radar equipment

Abstract

To automatically detect and correct deviation in a short time when there is the deviation in a mounting angle of a radar device.SOLUTION: A side wall parallel to a traveling direction is extracted based on a distance and a relative velocity, which are the positioning results of a radar device, by function fitting processing from a function fitting processing unit (112). Next, a calculation processing unit (114) calculates an angular orientation β of each point extracted as the side wall, based on a vehicle speed and a distance to the side wall calculated during the function fitting. A comparison processing unit (115) compares the calculated angular orientation β and an angular orientation Θ, which is the positioning result, to detect the mounting angular deviation of the radar device, and a correction processing unit (116) corrects the mounting angular deviation.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a radar device, and more particularly to a technique for correcting angular deviations that may occur when an on-board radar device is mounted on a vehicle. [Background technology]

[0002] In recent years, high-precision target position detection has become increasingly important for autonomous driving vehicles. Millimeter-wave radar devices used in autonomous driving are generally installed inside the bumper, and installation errors that occur when installing the bumper, as well as vibrations and shocks after installation, can cause the installation angle of the millimeter-wave radar device to deviate from the intended angle.

[0003] If the mounting angle of the millimeter-wave radar is misaligned, for example, when detecting a vehicle ahead, a vehicle traveling in the same lane may be mistaken for one traveling in an adjacent lane, which could result in a delay in issuing an alarm or automatically braking. Furthermore, if there is a misalignment in the mounting angle when combining position information with other sensors such as cameras, the same target may not be combined properly, resulting in the target being identified as a different object, or the target being overlooked. Even if the mounting angle misalignment is relatively small, it is necessary to widen the gate range for capturing the target when tracking it, taking into account the misalignment in the mounting angle. Widening the gate range increases the frequency of picking up noise such as clutter, which increases the probability of incorrect tracking and, as a result, increases the possibility of overlooking the target.

[0004] A known technology for correcting the mounting angle of a vehicle's millimeter-wave radar is, for example, that disclosed in Patent Document 1. In Patent Document 1, the axle is calculated by sensing the entire vehicle using a vehicle surroundings recognition sensor, and the amount of deviation in the sensor mounting angle is calculated from the relationship between a target installed in front of the vehicle, the calculated axle, and the angular orientation of the target detected by the sensor. While this method can accurately calculate the axle, it requires a dedicated system, which means that after shipping, the vehicle must be brought into a dedicated facility where the system is installed.

[0005] In Patent Document 2, targets with a relative velocity of zero are detected when the vehicle is traveling straight ahead, and the amount of deviation in the mounting angle is calculated by utilizing the fact that targets with a relative velocity of zero are at an angle of 90 degrees relative to the traveling direction. This method does not require the vehicle to be brought into a dedicated facility, but it does require the accumulation of targets with a relative velocity of zero. In particular, front radars installed in front of the vehicle almost never detect targets oriented directly to the side (90 degrees relative to the traveling direction), so even if the mounting angle is deviated, it is necessary to drive for a long period of time in a state where the angle deviation is not detected. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 2019-74398 [Patent Document 2] Japanese Patent Application Laid-Open No. 2014-153256 Summary of the Invention [Problem to be solved by the invention]

[0007] An object of the present invention is to realize a radar device that automatically detects and corrects any deviation in the mounting angle in a short time when such deviation occurs in the radar device.

[0008] The above and other objects and novel features of the present invention will become apparent from the description of this specification and the accompanying drawings. [Means for solving the problem]

[0009] Representative inventions disclosed in this application will be briefly described below.

