Obstacle detection device

Abstract

To accurately identify a position of an obstacle even in a situation where multiple reflection occurs.SOLUTION: An obstacle detection device 100 uses a first distance sensor 10A and a second distance sensor 10B to detect a position of an obstacle. The first distance sensor transmits a first survey wave having a first frequency and receives a first reflected wave that is a reflected wave obtained by reflecting the first survey wave on the obstacle and has the first frequency. The second distance sensor transmits a survey wave having a second frequency that is a frequency different from the first frequency and receives a second reflected wave that is a reflected wave obtained by reflecting a second survey wave on the obstacle and has the second frequency. The obstacle detection device includes: a measuring section 110 that measures a first distance from the first distance sensor to the obstacle from the first reflected wave, and measures a second distance from the second distance sensor to the obstacle from the second reflected wave; and a position calculation section 120 that obtains position coordinates indicating the position of the obstacle from an interval between the measured first and second distances and the first and second distance sensors.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present disclosure relates to an obstacle detection device. [Background technology]

[0002] Patent Document 1 describes an obstacle detection device that detects obstacles around a vehicle. In this obstacle detection device, multiple distance sensors each transmit a probe wave and receive a reflected wave reflected by an obstacle. The reflected wave includes a direct wave, which is a reflected wave of the probe wave transmitted by the distance sensor itself, and an indirect wave, which is a reflected wave of the probe wave transmitted by the other distance sensor. This obstacle detection device calculates the position of an intersection indicating the location of a reflection point by using a two-circle intersection process from pairs of distance information obtained from the direct waves received by each of the multiple sensors. This obstacle detection device also calculates the position of an intersection by using a two-circle intersection process from pairs of distance information obtained from the direct wave and the indirect wave received by one sensor. Furthermore, this obstacle detection device groups the multiple intersections obtained based on the distance between the intersections and determines whether the obstacles are the same object based on the grouping results. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] International Publication No. 2019 / 058507 Summary of the Invention [Problem to be solved by the invention]

[0004] In the technology described in Patent Document 1, as the indirect wave, an indirect wave is assumed in which the exploration wave transmitted from the other distance sensor is reflected only at one reflection point. However, for example, depending on the position or shape of an obstacle in the direction in which the exploration wave is transmitted, multiple reflections occur in which the exploration wave is reflected multiple times by the same object. In this case, it is assumed that the distance sensor receives an indirect wave of multiple reflections, which is a reflected wave in which the exploration wave transmitted from the other distance sensor is reflected at two or more reflection points. In a situation where multiple reflections occur, the distance information obtained from the indirect wave includes an error. Therefore, for example, when two-circle intersection processing is performed using distance information including an error, an intersection may be calculated at a position where there is no obstacle. As described above, when the conventional method is used, there is a problem that the position of the obstacle cannot be accurately specified in a situation where multiple reflections occur.

Means for Solving the Problems

[0005] The present disclosure can be realized in the following forms.

[0006] According to one aspect of the present disclosure, an obstacle detection device is provided. This obstacle detection device (100) Installed in the car (A1) uses a first distance sensor (10A) and a second distance sensor (10B) to detect the position of an obstacle. The first distance sensor transmits a first exploration wave having a first frequency, and receives a first reflected wave that is a reflected wave of the first exploration wave reflected by the obstacle and has the first frequency. The second distance sensor transmits an exploration wave having a second frequency that is different from the first frequency, and receives a second reflected wave that is a reflected wave of the second exploration wave reflected by the obstacle and has the second frequency. The obstacle detection device includes a measurement unit (110) that measures a first distance from the first distance sensor to the obstacle from the first reflected wave received by the first distance sensor, and measures a second distance from the second distance sensor to the obstacle from the second reflected wave received by the second distance sensor, and a position calculation unit (120) that obtains position coordinates indicating the position of the obstacle from the measured first distance and second distance and the distance between the first distance sensor and the second distance sensor. When the first distance sensor receives the first reflected wave reflected by the obstacle and the second distance sensor receives the second reflected wave reflected by the same obstacle, and the signal waveform representing the first reflected wave and the signal waveform representing the second reflected wave each contain at least two maximum points, the method for calculating the angle indicating the direction in which the obstacle exists is changed depending on the magnitude relationship between the reception intensities of the at least two maximum points included in the signal waveform representing the first reflected wave received by the first distance sensor and the magnitude relationship between the reception intensities of the at least two maximum points included in the signal waveform representing the second reflected wave received by the second distance sensor.

[0007] According to this aspect, the obstacle detection device uses only the first reflected wave received by the first distance sensor at the first frequency and the second reflected wave received by the second distance sensor at the second frequency as reflected waves to measure the distance to the obstacle. That is, the obstacle detection device uses only direct waves as reflected waves to measure the distance to the obstacle. Because indirect waves are not used to measure the distance to the obstacle, no error is included in the measured distance even in a situation where multiple reflections occur, and the position of the obstacle can be accurately detected. [Brief explanation of the drawings]

[0008] [Figure 1] 1 is a block diagram showing a schematic configuration of an obstacle detection system 1. FIG. [Figure 2] FIG. 2 is a diagram showing the positional relationship between a distance sensor mounted on a vehicle and an object. [Figure 3] FIG. 2 is a diagram showing angles indicating the direction of an object as seen from each distance sensor. [Figure 4] FIG. 10 is a diagram for explaining a method for detecting reception of a reflected wave. [Figure 5] FIG. 10 is a diagram illustrating a situation in which multiple reflections occur. [Figure 6] FIG. 10 is a diagram illustrating an example of bandwidth setting. [Figure 7] FIG. 10 is a diagram illustrating another example of bandwidth setting. [Figure 8] FIG. 2 is a diagram for explaining road surface reflected waves; [Figure 9] 10 is a diagram showing the change over time in the reception intensity of reflected waves including road surface reflected waves. [Figure 10] FIG. 10 is a diagram showing the relationship between road surface reflection intensity and incident angle. [Figure 11] FIG. 10 is a diagram for explaining a method for calculating the time when the reception intensity becomes 0. [Figure 12] FIG. 10 is a diagram illustrating another method for calculating the time when the reception intensity becomes 0. [Figure 13]FIG. 10 is a diagram for explaining a method for determining whether or not multiple peaks are reflections from the same object. [Figure 14] 10 is a diagram for explaining a method for calculating an angle indicating the direction of a reflection point according to the magnitude relationship between a plurality of peaks. FIG. [Figure 15] FIG. 2 is a diagram showing angles indicating the direction of an object as seen from each distance sensor. [Figure 16] 10 is a diagram for explaining a method for calculating an angle indicating the direction of a reflection point according to the magnitude relationship between a plurality of peaks. FIG. [Figure 17] 10 is a diagram for explaining a method for calculating an angle indicating the direction of a reflection point according to the magnitude relationship between a plurality of peaks. FIG. [Figure 18] 10A and 10B are diagrams illustrating another example of the positional relationship between the distance sensor and the object. [Figure 19] FIG. 10 is a diagram illustrating an example of setting the frequency of a transmission signal in another embodiment 2. DETAILED DESCRIPTION OF THE INVENTION

[0009] A1. Embodiment 1 1, in the embodiment, an obstacle detection system 1 that detects obstacles present around a vehicle will be described. The obstacle detection system 1 includes distance sensors 10A and 10B, a transmitting / receiving circuit 20, an alarm unit 30, and an obstacle detection device 100.

