, the in-vehicle radar device 1 is installed in a state where the central axis direction of the detection range in the horizontal plane is tilted by a predetermined angle φinst with respect to the backward direction (hereinafter referred to as the front direction) along the straight direction of the vehicle. Hereinafter, based on this installation state, the front direction detected by the in-vehicle radar device 1 is referred to as a reference direction, and the angle in the reference direction expressed with reference to the optical axis direction of the in-vehicle radar device 1 is referred to as a reference angle −φinst.

As shown in FIG. 1, the in-vehicle radar device 1 includes an antenna unit 2, a transmission / reception unit 3, and a signal processing unit 4.
The antenna unit 2 includes a plurality of antennas arranged in a row in the horizontal direction, and transmits / receives electromagnetic waves as radar waves.

The transmission / reception unit 3 periodically transmits / receives a frequency-modulated continuous wave (hereinafter referred to as FMCW) as a radar wave via the antenna unit 2 at regular time intervals. That is, the FMCW radar is used as the in-vehicle radar device 1. The transmission / reception unit 3 generates a beat signal composed of a frequency component of the difference between the reception signal and the transmission signal for each reception signal received by each antenna constituting the antenna unit 2. The transmission / reception unit 3 A / D-converts the beat signal, and supplies the received data obtained as a result of the A / D conversion to the signal processing unit 4. The FMCW uses millimeter waves whose center frequency is on the order of GHz, and the modulation width is set to about several tens of MHz.

The signal processing unit 4 is mainly composed of a well-known microcomputer having a CPU 41 and a semiconductor memory (hereinafter, memory 42) such as RAM, ROM, and flash memory. Further, the signal processing unit 4 is communicably connected to another vehicle-mounted device via an vehicle-mounted local area network (hereinafter, vehicle-mounted LAN) (not shown).

Various functions of the signal processing unit 4 are realized by the CPU 41 executing a program stored in a non-transitional substantive recording medium. In this example, the memory 42 corresponds to a non-transitional substantive recording medium in which a program is stored. In addition, when this program is executed, the method corresponding to the program is executed. The number of microcomputers constituting the signal processing unit 4 may be one or a plurality.

The signal processing unit 4 detects an object that reflects radar waves by executing a program by the CPU 41, and at least executes a main process of generating information about the object. The method for realizing the function of the signal processing unit 4 is not limited to software, and a part or all of the elements may be realized by using one or a plurality of hardware. For example, when the above function is realized by an electronic circuit which is hardware, the electronic circuit may be realized by a digital circuit including a large number of logic circuits, an analog circuit, or a combination thereof.

A part of the memory 42 is composed of a non-volatile memory in which the contents of the memory are retained even when the power of the vehicle-mounted radar device 1 is turned off. In this non-volatile memory, an orientation correction table showing the correspondence between the measurement orientation, which is the orientation of the object obtained by frequency analysis of the beat signal, and the orientation error in the measurement orientation is stored.

[2. process]
[2-1. Main processing]
Next, the main processing executed by the signal processing unit 4 will be described with reference to the flowchart of FIG. This process is activated every measurement cycle in which radar waves are transmitted and received.

When this processing is activated, the signal processing unit 4 acquires sampling data of the beat signal obtained by the transmission / reception unit 3 transmitting / receiving radar waves in S110.
In S120, the frequency spectrum is calculated for each of the uplink and downlink modulations of the FMCW and for each antenna constituting the antenna unit 2 by frequency analysis of the sampling data. Here, a fast Fourier transform is used as the frequency analysis.

In S130, the average frequency spectrum is calculated for each of the uplink modulation and the downlink modulation based on the frequency spectrum obtained in S120.
In S140, a frequency bin in which a peak value at which the reception intensity is equal to or higher than a preset threshold value is detected is extracted from each average frequency spectrum, and direction estimation processing is executed for each frequency bin. High-resolution estimation processing such as MUSIC is desirable for the orientation estimation processing, but beamforming or the like may be used.

In S150, the direction estimated in S140 (hereinafter referred to as the estimated direction) is corrected by using the direction correction table stored in the memory 42. Specifically, the directional error is acquired from the directional correction table using the estimated directional index as an index, and the acquired directional error is added to the estimated directional error to correct the directional error. In the following, the correction result is referred to as the detection direction.