[0010] A function fitting processing unit (112) performs function fitting processing from the distance and relative speed, which are the positioning results of the radar device, to extract side walls parallel to the traveling direction. Next, a calculation processing unit (114) calculates the angular direction β of each point extracted as a side wall based on the vehicle speed and the distance to the side wall calculated during function fitting. A comparison processing unit (115) compares the calculated angular direction β with the angular direction Θ, which is the positioning result, to detect a mounting angle deviation in the radar device, which is then corrected by a correction processing unit (116). [Effects of the Invention]

[0011] This makes it possible to automatically detect the radar mounting angle on a vehicle with high accuracy in a short time. It also makes it possible to detect the vehicle speed with high accuracy at the same time as the mounting angle.

[0012] Problems, configurations, and effects other than those described above will become apparent from the following description of the preferred embodiments of the invention. [Brief explanation of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram illustrating the configuration of a radar device according to an embodiment of the present invention. [Figure 2] FIG. 2 is an image diagram showing the relationship between the side wall parallel to the host vehicle speed vector and the relative speed. [Figure 3] Figure 3 shows an image of a curve that shows a side wall parallel to the vehicle's speed vector on a two-dimensional plane of distance R and relative speed V, and the equation of the curve. [Figure 4] FIG. 4 is a diagram showing the results of a group of points detected as targets on a two-dimensional plane of distance R and relative velocity V, and the results of extracted curve fitting. [Figure 5] Figure 5 shows the results of displaying the point cloud extracted as the side wall, with the horizontal axis representing the angle orientation Θ and the vertical axis representing the angle orientation β, as well as the results of linear approximation. [Figure 6] FIG. 6 shows an example of the data structure output to the memory. [Figure 7A]FIG. 7A is an image diagram of a curve on a two-dimensional plane of the distance R and the relative speed V when there is a side wall that curves convexly relative to the vehicle. [Figure 7B] FIG. 7B is an image diagram of a curve on a two-dimensional plane of the distance R and the relative speed V when there is a side wall that curves concavely relative to the vehicle. [Figure 8A] FIG. 8A is an image diagram of a curve on a two-dimensional plane of the distance R and the relative speed V when there is a diagonal sidewall in the approaching direction relative to the host vehicle speed Vr vector. [Figure 8B] FIG. 8B is an image diagram of a curve on a two-dimensional plane of the distance R and the relative velocity V when there is a diagonal sidewall in the direction away from the host vehicle velocity Vr vector. [Figure 9] FIG. 9 is a diagram illustrating a block configuration of a radar device according to an embodiment of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0014] Hereinafter, embodiments of the present invention will be described with reference to the drawings. The embodiments or examples are merely examples for explaining the present invention, and some omissions and simplifications have been made as appropriate for clarity of explanation. The present invention can also be implemented in various other forms. Unless otherwise specified, each component may be singular or plural.

[0015] In order to facilitate understanding of the invention, the position, size, shape, range, etc. of each component shown in the drawings may not represent the actual position, size, shape, range, etc. Therefore, the present invention is not necessarily limited to the position, size, shape, range, etc. disclosed in the drawings.

[0016] When there are multiple components with the same or similar functions, they may be described using the same reference numeral with different subscripts. When there is no need to distinguish between these multiple components, the subscripts may be omitted. [Example]

[0017] Representative embodiments of the invention disclosed in this application will be described with reference to the drawings. Fig. 1 is a diagram illustrating the block configuration of a radar device according to an embodiment. Fig. 2 is an image diagram showing the relationship between a side wall parallel to the host vehicle speed vector and the relative speed. Fig. 3 is an image diagram and equation of a curve showing the side wall parallel to the host vehicle speed vector on a two-dimensional plane of distance R and relative speed V.

[0018] A radar device (201) according to a representative embodiment of the present invention is an on-board millimeter wave radar device attached to the left and right front corners of an automobile, and as shown in FIG. 1, is composed of an analog processing unit (101), a positioning processing unit (102), and an angle deviation detection unit (103).

[0019] The analog processing unit (101) outputs millimeter wave radar from the transmitting antenna (104) as a transmission signal 1. Here, the transmission signal 1 uses the output signal of a synthesizer (105), and a chirp signal in which the frequency is linearly shifted over time is often used. The output signal from the synthesizer (105) is amplified by a transmission amplifier 3 and transmitted from the transmitting antenna (104).