[0010] As shown in FIG. 2, the obstacle detection system 1 is a system mounted on vehicle A1 to detect the position of an object M1, which is an obstacle around vehicle A1. The obstacle detection system 1 uses distance sensors 10A and 10B provided on vehicle A1 to detect the position of object M1. Object M1 is also referred to as the target object. In the embodiment, it is assumed that object M1 is an object having a rectangular parallelepiped shape. Distance sensors 10A and 10B are installed on the front bumper of vehicle A1. Distance sensors 10A and 10B are ultrasonic sensors using piezoelectric elements. Distance sensors 10A and 10B transmit ultrasonic waves as the probing waves. Also, distance sensors 10A and 10B receive the reflected waves reflected by object M1 around vehicle A1. Distance sensor 10A is also referred to as the first distance sensor. Distance sensor 10B is also referred to as the second distance sensor.

[0011] As shown in FIG. 2, in the embodiment, the vehicle width direction is set as the X-axis, the vehicle height direction is set as the Y-axis, and the traveling direction of vehicle A1 is set as the Z-axis. Distance sensor 10A is arranged at a position higher than that of distance sensor 10B in the vehicle height direction. It is preferable that the positions of distance sensors 10A and 10B on the X-axis and the positions on the Z-axis are the same. When there is a deviation between the positions of distance sensors 10A and 10B on the X-axis and the positions on the Z-axis, it is assumed that the amount of deviation is set within a preset range. Distance sensor 10A transmits a probing wave at a preset frequency and receives the reflected wave. The probing wave transmitted by distance sensor 10A is also referred to as the first probing wave. The reflected wave received by distance sensor 10A is also referred to as the first reflected wave. Distance sensor 10B transmits a probing wave at a frequency different from the frequency used by distance sensor 10A and receives the reflected wave. The probing wave transmitted by distance sensor 10B is also referred to as the second probing wave. The reflected wave received by distance sensor 10B is also referred to as the second reflected wave.

[0012] Setting the frequency used by distance sensor 10A and the frequency used by distance sensor 10B to be different means setting the bandwidth used by distance sensor 10A and the bandwidth used by distance sensor 10B so that they do not overlap. For example, if distance sensors 10A and 10B are sensors using piezoelectric elements, the frequency settings of the two sensors can be made different by changing the shape of the piezoelectric elements used in each sensor. The bandwidth used by distance sensor 10A is also called the first bandwidth. The bandwidth used by distance sensor 10B is also called the second bandwidth.

[0013] The frequency used by distance sensor 10A and the frequency used by distance sensor 10B are set to be different in order to prevent distance sensors 10A and 10B from receiving indirect waves. This is because in the obstacle detection system 1, only direct waves are used as reflected waves to measure the distance to object M1. Direct waves are reflected waves received by distance sensors 10A and 10B that are the probe waves transmitted by each of them and reflected by object M1 around vehicle A1. Indirect waves are reflected waves that are the probe waves transmitted by the other distance sensor and reflected by object M1 around vehicle A1. The specific method for setting the frequencies will be described later. The frequency at which distance sensor 10A transmits probe waves is also referred to as the first frequency. The frequency at which distance sensor 10B transmits probe waves is also referred to as the second frequency.

[0014] 1, the transmission / reception circuit 20 drives the distance sensors 10A and 10B to transmit search waves under the control of the obstacle detection device 100. The transmission / reception circuit 20 also supplies the obstacle detection device 100 with reception signals indicating the reflected waves received by the distance sensors 10A and 10B.

[0015] The warning unit 30 outputs a warning when the obstacle detection device 100 detects that an object M1 that may collide with the vehicle A1 is present around the vehicle A1.

[0016] The obstacle detection device 100 detects the position of an object M1 using distance sensors 10A and 10B. In the example shown in FIG. 2, there is one object M1 around the vehicle A1, but there may be two or more objects M1. As shown in FIG. 1, the obstacle detection device 100 has a measurement unit 110 and a position calculation unit 120. The functions of the measurement unit 110 and the position calculation unit 120 are realized by an ECU (Electronic Control Unit) or ASIC (Application Specific Integrated Circuit) provided in the vehicle A1.

[0017] The measurement unit 110 causes the distance sensors 10A and 10B to transmit search waves at preset timings via the transmission / reception circuit 20. At this time, the measurement unit 110 stores the transmission times in a memory provided in the obstacle detection device 100. Furthermore, when the measurement unit 110 receives reception signals indicating reflected waves received by the distance sensors 10A and 10B via the transmission / reception circuit 20, the measurement unit 110 stores the reception times in a memory provided in the obstacle detection device 100.

[0018] As shown in FIG. 3, the line connecting the distance sensor 10A and the object M1 represents the path of the reflected wave from the probe wave transmitted from the distance sensor 10A, reflected by the object M1, and returning to the distance sensor 10A as a reflected wave. The line connecting the distance sensor 10B and the object M1 represents the path of the reflected wave from the probe wave transmitted from the distance sensor 10B, reflected by the object M1, and returning to the distance sensor 10B as a reflected wave. The measurement unit 110 measures the distance L1 from the distance sensor 10A to the reflection point on the object M1 based on a received signal indicating the reflected wave received by the distance sensor 10A. The measurement unit 110 also calculates the distance L2 from the distance sensor 10B to the reflection point on the object M1 based on a received signal indicating the reflected wave received by the distance sensor 10B. The distance from the distance sensor 10A to the object M1 is also referred to as the first distance. The distance from the distance sensor 10B to the object M1 is also referred to as the second distance.

[0019] Specifically, first, the measurement unit 110 calculates the time Tr1 between the time when the distance sensor 10A transmits the search wave and the time when the reflected wave is received.

[0020] As shown in the upper part of Fig. 4, the measurement unit 110 determines the time when the distance sensor 10A received the reflected wave from a signal waveform that represents the change over time in the reception strength V1 at the distance sensor 10A. The measurement unit 110 determines the time Tr1 as the period from time T0, when the distance sensor 10A transmitted the search wave, to time T1, when the reception strength V1 at the distance sensor 10A exceeded a preset threshold Vth1. Here, time T1, when the set threshold Vth1 was exceeded, is considered to be the time when the distance sensor 10A received the reflected wave.

