Collision hazard detection device, program, and recording medium

JP7917328B2Active Publication Date: 2026-09-08SUBARU CORP
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
JP2022107329
Authority / Receiving Office
JP · JP
Patent Type
Patents
Current Assignee / Owner
Filing Date
2022-07-01
Publication Date
2026-09-08
Estimated Expiration
2042-07-01

AI Technical Summary

Benefits of technology

【0013】 本発明の1またはそれ以上の実施形態によれば、車両周辺にある物体の位置に影響されることなく、精度よく衝突危険物体を検出することができるという効果がある。

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure 0007917328000001
    Figure 0007917328000001
  • Figure 0007917328000002
    Figure 0007917328000002
  • Figure 0007917328000003
    Figure 0007917328000003
Patent Text Reader

Abstract

To provide a collision risk object detection device that detects a collision risk object with high accuracy without being affected by the position of an object that is present near a vehicle, and a program and a recording medium.SOLUTION: A collision risk object detection device comprises: a sonar sensor 20; an on-road stationary object detection unit 30 that detects an on-road stationary object S from an image at a rear side of a vehicle V, and calculates a position of the on-road stationary object S; a position estimation unit 60 that, when the on-road stationary object S has become undetectable by the on-road stationary object detection unit 30 due to backward traveling of the vehicle V, estimates the last position of the on-road stationary object S detected by the on-road stationary object detection unit 30 and a position of the on-road stationary object S on the basis of traveling speed information and steering angle information for the vehicle V, and calculates an angle θ between a central axis of the sonar sensor 20 and the position of the on-road stationary object S estimated; a correction unit 70 that corrects sensitivity reduction of a sensor output value on the basis of the angle θ; and a collision risk object determination unit 80 that determines whether or not the on-road stationary object S is a collision risk object on the basis of the sensor output value corrected by the correction unit 70.SELECTED DRAWING: Figure 1
Need to check novelty before this filing date? Find Prior Art

Description

[Technical Field]

[0001] The present invention relates to a collision dangerous object detection apparatus, a program, and a recording medium. [Background Art]

[0002] In recent years, driving support systems that support safe driving of vehicles while reducing the burden on vehicle drivers during vehicle operation have been put into practical use and widely spread. For example, a driving support system is known that detects an object present around a vehicle when the driver reverses the vehicle, and alerts the driver if the object is a collision dangerous object. For example, as a driving support system for detecting collision dangerous objects around a vehicle, a technology has been proposed that includes a camera for capturing images of the surroundings of the vehicle and a sonar sensor, and detects a collision dangerous object based on the image of the vehicle surroundings captured by the camera and the sensor output value of the sonar sensor (see, for example, Patent Document 1). [Prior Art Literature] [Patent Literature]

[0003] [Patent Document 1] Japanese Unexamined Patent Application Publication No. 2019-159380 [Summary of the Invention] [Problem to be Solved by the Invention]

[0004] Generally, compared to sonar sensors, cameras have inferior detection performance for objects at very close distances. Therefore, with respect to the position of the vehicle, objects located from medium range to short range are detected based on camera images, and objects located from short range to very close range are detected based on the sensor output of ultrasonic sonar.

[0005] However, since the detection sensitivity of a sonar sensor has directivity, there has been a problem that, for example, the detection accuracy of an object located in a direction close to the end of the viewing angle of the sonar sensor decreases.

[0006] Therefore, the present invention has been made in view of the above-mentioned problems, and aims to provide a collision hazard detection device, program, and recording medium that can accurately detect collision hazard objects without being affected by the position of objects around a vehicle. [Means for solving the problem]

[0007] Embodiment 1; One or more embodiments of the present invention propose a collision hazard detection device comprising: a road stationary object detection unit that detects road stationary objects from an image taken of the rear of a vehicle and calculates the position of the road stationary object; a sonar sensor that receives reflected waves from the road stationary object and outputs a sensor output value corresponding to the reflected waves; a position estimation unit that, when the road stationary object can no longer be detected by the road stationary object detection unit due to the vehicle moving in reverse, estimates the position of the road stationary object based on the last detected position of the road stationary object by the road stationary object detection unit, the vehicle's driving speed information, and steering angle information, and calculates an angle between the central axis of the sonar sensor and the estimated position of the road stationary object; a correction unit that corrects the decrease in sensitivity of the sensor output value based on the calculated angle; and a collision hazard determination unit that determines whether or not the road stationary object is a collision hazard based on the sensor output value corrected by the correction unit.