[0020] The transmitted transmission signal 1 is reflected by a target object, and a portion of the reflected wave returns to the receiving array antenna (106) and is received as a received signal 2. The receiving array antenna (106) has multiple receiving antennas that receive reflected waves from the millimeter-wave radar. The received signals 2 received by the multiple receiving antennas of the receiving array antenna (106) are amplified by multiple receiving amplifiers 4, respectively, and input to multiple mixers (107) that are frequency converters, where they are down-converted by the multiple mixers (107). At this time, the output signal of the synthesizer (105) is used as a local signal for the mixer (107). As a result, a frequency corresponding to the time difference between the transmission signal and the received signal, i.e., the distance to the target object, is output as multiple analog output signals from the multiple mixers (107). The multiple analog output signals of the mixer (107) are filtered by multiple filters 5, then input to multiple A / D converters (analog-to-digital conversion circuits) (108) and converted into multiple digital signals, which are then transmitted as multiple digital output signals to the positioning processor (102). Here, targets include moving objects such as automobiles and people, and stationary objects on the ground such as walls and telephone poles.

[0021] The positioning processor (102) receives a plurality of digital output signals from the analog processor (101) as received signals, and performs FFT (FFT: Fast Fourier Transform) processing on each received signal in a plurality of time / frequency FFT processing circuits (109). The FFT processing includes a frequency FFT and a time FFT. The distance R can be determined by performing the frequency FFT, and the relative velocity V can be determined by performing the time FFT. Once the range R and relative velocity V of the target are determined, the complex signal of each received signal can be extracted from the results of the FFT processing. The angle and azimuth processor (110) calculates the angular azimuth Θ (angle and azimuth 1) of the target using angle and azimuth processing such as spatial FFT processing, digital beamforming processing, and MUSIC (Multiple Signal Classification) processing based on the regularity of the complex signal of each received signal. The output of the positioning processor (102) is output as a point cloud (111) having information on the distance R, relative velocity V, and angular azimuth Θ according to the number of targets.

[0022] Here, the relative velocity V calculated by the time / frequency FFT processing circuit (109) corresponds to the amount of change per unit time in the distance of a concentric circle centered on the millimeter-wave radar device (201), and as shown in Fig. 2, the relative velocity V of a stationary target (203) in the direction of angle β (angular direction 2) with respect to the traveling direction of a vehicle 210 traveling in the direction of a host vehicle speed Vr vector (202) is a value obtained by multiplying the host vehicle speed Vr by the term of cosine β (V = Vr cos β). The vehicle 210 shown in Fig. 2 shows an example in which a millimeter-wave radar device (201) is also provided at the left front corner.

[0023] Furthermore, a side wall (204), which is an object parallel to the host vehicle speed Vr vector (202) shown in Figure 2, draws a specific curve (301) as shown in Figure 3 when represented on a two-dimensional plane of distance R and relative speed V. This curve (301) passes through proximity distance X (205) when relative speed V becomes zero, and as distance R increases, it asymptotically approaches host vehicle speed Vr, and is a curve that is uniquely determined by host vehicle speed Vr and proximity distance X, as shown by equation (302) for curve (301) of side wall (204). Here, a feature is an object that is stationary on the ground, such as a wall or a telephone pole.