[0021] Similarly, as shown in the lower part of Figure 4, the measurement unit 110 determines the time when the distance sensor 10B received the reflected wave from a signal waveform that represents the time change in the reception strength V2 at the distance sensor 10B. The measurement unit 110 determines the time Tr2 as the period from time T0, when the distance sensor 10B transmitted the search wave, to time T2, when the reception strength V2 at the distance sensor 10B exceeded a preset threshold Vth2. The time T2, when the set threshold Vth2 was exceeded, is considered to be the time when the distance sensor 10B received the reflected wave. In the example shown in Figure 4, it is assumed that the distance sensors 10A and 10B transmitted search waves simultaneously.

[0022] The measurement unit 110 calculates the distance L1 from the distance sensor 10A to the reflection point on the object M1 using the following formula (1): where C represents the speed of sound. L1=C Tr1 / 2 (1)

[0023] Similarly, the measurement unit 110 calculates the distance L2 from the distance sensor 10B to the reflection point on the object M1 using the following equation (2). L2=C Tr2 / 2 (2)

[0024] As described above, by setting the frequency, the distance sensors 10A and 10B are each controlled so as not to receive the reflected wave of the exploration wave transmitted from the other distance sensor. Therefore, the measurement unit 110 calculates the distances L1 and L2 using only the direct wave. The advantages of calculating the distances L1 and L2 using only the direct wave are described below.

[0025] As shown in FIG. 5, multiple reflections may occur depending on the position or shape of an obstacle in the direction in which the exploration wave is transmitted. For example, a part of the wall of an indoor parking space may protrude. In the illustrated example, the distance sensors 10A and 10B are arranged along the vehicle width direction. The exploration wave transmitted from the distance sensor 10A indicated by the broken line is reflected at two points and then reaches the distance sensor 10B. The exploration wave transmitted from the distance sensor 10B indicated by the broken line is also reflected at two points and then reaches the distance sensor 10A. Thus, in FIG. 5, multiple reflections occur in which the exploration wave is reflected multiple times by the same object. The solid line represents the direct wave. The measurement unit 110 calculates the distance to the obstacle from the time from when the exploration wave is transmitted until the reflected wave is received. However, in a situation where multiple reflections occur, since the exploration wave is reflected at two or more points, it is assumed that the time until the reflected wave is received becomes longer compared to the case where the exploration wave is reflected at one point. For this reason, an error is included in the calculated distance to the obstacle.

[0026] On the other hand, in the embodiment, since only the direct wave is used to calculate the distances L1 and L2, no error is included in the measured distances L1 and L2.

[0027] As shown in FIG. 3, the position calculation unit 120 calculates position coordinates indicating the position of the reflection point on the object M1 using the distances L1 and L2 calculated by the measurement unit 110. The point where the vertical line from the distance sensor 10A intersects the road surface is set as the origin (0, 0, 0). The position calculation unit 120 obtains position coordinates in the XYZ orthogonal coordinate system as the position of the reflection point on the object M1.

[0028] Specifically, first, the position calculation unit 120 obtains an angle θ1 representing the direction of the reflection point as seen from the distance sensor 10A and an angle θ2 representing the direction of the reflection point as seen from the distance sensor 10B, respectively. The angle θ1 refers to the angle formed by the direction of the reflection point as seen from the distance sensor 10A with respect to the horizontal direction. The angle θ2 refers to the angle formed by the direction of the reflection point as seen from the distance sensor 10B with respect to the horizontal direction. For ease of understanding, in FIG. 3, the illustration of the exploration waves transmitted by the distance sensors 10A and 10B is omitted.

[0029] The position calculation unit 120 calculates the angle θ by the following formula (3). Let the distance between the distance sensor 10A and the distance sensor 10B be the interval d1. The interval d1 is the difference between the height hs1 representing the position of the distance sensor 10A in the Y-axis direction and the height hs2 representing the position of the distance sensor 10B in the Y-axis direction. θ1 = arcsin{-(L1 2 + d1 2 - L2 2 ) / (2·L1·d1)} ··· (3)

[0030] The position calculation unit 120 calculates the angle θ2 by the following formula (4). In the illustrated example, since the reflected waves received by the distance sensors 10A and 10B are both reflected from the same reflection point, the position calculation unit 120 only needs to calculate either the angle θ1 or the angle θ2. θ2 = arcsin{-(L2 2 + d1 2 - L1 2 ) / (2·L2·d1)} ··· (4)

[0031] After that, the position calculation unit 120 calculates the position y1 on the Y-axis of the reflection point where the exploration wave transmitted from the distance sensor 10A is reflected by the following formula (5). Further, the position calculation unit 120 calculates the position y2 on the Y-axis of the reflection point where the exploration wave transmitted from the distance sensor 10B is reflected by the following formula (6). In the illustrated example, the reflected waves received by the distance sensors 10A and 10B are both reflected from the same reflection point. In this case, y1 = y2 can be considered. The positions y1 and y2 can be calculated by formula (5) or formula (6). y1 = hs1 - L1sinθ1 ···(5) y2 = hs2 - L2sinθ2 ···(6)

[0032] In addition, the position calculation unit 120 calculates the position z1 on the Z-axis of the reflection point where the exploration wave transmitted from the distance sensor 10A is reflected by the following formula (7). Further, the position calculation unit 120 calculates the position z2 on the Z-axis of the reflection point where the exploration wave transmitted from the distance sensor 10B is reflected by the following formula (8). In the illustrated example, the reflected waves received by the distance sensors 10A and 10B are both reflected from the same reflection point. In this case, z1 = z2 can be regarded as. The positions z1 and z2 can be calculated by formula (7) or formula (8). z1 = L1cosθ1 ···(7) z2 = L2cosθ2 ···(8)

[0033] When the distance between the object M1 and the vehicle A1 is less than or equal to a preset distance from the calculated position coordinates of the reflection point on the object M1, for example, the position calculation unit 120 determines that the object M1 may collide with the vehicle A1. In this case, the position calculation unit 120 notifies the alarm unit 30 to that effect.

[0034] As described above, in the embodiment, since the indirect wave is not used for measuring the distances L1 and L2 to the object M1, even in a situation where multiple reflections occur, the measured distances do not include errors, and the position of the obstacle can be accurately detected. Also, since the indirect wave is not used for measuring the distances L1 and L2, the amount of calculation can be reduced compared to the case where the distance is calculated using the direct wave and the indirect wave. Therefore, the time required for measuring the distance can be shortened.