In S160, object information is generated for each object that has a peak in the average frequency spectrum. The object information includes the corrected direction obtained in S150, the relative distance to the object, and the relative velocity. The relative distance and relative velocity with respect to the object are obtained from the peaks detected from each average frequency spectrum by using a known method in the FMCW radar. The generated object information is provided to each in-vehicle device that uses the object information via the in-vehicle LAN.

In S170, the first learning process of learning the directional error other than the front direction and updating the directional correction table is executed by using the processing results of S140 and S160.
In S180, the second learning process of learning the directional error in the front direction and updating the directional correction table is executed using the processing results of the previous S140 and S160, and this process is terminated.

[2-2. Learning process]
The first learning process executed in the above S170 will be described.
In the first learning process, the observed value of the direction obtained in S140 is separated into the actual direction and the direction error by using the maximum likelihood method, and the distribution of points representing the relationship between the separated actual direction and the direction error is used. Find the regression function that expresses the relationship between the two .

However, in the vehicle-mounted radar device 1, the radar detection range includes the front direction. In this front direction, the lower limit of the variance of the error, which is one of the parameters used in the maximum likelihood method, becomes infinite, and the calculation by the maximum likelihood method becomes impossible, so that the orientation error in the front direction cannot be obtained. Therefore, in the first learning process, the direction error is learned for the direction other than the front direction, and the direction correction table is updated for the direction other than the front direction.

Next, the second learning process executed in S180 will be described with reference to the flowchart of FIG.
When this processing is activated, the signal processing unit 4 determines in S210 whether or not the preset straight-ahead condition is satisfied. If the straight-ahead condition is not satisfied, this process is terminated, and if the straight-ahead condition is satisfied, the process proceeds to S220. The straight-ahead condition is a condition for determining whether or not the own vehicle is going straight. Specifically, from information such as the steering angle that indicates the behavior of the own vehicle, it can be considered that the state that can be regarded as going straight has continued for a predetermined time or more, and that the road shape of the traveling path within the predetermined range before and after the own vehicle is a straight line. This is used as a straight-ahead condition. The detected value of the steering angle to be determined and the map information representing the road shape are acquired, for example, via the in-vehicle LAN.

In S220, a rest point which is a reflection point related to a stationary object is extracted based on the information about each reflection point reflecting the radar wave detected in S130 and S140. Specifically, the reflection points distributed within the predetermined velocity range shown in FIG. 5 (for example, the solid line ± 5 m / s in the graph) are extracted as static points. The solid line in the graph of FIG. 5 is a graph of the theoretical value of the relationship between the apparent relative velocity of the stationary object observed under ideal conditions without noise and the direction φ of the stationary point.

In S230, among the stationary points extracted in S220, the front points estimated to be in the front direction are extracted. Specifically, as shown in FIG. 6, a predetermined angle range centered on the reference angle −φinst is set as the front range, and a stationary point existing in the front range is set as the front point. The right figure in the figure shows the relationship between the position sinφ of the stationary point and the distance R to the stationary point, and the left figure in the figure shows the relationship between the reflection intensity P of the stationary point and the distance R to the stationary point. In the figure, only the front point when −φinst ± 10 ° is the front range is shown. In addition, the parabolic graph shown in the figure on the right shows the theoretical value of the reflection point on the wall-shaped roadside object when there is a wall-shaped road survey object such as a guardrail or noise barrier on the side of the road. show. As shown in the figure, it can be seen that the reflection point on the wall-shaped roadside object needs to be removed because it becomes noise when the front direction is obtained.

In S240, by tracking each of the front points extracted in S230, a height pattern showing the relationship between the distance R from the own vehicle to the front point and the reflection intensity P from the front point is generated for each front point. create.

In S250, it is determined whether or not the front point is a high point existing at a position higher than the preset high threshold value according to the height pattern generated in S240. Here, the high threshold value is set to a height equal to or higher than the lower limit of the height of the upper structure (for example, a signboard or the like) installed above the road. When it is determined that the front point is a high point, the information of the high point is stored in the memory.

As for the height pattern, the higher the height H of the reflection point from the road surface, the narrower the interval between the null points, and the lower the height H of the reflection point from the road surface, as shown in FIG. In addition, the interval between null points becomes narrower. The null point is generated by the interference between the direct wave from the reflection point and the indirect wave that is reflected from the reflection point to the road surface and arrives. However, the case where the height H of the reflection point from the road surface is H = 6 m in FIG. 7 and the case where H = 0.5 m is shown in FIG. The figure shows an example of the theoretical value and the measured value of the height pattern.