[0024] In the angle deviation detection unit (103), a function fitting processing unit (112) performs function fitting processing on point cloud information of the distance R and the relative velocity V among the distance R, relative velocity V, and angular orientation Θ of the point cloud (111), thereby extracting a side wall (204) parallel to the host vehicle velocity Vr vector (202). Specifically, the function fitting processing involves sweeping the host vehicle velocity Vr and the proximity distance X to determine the curve (301) in FIG. 3 that best fits the point cloud. During fitting, points far from the curve (301) are not considered to be side walls (204), and if the number of points corresponding to the fitting is greater than a predetermined threshold, the side wall (204) is detected. When a side wall (204) is detected, the host vehicle velocity Vr and the side wall distance X are calculated as information (113). This information (113) can be used as a correction unit (119) for the host vehicle speed Vr0 output from the vehicle speed sensor (118) or as automatic driving information. If the comparison unit (120) compares the host vehicle speed Vr calculated by the function fitting processing unit (112) with the host vehicle speed Vr0 calculated by the vehicle speed sensor (118) and finds that the difference is greater than a predetermined value, a sensor abnormality warning unit (121) issues a warning indicating a sensor abnormality, thereby notifying the driver of the vehicle 210 of a malfunction or deterioration in accuracy of the vehicle speed sensor (118) by sound or display. In addition, the error between the point cloud and the function, which are function-fitted by the function fitting processing unit (112), can be used as accuracy information for the detection accuracy of the side wall (204). The comparison unit (120) and correction unit (119) can be collectively referred to as a correction function for the host vehicle speed Vr0 calculated by the vehicle speed sensor (118).

[0025] 4 shows the point cloud (111) of the detected target and the curve (301) extracted by the function fitting processing unit (112). It can be seen that the function fitting processing unit (112) detected the vehicle speed of 93 km / h, the right wall at a distance of 8.5 m, and the left wall at a distance of 4.8 m.

[0026] Since the vehicle speed Vr or the sidewall distance X is known through calculation by the function fitting processing unit (112), the angle direction β of each point of the point cloud can be calculated by the angle direction β calculation unit (114) using the curve equation (302). As a result, each point of the point cloud extracted as the sidewall (204) can have two types of angle direction information: the angle direction β calculated by the angle direction β calculation processing unit (114) and the angle direction Θ calculated in advance by the angle direction processing unit (110). The angle direction β is based on the vehicle speed Vr vector (202), and the angle direction Θ is based on the mounting direction of the millimeter-wave radar device (201), i.e., the mounting axis 206. Therefore, by comparing these two angles (β, Θ) in the angle comparison processing unit (115), the mounting angle of the millimeter-wave radar device (201) can be calculated. The mounting angle of the millimeter wave radar device (201) means the angle between the vehicle speed Vr vector (axial direction or longitudinal direction of the vehicle) and the mounting axis 206. When comparing these two angles (β, Θ), for example, it is advisable to form a point cloud on a two-dimensional plane of the angular orientation Θ and the angular orientation β and perform linear approximation on the point cloud.

[0027] Figure 5 shows the results of linear approximation of the point cloud extracted as a sidewall, with the horizontal axis representing angular orientation Θ and the vertical axis representing angular orientation β. The dotted line (501) in Figure 5 is a straight line representing the linear approximation of the point cloud. Considering that Θ and β are angular orientations on the same scale, the slope of the line is understood to be 1. If the slope deviates significantly from 1, the data can be determined to be abnormal, and measures such as discarding the data can be taken. The linear approximation equation (502: y = 0.9988x + 0.4285) in Figure 5 can be determined to be normal data because the slope (0.9988) is close to 1. The y-intercept of the linear approximation equation (502), 0.4285, represents the angular deviation between angular orientation β and angular orientation Θ, and the value of this y-intercept represents the installation angle deviation of the millimeter-wave radar device (201). By statistically accumulating the values ​​of these y-intercepts, it is possible to calculate an accurate mounting angle deviation of the millimeter wave radar device (201).