[0035] A2. Embodiment 2 In Embodiment 2, a specific method for setting the frequencies used by distance sensor 10A and distance sensor 10B will be described. As described above, the bandwidths of distance sensor 10A and distance sensor 10B are set so as not to overlap. The bandwidth is the range between the upper limit frequency and the lower limit frequency.

[0036] In the example shown in FIG. 6, the reception sensitivity of distance sensor 10A is highest at resonance frequency f01. The reception sensitivity of distance sensor 10B is highest at resonance frequency f02. Here, of the frequencies when the reception sensitivity of distance sensor 10A has decreased by 3 dB from its maximum value, the lower one is set as lower limit frequency f11 and the higher one is set as upper limit frequency f12. Also, of the frequencies when the reception sensitivity of distance sensor 10B has decreased by 3 dB from its maximum value, the lower one is set as lower limit frequency f21 and the higher one is set as upper limit frequency f22. From lower limit frequency f11 to upper limit frequency f12 is defined as the bandwidth BW1 of distance sensor 10A. From lower limit frequency f21 to upper limit frequency f22 is defined as the bandwidth BW2 of distance sensor 10B. At this time, bandwidths BW1 and BW2 are set so as to satisfy the following formula (9). Note that BW1 = f12 - f11 and BW2 = f22 - f21. f01 + 1 / 2·BW1 ≦ f02 - 1 / 2·BW2 ···(9)

[0037] Note that FIG. 6 shows an example where there is a difference between the upper limit frequency f12 of distance sensor 10A and the lower limit frequency f21 of distance sensor 10B.

[0038] As shown in FIG. 7, bandwidths BW1 and BW2 may be set so that they approach each other compared to FIG. 6. In the illustrated example, the upper limit frequency f12 of distance sensor 10A and the lower limit frequency f21 of distance sensor 10B overlap. Also in this case, in order to satisfy the above formula (9), the same effect as the example shown in FIG. 6 can be expected.

[0039] In this way, the bandwidth BW1 of the distance sensor 10A and the bandwidth BW2 of the distance sensor 10B are set so as not to overlap with each other within the range from maximum to half of the receiving sensitivity. Therefore, the distance sensors 10A and 10B cannot obtain sufficient receiving sensitivity for the reflected waves of the search waves transmitted by the other distance sensor. For example, even if the reflected wave of the search wave transmitted by the distance sensor 10B reaches the distance sensor 10A, the distance sensor 10A is likely unable to detect the signal due to its weak signal strength. Alternatively, if the signal received by the distance sensors 10A and 10B has a receiving strength below a predetermined level, the measurement unit 110 may detect the signal as an indirect wave. This simple method allows the distance sensors 10A and 10B to be controlled so that they only receive the reflected waves of the search waves transmitted by themselves.

[0040] In this case, as in the first embodiment, even in a situation where multiple reflections occur, the measured distance does not contain any error, and the position of the obstacle can be detected accurately.

[0041] A3. Embodiment 3 In the third embodiment, the frequency used by the distance sensor 10A or the distance sensor 10B that is located at a higher position in the vehicle height direction is set to be lower than that of the other distance sensor.

[0042] As shown in FIG. 8, it is assumed that the distance sensor 10A is positioned higher in the vehicle height direction than the distance sensor 10B. In this case, it is preferable that the frequency used by the distance sensor 10A is set to be lower than the frequency used by the distance sensor 10B. The reason for this is explained below. As shown in FIG. 8, the search waves transmitted by the distance sensors 10A and 10B may be reflected by the road surface. Hereinafter, the waves reflected by the road surface will be referred to as road surface reflected waves. In the example shown in the figure, a road surface reflected wave having an incident angle θG1 is received by the distance sensor 10A.

[0043] As shown in Figure 9, distance sensor 10A receives a composite wave W3, which is a combination of a reflected wave W1 reflected by object M1 and a road surface reflected wave W2. In the waveform shown, the wave represented by the dashed line is the reflected wave W1. The wave represented by the broken line is the road surface reflected wave W2. The wave represented by the solid line is the composite wave W3. Similarly, distance sensor 10B receives a composite wave W3, which is a combination of a reflected wave W1 reflected by object M1 and a road surface reflected wave W2. Because what distance sensors 10A and 10B are supposed to detect is object M1, road surface reflected wave W2 can be considered a noise component.

[0044] Furthermore, since the distance sensor 10A is disposed at a higher position than the distance sensor 10B, the incident angle θG1 of the road surface reflected wave to the distance sensor 10A is greater than the incident angle of the road surface reflected wave to the distance sensor 10B.

[0045] As shown in Figure 10, the road surface reflection intensity, which is the reception strength of waves reflected from the road surface, tends to increase as the angle of incidence to the distance sensor increases. Therefore, the reception strength of road surface reflected waves received by distance sensor 10A is stronger than the reception strength of road surface reflected waves received by distance sensor 10B. In other words, the noise components received by distance sensor 10A are greater than the noise components received by distance sensor 10B. In the obstacle detection system 1, obstacles are detected using distance sensors 10A and 10B, so it is undesirable for the difference between the noise components input to one distance sensor and the other to be large.

[0046] Furthermore, as shown in FIG. 10, when comparing road surface reflected waves having a high frequency with road surface reflected waves having a low frequency, the road surface reflection intensity R1 of the road surface reflected waves having a high frequency tends to be higher than the road surface reflection intensity R2 of the road surface reflected waves having a low frequency.

[0047] Therefore, the frequency used by the distance sensor located at a higher position in the vehicle height direction is set lower than that of the other distance sensor. The frequency used by distance sensor 10A, which receives road surface reflected waves with a stronger intensity due to a larger incident angle of the road surface reflected waves, is set lower than the frequency used by distance sensor 10B, which receives road surface reflected waves with a lower incident intensity. By setting the frequency used by distance sensor 10A lower than that of distance sensor 10B, the road surface reflection intensity of road surface reflected waves at distance sensor 10A can be reduced. Therefore, the difference in magnitude between the noise caused by road surface reflected waves received by distance sensor 10A and the noise caused by road surface reflected waves received by distance sensor 10B can be reduced.

[0048] A4. Embodiment 4 In the first embodiment, the measurement unit 110 calculates the distance between the distance sensor and the reflection point by determining the time when the received strength exceeds the threshold as the time when the reflected wave is received. Alternatively, the time when the reflected wave is received may be determined by other methods. For example, the measurement unit 110 may determine the time when the received strength of the reflected wave calculated from the peak time becomes zero as the time when the reflected wave is received. A peak refers to, for example, a point in a waveform representing the change over time in the received strength of the reflected wave where the value indicating the slope changes from a positive value to a negative value. A peak is also called a local maximum point.