Then, specifically, when the height pattern satisfies the high-level condition, all the front points belonging to the height pattern are determined to be high-level points. As the high-level condition, for example, in a distance region of the lower limit distance or more, a height pattern over a predetermined distance is required, and the degree of fluctuation of the reflection intensity in the height pattern is larger than the fluctuation threshold value. Here, the lower limit distance is based on, for example, an ideal height pattern required for a stationary object having a height that the vehicle can overcome, and the farthest null point within the detection range of the FMCW radar existing in this height pattern. It is set to a value larger than the distance to. If the setting is based on FIG. 8, the lower limit distance may be, for example, about 45 m. Further, the predetermined distance may be, for example, 50 m or more. The degree of variation in the reflection intensity may be a parameter that can express the difference in smoothness of the graphs showing the actually measured values shown in FIGS. 7 and 8. For example, a distance range equal to or larger than the lower limit distance may be divided into small sections (for example, in units of 10 m), and the average value of the difference between the maximum value and the minimum value obtained for each small section over all the small sections may be used.

In S260, it is determined whether or not the calculation condition for obtaining the directional error is satisfied. The calculation condition may be determined based on whether or not the elapsed time, the mileage, the number of detected high points, etc. measured or detected starting from the previous calculation exceeds a preset threshold value. .. The calculation condition may be one of these conditions, or may be and or or of a plurality of conditions. If it is determined in S260 that the calculation condition is satisfied, the process proceeds to S270, and if it is determined that the calculation condition is not satisfied, the present process is terminated as it is.

In S270, the orientation error in the front direction is calculated based on the distribution of the accumulated high points. Specifically, as shown in FIG. 9, the cumulative distribution with respect to the detection direction (that is, the angle φ) of the high point is obtained from the distribution of the high point, and the angle corresponding to the median value is used as the measured value. Then, the difference between the measured value and the reference angle −φinst is obtained as the directional error Δφo in the front direction. The cumulative distribution may be created for the entire distance range, but may be created for only the range above the lower limit distance (for example, 70 m or more). In the latter case, the processing load when obtaining the directional error Δφo from the cumulative distribution function can be reduced.

The high-level points stored in the memory may be deleted each time the directional error is calculated, or when the maximum number that can be stored is accumulated, the old high-level points are extracted every time a new high-level point is extracted. It may be deleted in order from.

In S280, the direction correction table stored in the memory 42 is updated according to the direction error Δφo in the front direction obtained in S270, and this process is terminated.
In the present embodiment, S220 is the first step and static extraction unit, S230 is the second step and front extraction unit, S240 to S250 are the third step and high-level extraction unit, and S270 is the fourth step and error calculation unit. Corresponds to. Further, the signal processing unit 4 corresponds to the direction error detection device.

[3. effect]
According to the embodiment described in detail above, the following effects are obtained.
(1) In the in-vehicle radar device 1, a reflection point on a superstructure such as a signboard installed above the road, that is, a reflection point on an object existing in the front direction of the vehicle is extracted as a high-level point, and this extraction is performed. The orientation error Δφo in the front direction is obtained by using the obtained high point. That is, in order to obtain the directional error Δφo in the front direction without using the reflection point from a low position containing many reflection points (that is, noise) based on an object existing in the direction other than the front direction, the directional error Δφo in the front direction is used. , Can be obtained accurately in a short time.

(2) In the in-vehicle radar device 1, it is possible to determine whether or not it is a high point using the height pattern of the front point. Therefore, for that determination, a device for detecting the height of the object is separately prepared. There is no need, and the device configuration can be simplified.

[4. Other embodiments]
Although the embodiments of the present disclosure have been described above, the present disclosure is not limited to the above-described embodiments, and can be variously modified and implemented.

(A) In the above embodiment, the case where the front direction included in the detection range of the in-vehicle radar device 1 is the rear along the straight direction of the vehicle has been described, but the front direction is along the straight direction of the vehicle. The techniques of the present disclosure may be applied where forward.

(B) In the above embodiment, a method using the maximum likelihood method is adopted as the first learning process, but the method is not limited thereto. As the method of the first learning process, in principle, another method may be adopted, which is characterized in that the orientation error in the front direction cannot be obtained or the calculation accuracy of the orientation error in the front direction is lowered.