[0028] FIG. 6 shows an example of the data structure output to memory, illustrating an example of the data structure when statistically accumulating data. The elapsed time (601) represents the time elapsed since the angle deviation correction and alarm issuance were performed. When the linear approximation formula (502) is calculated at 0.5-second intervals, the data is accumulated in 0.5-second increments. The number of fitted points (602) indicates the number of points extracted when the curve (301) is function-fitted by the function fitting processing unit (112). The more points used, the more accurate the angle deviation detection. Therefore, by using these points (602) as weights, the accuracy of data utilization can be improved. The mounting angle (603) is a numerical value (mounting angle deviation) corresponding to the y-intercept of the linear approximation formula (502). The sidewall distance (604) and the vehicle speed (605) are values ​​calculated by the function fitting processing unit (112) and correspond to the vehicle speed Vr and the sidewall distance X information (113) in FIG. 1.

[0029] An example of a method for calculating the mounting angle deviation is to calculate a weighted average of the number of fitted points (602) for the mounting angle (603). Here, by providing a mechanism for calculating the weighted average value described above when the cumulative value of the number of fitted points (602) exceeds a predetermined threshold, it is possible to detect the mounting angle deviation value with the desired error accuracy. For the detected mounting angle deviation, a simple correction process is performed by the correction processing unit (116) by subtracting the angle deviation (603) from the angle orientation Θ calculated by the angle orientation processing unit (110). This allows the calculation of a corrected angle orientation Θ' without any mounting angle error, and the output of a positioning result (117) related to the corrected point cloud. Furthermore, by providing a mechanism (sensor abnormality warning unit 121 in FIG. 1) that issues an alarm as a sensor abnormality when the detected mounting angle deviation value (603) exceeds a predetermined threshold, it is possible to guarantee the beam width in antenna design and the field of view in system design.

[0030] One concern here is that the error in the calculated angular orientation β may be large if the detected sidewall is significantly curved or oblique. Figures 7A, 7B, 8A, and 8B show deviations from the curve (301) extracted by the function fitting processor (112) in these cases.

[0031] 7A shows a curve (702) on a two-dimensional plane of the distance R and relative velocity V of a side wall (701) that curves convexly relative to the vehicle. FIG. 7B shows a curve (704) on a two-dimensional plane of the distance R and relative velocity V of a side wall (703) that curves concavely relative to the vehicle. These curves (702) and (704) are different from the curve (301) on a two-dimensional plane of the side wall (204) that is parallel to the host vehicle speed Vr vector (202), and therefore are likely to be omitted during function fitting processing by the function fitting processor (112). In other words, since they are not detected as side walls (204), they do not affect the error in the angular orientation β.

[0032] 8A and 8B show cases where a sidewall is present at an angle to the host vehicle speed Vr vector (202). Fig. 8A shows a curve (802) on a two-dimensional plane of the distance R and the relative speed V when a sidewall (801) is present at an angle in the direction approaching the host vehicle. Fig. 8B shows a curve (804) on a two-dimensional plane of the distance R and the relative speed V when a sidewall (803) is present at an angle in the direction receding from the host vehicle. These curves (802) and (804) are characterized by being asymmetric on the positive and negative sides of the relative speed V, and are different from the curve (301) on a two-dimensional plane of the sidewall (204) parallel to the host vehicle speed Vr vector (202), shown by the dotted line. Therefore, as with the curved sidewalls (701) and (703), they are likely to be omitted during function fitting processing by the function fitting processing unit (112).

[0033] However, for sidewalls with gentle curves or gentle inclinations, they may be detected as sidewalls parallel to the traveling direction because they match the curve (301) extracted by the function fitting processor (112) to some extent. In this case, it is possible to suppress deterioration in the detection accuracy of the mounting angle deviation by performing a weighted cumulative average of the error between the point cloud function-fitted by the function fitting processor (112) and the function curve (301) as accuracy information of the sidewall detection accuracy. As another method, a steering angle sensor may be used to prevent sidewall detection when the steering angle exceeds a predetermined steering angle range. This can more completely omit the detection of curves or inclined sidewalls, thereby improving the detection accuracy of the mounting angle deviation. For example, sidewall extraction may be performed within a steering angle range of ±5 degrees. In this case, sidewall extraction is not performed for steering angle ranges other than ±5 degrees.