[0049] As shown in FIG. 11, when the received intensity V1 exceeds the threshold value Vth1, the measurement unit 110 calculates the slope K1 of the received intensity by K1 = ΔV1 / Δt. The slope K1 is the slope of the tangent line of the signal waveform at the time when the received intensity V1 exceeds the threshold value Vth1. Δt is a short period determined in advance. ΔV1 is the change amount of the received intensity V1 during the period represented by Δt. The measurement unit 110 calculates the time T1 when the received intensity V1 becomes 0 on the tangent line having the slope K1. Similarly, when the received intensity V2 exceeds the threshold value Vth2, the measurement unit 110 calculates the slope K2 of the received intensity by K2 = ΔV2 / Δt. The slope K2 is the slope of the tangent line of the signal waveform at the time when the received intensity V2 exceeds the threshold value Vth2. The measurement unit 110 calculates the time T2 when the received intensity V2 becomes 0 on the tangent line having the slope K2. By using the times T1 and T2 obtained in this way, the measurement unit 110 can calculate more accurate distances L1 and L2.

[0050] As shown in FIG. 12, as another method, the measurement unit 110 enlarges or reduces the received waveform of the distance sensor 10A stored in advance in the memory provided in the obstacle detection device 100 according to the received intensity of the peak generated immediately after exceeding the threshold value. The measurement unit 110 may calculate, as the time t11, the time when the received intensity becomes 0 before the peak occurs, and as the time t12, the time when the received intensity becomes 0 after the peak occurs, in the enlarged or reduced waveform. Similarly, the measurement unit 110 may calculate the times t21 and t22 from the received waveform of the distance sensor 10B stored in advance. The received waveforms stored in advance can be prepared as follows. Each of the distance sensors 10A and 10B transmits a probing wave toward, for example, a facing wall and receives the reflected wave reflected by the wall. The waveform of the received reflected wave may be stored in advance in the memory as the received waveform. By using the time t11 obtained in this way as the time T1 and the time t21 as T2, the measurement unit 110 can calculate more accurate distances L1 and L2.

[0051] A5. Embodiment 5 In the embodiment, as shown in FIG. 4, an example in which one peak is included in the signal waveforms representing the signals respectively received by the distance sensors 10A and 10B has been described. However, there may be a plurality of peaks included in each of the signal waveforms representing the received signals of the distance sensors 10A and 10B.

[0052] As shown in FIG. 13, in the signal waveform representing the received signal, peaks occur at time Tpk1, time Tpk2, and time Tpk3. The reflected waves having each peak may be reflected waves reflected from different objects. Alternatively, the reflected waves having each peak may be reflected waves reflected from the same object. For example, the probing wave may be reflected at the upper end and the lower end of the same object, respectively. In this case, there is a difference in the lengths of the respective paths between the reflected wave reflected at the upper end and the reflected wave reflected at the lower end. In such a case, in the signal waveform representing the received signal, the peak of the reflected wave reflected at the upper end may appear first, and then the peak of the reflected wave reflected at the lower end may appear.

[0053] Therefore, when the time difference between two adjacent peaks on the time axis is equal to or less than a preset time, the measurement unit 110 determines that all of the reflected waves having each peak are reflected from the same object. The preset time is also referred to as a reference value. Further, when the difference in time between two adjacent peaks on the time axis exceeds the preset time, the measurement unit 110 determines that the reflected waves having each peak are reflected from different objects. The measurement unit 110 calculates the time difference between two adjacent peaks for each combination of two peaks on the time axis, and determines whether or not the reflected waves represented by each of the two peaks are all reflected from the same object.

[0054] In the example shown in FIG. 13, assume that the difference T12 between time Tpk1 and time Tpk2 is within a preset time. In this case, the measurement unit 110 determines that the reflected wave having a peak at time Tpk1 and the reflected wave having a peak at time Tpk2 are reflected from the same object. For example, the measurement unit 110 determines that the reflected waves received in period T4, which is enclosed by the dashed line and includes the preset periods before and after time Tpk1 and the preset periods before and after time Tpk2, are reflected waves reflected from the same object.

[0055] Also, assume that the difference T23 between time Tpk2 and time Tpk3 exceeds a preset time. In this case, the measurement unit 110 determines that the reflected wave having a peak at time Tpk2 and the reflected wave having a peak at time Tpk3 represent reflected waves reflected from different objects, respectively.

[0056] Also, when three or more peaks occur within a preset time, the measurement unit 110 may determine that all the reflected waves having each peak are reflected from the same object. With the above configuration, it is possible to easily determine whether the reflected waves are from the same object.

[0057] In the signal waveform representing the received signal, when a plurality of peaks occur, the distance to the reflection point can be calculated as follows.

[0058] As shown in the upper part of FIG. 14, assume that the signal waveform representing the signal received by the distance sensor 10A includes peak p11 and peak p12. Assume that the measurement unit 110 determines that the reflected wave having the earlier-occurring peak p11 and the reflected wave having the later-occurring peak p12 are reflected waves reflected from the same object M1. Also, as shown in the lower part of FIG. 14, assume that the signal waveform representing the signal received by the distance sensor 10B includes peak p21 and peak p22. Assume that the measurement unit 110 determines that the reflected wave having the earlier-occurring peak p21 and the reflected wave having the later-occurring peak p22 are reflected waves reflected from the same object M1.

[0059] In FIG. 15, the reflected wave received by distance sensor 10A is represented by a dashed line, and the reflected wave received by distance sensor 10B is represented by a dashed line. For ease of understanding, the detection waves transmitted by each distance sensor are not shown. Distance sensors 10A and 10B receive reflected waves reflected by an object having a surface directly facing them. The upper and lower ends of object M1 have limited surfaces directly facing distance sensors 10A and 10B. Furthermore, the upper end of object M1 has a smaller surface area directly facing distance sensors 10A and 10B than the lower end of object M1, which is in contact with the road surface. Therefore, the received strength of the reflected wave from the upper end of object M1 is smaller than the received strength of the reflected wave from the lower end of object M1.

[0060] Therefore, the later-occurring peak p12 shown in the upper part of FIG. 14 is a peak of the reflected wave received by the distance sensor 10A and reflected near the road surface of the object M1. This reflected wave received by the distance sensor 10A is also called the third reflected wave. As shown in FIG. 15, the distance between the distance sensor 10A and the reflection point near the road surface of the object M1 is defined as distance L12. Furthermore, the earlier-occurring peak p11 shown in the upper part of FIG. 14 is a peak of the reflected wave received by the distance sensor 10A and reflected near the top end of the object M1. This reflected wave received by the distance sensor 10A is also called the first reflected wave. As shown in FIG. 15, the distance between the distance sensor 10A and the reflection point near the top end of the object M1 is defined as distance L11.