(C) A plurality of functions possessed by one component in the above embodiment may be realized by a plurality of components, or one function possessed by one component may be realized by a plurality of components. .. Further, a plurality of functions possessed by the plurality of components may be realized by one component, or one function realized by the plurality of components may be realized by one component. Further, a part of the configuration of the above embodiment may be omitted. Further, at least a part of the configuration of the above embodiment may be added or replaced with the configuration of the other above embodiment. It should be noted that all aspects included in the technical idea specified from the wording described in the claims are embodiments of the present disclosure.

(D) In addition to the orientation error detection device and the orientation error detection method realized by the signal processing unit 4 described above, a system having the orientation error detection device as a component, and a program for operating a computer as the orientation error detection device. The present disclosure can also be realized in various forms such as a non-transitional actual recording medium such as a semiconductor memory in which this program is recorded.

1 ... In-vehicle radar device, 2 ... Antenna unit, 3 ... Transmission / reception unit, 4 ... Signal processing unit, 41 ... CPU, 42 ... Memory.

Claims (7)

The reflection point observed by the FMCW radar mounted on the vehicle is acquired so that the detection range includes the front direction of the front or the rear along the straight direction of the vehicle, and the observed value of the reflection point is the most likely method. Is used to separate the actual orientation and the orientation error, and by obtaining a regression function that expresses the relationship between the actual orientation and the orientation error from the distribution of points that represent the relationship between the actual orientation and the orientation error, the orientation error other than the front direction is obtained. The first error calculation step to find
The first step (S220) of extracting the resting point presumed to be stationary among the reflection points, and
It is located within the front range set around the reference direction among the stationary points extracted in the first step with the front direction obtained from the installation state of the FMCW radar on the vehicle as the reference direction. The second step (S230) of extracting the front point and
Among the front points extracted in the second step, the third step (S240 to S250) of extracting the high point presumed to exist at a position higher than the preset high threshold value, and the third step (S240 to S250).
The fourth step (4th step) in which the front direction obtained from the distribution of high-order points extracted in the third step is used as the measurement direction, and the deviation of the measurement direction with respect to the reference direction is obtained as the orientation error in the front direction. S270) and
Orientation error detection method. The reflection point observed by the FMCW radar mounted on the vehicle is acquired so that the detection range includes the front direction of the front or the rear along the straight direction of the vehicle, and the observed value of the reflection point is the most likely method. Is used to separate the actual orientation and the orientation error, and by obtaining a regression function that expresses the relationship between the actual orientation and the orientation error from the distribution of points that represent the relationship between the actual orientation and the orientation error, the orientation error other than the front direction is obtained. 1st error calculation unit (S110 to S170) for obtaining
A static extraction unit (S220) configured to extract a stationary point that is presumed to be stationary among the reflection points, and a static extraction unit (S220).
Of the stationary points extracted by the stationary extraction unit, the front surface located within the frontal range set around the reference direction with the frontal direction obtained from the installed state of the FMCW radar on the vehicle as the reference direction. A front extraction unit (S230) configured to extract points, and
Among the front points extracted by the front extraction unit, the high-level extraction unit (S240 to S250) configured to extract the high-level points presumed to exist at a position higher than the preset high-level threshold value, and the high-level extraction unit (S240 to S250).
The measurement direction is the front direction obtained from the distribution of high points extracted by the high-level extraction unit, and the deviation of the measurement direction with respect to the reference direction is obtained as the orientation error in the front direction . 2 Error calculation unit (S270) and
A directional error detector equipped with. The directional error detection device according to claim 2.
The high-level extraction unit tracks the front point to generate a height pattern showing the relationship between the distance between the vehicle and the front point and the reflection intensity from the front point, and the height pattern is used to generate the height pattern. It is configured to determine whether the front point is the high point.
Directional error detector. The directional error detection device according to claim 3.
The high-level extraction unit is configured to use the height pattern obtained for a region where the distance to the vehicle is longer than the preset lower limit distance for determining whether or not the high-level point is the high-level point.
Directional error detector. The directional error detection device according to claim 4.
The lower limit distance is a value larger than the distance to the farthest null point within the detection range of the FMCW radar, which exists in the ideal height pattern required for a stationary object having a height that the vehicle can overcome. Set,
Directional error detector. The directional error detection device according to any one of claims 2 to 5.
The high threshold value is set to be equal to or higher than the lower limit of the height from the road surface of the signboard installed above the road.
Directional error detector. The directional error detection device according to any one of claims 2 to 6.
The second error calculation unit is configured to obtain the measurement direction by using a cumulative distribution function that accumulates the high points with the orientation as a parameter.
Directional error detector.

Images