[0034] Another concern regarding deterioration in accuracy of mounting angle misalignment is that when the relative velocity V is faster than a predetermined velocity, a slower relative velocity V is output due to FFT aliasing during calculation by the time / frequency FFT processing circuit (109). This may result in failure during function fitting processing by the function fitting processing unit (112), which may degrade the detection accuracy of the mounting angle misalignment. Regarding this aliasing of the relative velocity, it is possible to suppress deterioration in the detection accuracy of the mounting angle misalignment by repeatedly arranging a two-dimensional plane of the distance R and the relative velocity V and then performing function fitting processing by the function fitting processing unit (112). Furthermore, by using a speed sensor to perform sidewall extraction only at a predetermined traveling speed, it is possible to further suppress deterioration in the detection accuracy of the mounting angle misalignment. For example, if the aliasing speed of the relative velocity is 70 km / h, sidewall extraction can be performed by limiting the speed detected by the speed detection output of the speed sensor to a predetermined speed range, such as 5 km / h to 70 km / h. In this case, sidewall extraction will not be performed outside the range of 5 km / h to 70 km / h. That is, the predetermined steering angle range and the predetermined speed range can be used as activation conditions for the angle deviation detection unit (103) or the function fitting processing unit (112).

[0035] Fig. 9 is a diagram illustrating the block configuration of a radar device according to an embodiment of the present invention. Fig. 9 differs from Fig. 1 in that in radar device 201A in Fig. 9, the steering angle detection output of steering angle sensor 130 and the speed detection output of the speed sensor are input to a function fitting processing unit (112) in angle deviation detection unit 103. The other configurations and operations of radar device 201A are the same as those of radar device 201 in Fig. 1, and redundant explanations will be omitted.

[0036] As a result, the operation of the function fitting processor (112) is controlled so that the steering angle detection output of the steering angle sensor 130 extracts the sidewalls within a predetermined steering angle range, such as within a steering angle of ±5 degrees, and the speed detected by the speed detection output of the speed sensor extracts the sidewalls within a predetermined speed range, such as from 5 km / h to 70 km / h. Of course, the function fitting processor (112) may be controlled to perform the sidewall extraction operation within a steering angle range of ±5 degrees and within a speed range of from 5 km / h to 70 km / h.

[0037] The predetermined steering angle range and the predetermined speed range can be set as activation conditions for the angle deviation detection unit (103) or the function fitting processing unit (112). This allows automatic detection and correction of deviation in a short time when deviation occurs in the mounting angle of the radar device. It is also possible to improve the detection accuracy of the mounting angle deviation. Furthermore, it is possible to suppress deterioration of the detection accuracy of the mounting angle deviation.

[0038] 1 and 9, the positioning processing unit 102, angle deviation detection unit 103, angle azimuth Θ correction processing unit 116, vehicle speed Vr0 correction unit 119, vehicle speed comparison unit 120, and sensor abnormality warning unit 121 may each be configured as a dedicated hardware circuit or as software. When configured as software, the system is configured using a microcomputer or microprocessor equipped with a central processing unit CPU, read-only memory ROM, random access memory RAM, etc., and the software stored in the ROM is executed by the CPU, and the execution results, etc. are stored in the RAM. This makes it possible to automatically detect the radar mounting angle to the vehicle with high accuracy in a short time.