[0061] The later-occurring peak p22 shown in the lower part of FIG. 14 is a peak of the reflected wave received by the distance sensor 10B and reflected near the road surface in the object M1. This reflected wave received by the distance sensor 10B is also referred to as the fourth reflected wave. As shown in FIG. 15, let the distance between the distance sensor 10B and the reflection point near the road surface in the object M1 be the distance L22. Also, the previously-occurring peak p21 shown in the lower part of FIG. 14 is a peak of the reflected wave received by the distance sensor 10B and reflected near the upper end in the object M1. This reflected wave received by the distance sensor 10B is also referred to as the second reflected wave. As shown in FIG. 15, let the distance between the distance sensor 10B and the reflection point near the upper end in the object M1 be the distance L21.

[0062] The measurement unit 110 calculates the distances L11, L12, L21, and L22 as follows. First, the measurement unit 110 calculates the distance L11 according to the above formula (1). Specifically, the measurement unit 110 multiplies the time from the time when the exploration wave is transmitted to the time t11 by the speed of sound C and divides the obtained value by 2 to calculate the distance L11. The measurement unit 110 multiplies the time from the time when the exploration wave is transmitted to the time t12 by the speed of sound C and divides the obtained value by 2 to calculate the distance L12. Similarly, the measurement unit 110 multiplies the time from the time when the exploration wave is transmitted to the time t21 by the speed of sound C and divides the obtained value by 2 to calculate the distance L21. The measurement unit 110 multiplies the time from the time when the exploration wave is transmitted to the time t22 by the speed of sound C and divides the obtained value by 2 to calculate the distance L22. The times t11, t12, t21, and t22 at which the reception intensity of the reflected wave becomes zero can be calculated respectively by, for example, the methods shown in FIG. 11 or FIG. 12.

[0063] Furthermore, it is assumed that at least two peaks included in the signal waveform representing the signal received by the distance sensor 10A and at least two peaks included in the signal waveform representing the signal received by the distance sensor 10B are determined to represent reflected waves reflected by the same object. In this case, the position calculation unit 120 changes the method of calculating the angle indicating the direction in which the object M1 exists depending on the magnitude relationship between the reception intensities of the at least two peaks in the signal waveform representing the signal received by the distance sensor 10A and the magnitude relationship between the reception intensities of the at least two peaks in the signal waveform representing the signal received by the distance sensor 10B. Below, when (a) V11≦V12 and V21≦V22, (b) V11 <V12かつV21≧V22である場合、(c)V11≧V12かつV21≧V22である場合、それぞれについて、角度θ11、θ12、θ21、θ22の算出方法を説明する。

[0064] As shown in Figure 15, angle θ11 refers to the angle formed with the horizontal direction in the direction of a reflection point near the top of object M1 as seen from distance sensor 10A. Angle θ12 refers to the angle formed with the horizontal direction in the direction of a reflection point near the road surface as seen from distance sensor 10A. Angle θ21 refers to the angle formed with the horizontal direction in the direction of a reflection point near the top of object M1 as seen from distance sensor 10B. Angle θ22 refers to the angle formed with the horizontal direction in the direction of a reflection point near the road surface as seen from distance sensor 10B.

[0065] (a) Calculation method for angles when V11≦V12 and V21≦V22 As shown in FIG. 14, a method for calculating angles θ11, θ12, θ21, and θ22 when V11 ≤ V12 and V21 ≤ V22 with respect to the reception intensity of each peak will be described. For example, when the height ht of the object M1 is lower than the height hs2 of the lower distance sensor 10B, it is conceivable that V11 ≤ V12 and V21 ≤ V22. This is because it is assumed that among the reflected waves received by the distance sensor 10A, the reflected wave reflected from the lower end of the object M1 has a stronger reception intensity than the reflected wave reflected from the upper end of the object M1. The same applies to the reflected waves received by the distance sensor 10B.

[0066] Specifically, the position calculation unit 120 calculates the angle θ11 by the above-described formula (3) using the calculated distances L11 and L21. The position calculation unit 120 calculates θ21 by the above-described formula (4) using the distances L11 and L21. Further, the position calculation unit 120 calculates the angle θ12 by the above-described formula (3) using the calculated distances L12 and L22. The position calculation unit 120 calculates θ22 by the above-described formula (4) using the distances L12 and L22.

[0067] (b) Method for calculating the angle when V11 < V12 and V21 ≥ V22 As shown in FIG. 16, a method for calculating angles θ11, θ12, θ21, and θ22 when V11 < V12 and V21 ≥ V22 with respect to the reception intensity of each peak will be described. For example, when the height ht of the object M1 is smaller than the height hs1 of the distance sensor 10A and equal to or greater than the height hs2 of the distance sensor 10B, it is conceivable that V11 < V12 and V21 ≥ V22. This is because it is considered that the exploration wave transmitted from the distance sensor 10B is reflected by the object M1 in the front direction of the distance sensor 10B and near the road surface.

[0068] In this case, it is assumed that the reception intensity of the reflected wave transmitted from the distance sensor 10B and reflected by the object M1 in the front direction of the distance sensor 10B is equal to or greater than the reception intensity of the reflected wave transmitted from the distance sensor 10B and reflected near the road surface. Also, since the height ht of the object M1 is lower than the height hs1 of the distance sensor 10A, it is assumed that the reception intensity of the reflected wave transmitted from the distance sensor 10A and reflected near the road surface is stronger than the reception intensity of the reflected wave transmitted from the distance sensor 10A and reflected at other locations of the object M1. Other locations are, for example, near the upper end of the object M1. Also, the distance from the reflection point in the front direction of the distance sensor 10B to the distance sensor 10B is shorter than the distance from the reflection point near the road surface to the distance sensor 10B. Therefore, it is considered that the reflected wave having the previously generated peak is the reflected wave reflected by the object M1 in the front direction of the distance sensor 10B. The same applies to the reflected wave received by the distance sensor 10A.

[0069] Specifically, the position calculation unit 120 calculates the angle θ12 by the above-described formula (3) using the calculated distances L12 and L22. The position calculation unit 120 calculates θ22 by the above-described formula (4) using the distances L12 and L22. The position calculation unit 120 sets the angle θ21 to 0 degrees. The position calculation unit 120 determines that the angle θ11 cannot be calculated.