[0039] The invention made by the inventor has been specifically described above based on the embodiments, but it goes without saying that the present invention is not limited to the above embodiments and can be modified in various ways. [Explanation of symbols]

[0040] 101 Analog Processing Section 102 Positioning processing unit 103 Angle deviation detection unit 104 transmitting antenna 105 Synthesizer 106 Receiving array antenna 107 Mixer 108 A / D converters 109 Time / Frequency FFT Processing Circuit 110 Angle and direction processing section 111 Detected target point cloud 112 Function fitting processing section 113 Distance to side wall (X) 114 Angular direction (β) calculation processing unit 115 Angle orientation comparison section 116 Angular azimuth (Θ) correction processing unit 117 Positioning results without installation angle errors 118 Speed ​​Sensor 119 Speed ​​sensor detection speed correction unit 120 Vehicle speed comparison unit 121 Sensor abnormality alarm unit 130 Steering angle sensor 201 Millimeter wave radar equipment 202 Vehicle velocity vector (Vr) 203 Stationary target at angle β 204 Side walls parallel to the vehicle velocity vector 205 Melee Range (X) 206 Mounting shaft 301 Curve in the plane of distance-relative velocity of the side wall parallel to the vehicle's velocity vector 302 Equation of the curve of the side wall parallel to the vehicle velocity vector 501 Linear approximation 502 Linear approximation formula 601 Elapsed Time 602 Fitted points (detection results) 603 Mounting angle (detection result) 604 Sidewall distance (detection result) 605 Vehicle speed (detection result) 701 curved sidewalls 702 Curve of a curved side wall in the distance-relative velocity plane 703 Curved sidewalls 704 Curve of a curved side wall in the distance-relative velocity plane 801 Angled sidewall 802 Curve of the distance-relative velocity plane of an oblique side wall 803 Angled Sidewall 804 Curve of the distance-relative velocity plane of an oblique side wall

Claims

1. In an on-vehicle radar device capable of measuring the angular direction (1) of a target, the distance to the target, and the relative speed thereof using a plurality of receiving antennas, a function fitting processing unit that extracts features including side walls parallel to the vehicle traveling direction from the information on the distance and the relative speed by function fitting; a processing unit that calculates an angular orientation 2 of each point of the point cloud extracted as the feature from the result of the function fitting; a comparison unit to which the point cloud having the angular orientation 1 and the angular orientation 2 is input, the comparison unit displays the point cloud on a two-dimensional plane with the x-axis representing the angular orientation 1 and the y-axis representing the angular orientation 2, and calculates the angular deviation of the mounting axis of the radar device from the y-intercept of an approximate line obtained by linearly approximating the point cloud.

2. The radar device according to claim 1, a correction processing unit that corrects the angular deviation of the mounting shaft based on the calculation result of the angular deviation.

3. The radar device according to claim 1, The radar device is characterized by having a function of issuing an alarm when the calculated angular deviation exceeds a predetermined angular deviation.

4. The radar device according to claim 1, The radar device is characterized in that the function fitting processing unit is configured to calculate a vehicle speed from information on the distance to the feature and the relative speed during the function fitting.

5. The radar device according to claim 4, A radar device having a correction function for correcting an error of a speed sensor based on the result of the host vehicle speed.

6. The radar device according to claim 5, The radar device is characterized in that the function fitting processing unit performs the function fitting when the speed detection output from the speed sensor is within a predetermined speed range.

7. The radar device according to claim 5, The radar device is characterized in that the function fitting processing unit performs the function fitting when the steering angle detection output from the steering angle sensor is within a predetermined steering angle range.

8. The radar device according to claim 7, The radar device is characterized in that the function fitting processing unit performs the function fitting when the steering angle detection output is within the predetermined steering angle range and the speed detection output from the speed sensor is within a predetermined speed range.

9. A radar device mounted on a vehicle, extracting a sidewall parallel to the traveling direction of the vehicle by function fitting from the distance obtained as a result of the positioning of the radar device and the relative speed of the vehicle; calculating an angular orientation β of each point of the point cloud extracted as the side wall based on the vehicle speed of the vehicle and the distance to the side wall calculated during the function fitting; a radar device that displays the point cloud on a two-dimensional plane with the x-axis representing the angular orientation Θ that is the positioning result and the y-axis representing the calculated angular orientation β, calculates an installation angle deviation in the radar device from a y-intercept of an approximate line obtained by linearly approximating the point cloud, and corrects the installation angle deviation.

Images