[0070] (c) Method for calculating the angle when V11 ≧ V12 and V21 ≧ V22 As shown in FIG. 17, the calculation method for angles θ11, θ12, θ21, and θ22 for the reception intensities of the respective peaks when V11≧V12 and V21≧V22 is described below. For example, when the height ht of object M1 is equal to or greater than the height hs1 of the distance sensor 10A, which is positioned higher, V11≧V12 and V21≧V22 may be satisfied. This is because the search waves transmitted from distance sensors 10A and 10B are expected to be reflected from object M1 in the direction in front of each distance sensor and near the road surface. In this case, the reception intensity of the wave transmitted from distance sensor 10A and reflected by object M1 in the direction in front of distance sensor 10A is expected to be equal to or greater than the reception intensity of the wave transmitted from distance sensor 10A and reflected near the road surface. Furthermore, the distance from the reflection point in the direction in front of distance sensor 10A to distance sensor 10A is shorter than the distance from the reflection point near the road surface to distance sensor 10A. Therefore, the reflected wave having the peak that occurs first is considered to be the reflected wave reflected by the object M1 in the front direction of the distance sensor 10 A. The same applies to the distance sensor 10 B.

[0071] Specifically, the position calculation unit 120 uses the calculated distances L12 and L22 to calculate the angle θ12 according to the above-mentioned equation (3). The position calculation unit 120 uses the distances L12 and L22 to calculate the angle θ22 according to the above-mentioned equation (4). The position calculation unit 120 sets the angles θ11 and θ21 to 0 degrees.

[0072] As described above, the position calculation unit 120 changes the method for calculating the angle indicating the direction in which the object M1 exists depending on the magnitude relationship of the peak reception intensities. This makes it possible to prevent erroneous detection of the direction in which the object M1 exists when the magnitude relationship of the multiple maximum points included in the signal waveform of the reception signal from the distance sensor 10A differs from the magnitude relationship of the multiple maximum points included in the signal waveform of the reception signal from the distance sensor 10B.

[0073] Also, when two angles are calculated from the reflected waves received by one distance sensor as in Embodiment 5, the position calculation unit 120 calculates the coordinates of the reflection point by Expressions (10) to (17). y11 = hs1 - L11sinθ11 ···(10) z11 = L11cosθ11 ···(11) y12 = hs1 - L12sinθ12 ···(12) z12 = L12cosθ12 ···(13) y21 = hs2 - L21sinθ21 ···(14) z21 = L21cosθ21 ···(15) y22 = hs2 - L22sinθ22 ···(16) z22 = L22cosθ22 ···(17)

[0074] A6. Embodiment 6 In Embodiment 5, the position calculation unit 120 detects the height ht of the object M1 from the calculated coordinate values. The position calculation unit 120 may detect the height ht of the object M1 by Expression (18) using the y - coordinate values y11 and y12 of the reflection point calculated from the distances L11 and L12 measured by the distance sensor 10A, respectively. ht = y11 - y12 ···(18)

[0075] Alternatively, the position calculation unit 120 may detect the height ht of the object M1 by Expression (19) using the y - coordinate values y21 and y22 of the reflection point calculated from the distances L21 and L22 measured by the distance sensor 10B, respectively. ht = y21 - y22 ···(19)

[0076] Note that the height ht of the object M1 obtained by the above Expression (18) or (19) represents the height that the object M1 has at least.

[0077] However, the position calculation unit 120 does not use the above formula to calculate the height ht when the reception intensity V21 of the first peak in the signal waveform representing the reception signal of the distance sensor 10B is equal to or greater than the reception intensity V22 of the second peak, as shown in Figure 13 or 14. The position calculation unit 120 determines that the height is equal to or greater than the position hs2 in the height direction of the distance sensor 10B.

[0078] As shown in Figure 14, if the reception strength V11 of the peak that occurs first in the signal waveform of the reception signal of the distance sensor 10A is greater than or equal to the reception strength V12 of the peak that occurs later, the position calculation unit 120 determines that the height ht is greater than or equal to the heightwise position hs1 of the distance sensor 10A.

[0079] Furthermore, if there is variation in the calculated z-axis coordinate values z11, z12, z21, and z22 of the reflection point, the position calculation unit 120 determines that the reliability of the height ht determination result is low. Specifically, if a value indicating the variation in the values z11, z12, z21, and z22 exceeds a preset threshold, the position calculation unit 120 determines that there is variation in the values z11, z12, z21, and z22. With this configuration, it is possible to detect at least the height of the object M1 and determine the reliability of the height detection result.

[0080] This is because it is expected that the variations in the values z11, z12, z21, and z22 will be large if the search wave is reflected by an object M2 having a shape as shown in Fig. 18. On the other hand, if the object M1 has a rectangular parallelepiped shape as shown in Fig. 2, it is expected that the variations in the values z11, z12, z21, and z22 will be small.

[0081] B1. Other embodiment 1 The height of the object M1 can also be estimated using a trained machine learning model. For example, suppose that the reflected waves received by the distance sensor 10A and the reflected waves received by the distance sensor 10B are determined to be reflected waves from the same object M1. In this case, a certain range of waveforms including a peak is extracted from the waveforms representing the received signals of the distance sensors 10A and 10B. The reception strength for each preset period included in the extracted waveforms and the time information corresponding to each reception strength are input to the trained machine learning model. The machine learning model outputs the height of the object M1. Thus, the height of the object M1 can be automatically estimated.

[0082] As the training data, a set of data including the reception intensity and time information of a reflected wave reflected by an object having a known height and received by the distance sensor 10A, the reception intensity and time information of a reflected wave reflected by an object having a known height and received by the distance sensor 10B, and the height of the object is used. In the learning phase, machine learning is performed using the training data.

[0083] B2. Other embodiment 2 In the second embodiment, the resonant frequency of the distance sensor 10A is set to f01, and the resonant frequency of the distance sensor 10B is set to f02. Both distance sensors transmit signals of a constant frequency as search waves. However, the signals transmitted by the distance sensors 10A and 10B may have frequencies that change over time. Specifically, the distance sensors 10A and 10B may transmit signals modulated with a triangular wave. In this case, the signals transmitted by the distance sensors 10A and 10B include an up-segment in which the frequency increases over time and a down-segment in which the frequency decreases over time. In this case, as in the second embodiment, the bandwidth of the distance sensor 10A must be set so as not to overlap with the bandwidth of the distance sensor 10B.

[0084] As shown in FIG. 19, the frequencies of the transmission signals of the distance sensors 10A and 10B change over time. It is assumed that the frequency f1 of the transmission signal of the distance sensor 10A and the frequency f2 of the transmission signal of the distance sensor 10B at the same time satisfy the following equation. Let the center frequency of the transmission signal in the distance sensor 10A be f01 and the center frequency of the transmission signal in the distance sensor 10B be f02. Let the upper limit frequency of the transmission signal in the distance sensor 10A be f12 and the lower limit frequency be f11. Let the upper limit frequency of the transmission signal in the distance sensor 10B be f22 and the lower limit frequency be f21. f2 - f1 ≧ ((f02 - f21) + (f12 - f01)) ···(20)

[0085] B3. Other Embodiment 3 In Embodiments 1 to 5, an example in which the distance sensors 10A and 10B transmit ultrasonic waves as the detection wave has been described. However, the distance sensors 10A and 10B may transmit radio waves as the detection wave. Also in this case, the distance to the object, the position of the object, and the height that the object has at least can be detected from the reflected waves received by the distance sensors 10A and 10B.

[0086] In Embodiment 1, an example in which the positions of the distance sensors 10A and 10B on the Y-axis are different has been described. However, the distance sensors 10A and 10B may be arranged at different positions on the X-axis. In this case, it is preferable that the positions of the distance sensors 10A and 10B on the Y-axis and the positions on the Z-axis are the same. If there is a deviation between the positions of the distance sensors 10A and 10B on the Y-axis and the positions on the Z-axis, it is assumed that the amount of deviation is set within a preset range.

[0087] In Embodiments 1 to 5, an example in which the distance sensors 10A and 10B are provided on the front bumper of the vehicle A1 has been described. However, a pair of distance sensors may also be provided on the rear bumper of the vehicle A1.

[0088] In Embodiment 2, an example of setting the bandwidths BW1 and BW2 so as to satisfy Expression (9) has been described. This is to ensure that the bandwidth BW1 in the distance sensor 10A and the bandwidth BW2 in the distance sensor 10B do not overlap with each other within the range where the reception sensitivity decreases from the maximum value to half. However, if the resonance frequencies f01 and f02 are set so that the distance sensor 10A and the distance sensor 10B do not interfere with each other, it is not necessary to set the bandwidths BW1 and BW2 so as to satisfy Expression (9).

[0089] The present disclosure is not limited to the above-described embodiments, and can be implemented in various configurations without departing from the gist thereof. For example, the technical features in the embodiments corresponding to the technical features in each form described in the summary of the invention can be appropriately replaced or combined in order to solve some or all of the above-described problems or to achieve some or all of the above-described effects. Further, if the technical feature is not described as essential in this specification, it can be appropriately deleted.

Description of Reference Numerals

[0090] 10A, 10B... distance sensors, 100... obstacle detection device, 110... measurement unit, 120... position calculation unit

Claims

An obstacle detection device (100) that detects the position of an obstacle (M1) using a first distance sensor (10A) and a second distance sensor (10B) mounted on a vehicle (A1), The first distance sensor transmits a first detection wave having a first frequency, and receives a first reflected wave that is a reflected wave of the first detection wave reflected by the obstacle and has the first frequency, The second distance sensor transmits a second detection wave having a second frequency different from the first frequency, and receives a second reflected wave that is a reflected wave of the second detection wave reflected by the obstacle and has the second frequency, The obstacle detection device, A measurement unit (110) that measures a first distance from the first distance sensor to the obstacle from the first reflected wave received by the first distance sensor, and measures a second distance from the second distance sensor to the obstacle from the second reflected wave received by the second distance sensor; A position calculation unit (120) that obtains position coordinates indicating the position of the obstacle from the measured first distance and second distance and the distance between the first distance sensor and the second distance sensor; Comprising, When the first distance sensor receives the first reflected wave reflected by the obstacle, the second distance sensor receives the second reflected wave reflected by the same obstacle, and at least two maximum points are included in each of the signal waveforms representing the first reflected wave and the signal waveform representing the second reflected wave, Depending on the magnitude relationship of the reception intensities of at least two maximum points included in the signal waveform representing the first reflected wave received by the first distance sensor and the magnitude relationship of the reception intensities of at least two maximum points included in the signal waveform representing the second reflected wave received by the second distance sensor, a method for calculating an angle indicating the direction in which the obstacle exists is made different, Obstacle detection device.

2. The obstacle detection device according to claim 1, The resonance frequency of the first distance sensor is f01, and the resonance frequency of the second distance sensor is f02, which is lower than the resonance frequency of the first distance sensor, The frequency at which the reception sensitivity is reduced by 3 decibels from the reception sensitivity at the resonance frequency f01 of the first distance sensor. The lower frequency f11 is used as the lower limit frequency in the first bandwidth used by the first distance sensor, and the higher frequency f12 is used as the upper limit frequency in the first bandwidth. The frequency at which the reception sensitivity has decreased by 3 decibels from the reception sensitivity at the resonance frequency f02 of the second distance sensor, and when the lower frequency f21 is set as the lower limit frequency in the second bandwidth used by the second distance sensor and the higher frequency f22 is set as the upper limit frequency in the second bandwidth, {f01 + 1 / 2 · (f12 - f11)} ≤ {f02 - 1 / 2 · (f22 - f21)} The first bandwidth and the second bandwidth are set so that an obstacle detection device.

3. The obstacle detection device according to claim 1 or 2, wherein the first distance sensor is disposed at a position higher in the vehicle height direction than the second distance sensor, and the first frequency is set lower than the second frequency, an obstacle detection device.

4. The obstacle detection device according to any one of claims 1 to 3, wherein the measurement unit when a signal waveform representing a reflected wave includes a plurality of maximum points, calculates a time difference at which at least two adjacent maximum points in the combination occur for each combination of at least two adjacent maximum points on the time axis, and when the calculated time difference is equal to or less than a preset reference value, determines that the maximum points included in the combination are those included in the reflected wave from the same obstacle, an obstacle detection device.

5. The obstacle detection device according to claim 4, wherein the first distance sensor receives the first reflected wave and a third reflected wave, and determines that the height of the obstacle is not less than the difference between a value representing a position along the height direction in the position coordinates of the obstacle calculated from the first reflected wave and a value representing a position along the height direction in the position coordinates of the obstacle calculated from the third reflected wave, an obstacle detection device.

6. The obstacle detection device according to claim 5, wherein the second distance sensor receives the second reflected wave and a fourth reflected wave, When a value indicating the variation among a value representing the position along the traveling direction of the vehicle at the position coordinates of the obstacle calculated from the first reflected wave, a value representing the position along the traveling direction of the vehicle at the position coordinates of the obstacle calculated from the third reflected wave, a value representing the position along the traveling direction of the vehicle at the position coordinates of the obstacle calculated from the second reflected wave, and a value representing the position along the traveling direction of the vehicle at the position coordinates of the obstacle calculated from the fourth reflected wave exceeds a preset threshold value, it is determined that the reliability of the detection result regarding the height of the obstacle is low. Obstacle detection device. **Claim 7** The obstacle detection device according to any one of claims 1 to 6, wherein the first distance sensor receives the first reflected wave reflected by the obstacle, and the second distance sensor receives the second reflected wave reflected by the same obstacle, and the height of the obstacle is estimated by inputting information associating the intensity of the first reflected wave received by the first distance sensor with time and information associating the intensity of the second reflected wave received by the second distance sensor with time into a learned machine learning model. Obstacle detection device.

Images