[0008] Embodiment 2; One or more embodiments of the present invention propose a collision-hazard object detection device in which the correction unit corrects the sensor output value based on the sensitivity correction amount of the sensor output value at the angle calculated by the position estimation unit.

[0009] Embodiment 3; One or more embodiments of the present invention propose a collision-hazard object detection device comprising a direction control unit that changes the orientation of the central axis of the sonar sensor, wherein the correction unit outputs a control amount to the direction control unit so that the orientation of the central axis of the sonar sensor becomes the angle, thereby correcting the decrease in sensitivity of the sensor output value.

[0010] Embodiment 4; One or more embodiments of the present invention propose a collision hazard object detection device in which the collision hazard object determination unit determines that a stationary object on the road is a collision hazard object when the sensor output value corrected by the correction unit is greater than a predetermined threshold.

[0011] Embodiment 5; One or more embodiments of the present invention are programs for causing a computer to execute a method for detecting a collision hazard object in a collision hazard object detection device comprising a road stationary object detection unit, a sonar sensor, a position estimation unit, a correction unit, and a collision hazard object determination unit, wherein the method comprises: a first step in which the road stationary object detection unit detects a road stationary object from an image captured from the rear of a vehicle and calculates the position of the road stationary object; a second step in which the sonar sensor receives reflected waves from the road stationary object and outputs a sensor output value corresponding to the reflected waves; and a second step in which the position estimation unit determines that the road stationary object cannot be detected by the road stationary object detection unit due to the vehicle's reverse movement. The proposed program causes a computer to execute a collision hazard object detection method, which comprises: a third step of estimating the position of the stationary object on the road based on the position of the last stationary object detected by the road stationary object detection unit, the vehicle's driving speed information, and steering angle information, and calculating the angle between the central axis of the sonar sensor and the estimated position of the stationary object on the road; a fourth step of the correction unit correcting the decrease in sensitivity of the sensor output value based on the calculated angle; and a fifth step of the collision hazard object determination unit determining whether or not the stationary object on the road is a collision hazard object based on the sensor output value corrected by the correction unit.

[0012] Embodiment 6; One or more embodiments of the present invention are a non-transient recording medium readable by a computer, which records a program for causing a computer to execute a method for detecting a collision hazard in a collision hazard object detection device comprising a road stationary object detection unit, a sonar sensor, a position estimation unit, a correction unit, and a collision hazard object determination unit, wherein the road stationary object detection unit performs a first step of detecting a road stationary object from an image captured of the rear of a vehicle and calculating the position of the road stationary object; the sonar sensor performs a second step of receiving reflected waves from the road stationary object and outputting a sensor output value corresponding to the reflected waves; and the position estimation unit performs a function that causes the road stationary object detection unit to be unable to detect the road stationary object due to the vehicle moving in reverse. When this occurs, the proposed non-transient recording medium readable by a computer contains a program for causing a computer to execute a collision hazard object detection method, which includes: a third step of estimating the position of the stationary object on the road based on the position of the stationary object last detected by the road stationary object detection unit, the vehicle's driving speed information, and steering angle information, and calculating an angle between the central axis of the sonar sensor and the estimated position of the stationary object on the road; a fourth step of the correction unit correcting the decrease in sensitivity of the sensor output value based on the calculated angle; and a fifth step of the collision hazard object determination unit determining whether or not the stationary object on the road is a collision hazard object based on the sensor output value corrected by the correction unit. [Effects of the Invention]

[0013] According to one or more embodiments of the present invention, there is an effect that collision hazards can be detected with high accuracy without being affected by the position of objects around the vehicle. [Brief explanation of the drawing]

[0014] [Figure 1] This figure shows the configuration of a collision-hazard object detection device according to the first embodiment of the present invention. [Figure 2] This figure shows the positions of the camera and sonar sensor of the collision-hazard object detection device according to the first embodiment of the present invention. [Figure 3]FIG. 1 is a diagram showing position information of a road stationary object detected by a road stationary object detection unit of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 4] FIG. 1 is a diagram showing position information of a road stationary object calculated by a position estimation unit of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 5] FIG. 1 is a diagram showing an angle calculated by a position estimation unit of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 6] FIG. 1 is a diagram showing detection sensitivity of a sonar sensor of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 7] FIG. 1 is a diagram showing a processing flow of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 8] FIG. 1 is a diagram showing a processing flow of the collision dangerous object detection apparatus according to the first embodiment of the present invention. [Figure 9] FIG. 1 is a diagram showing a configuration of a collision dangerous object detection apparatus according to a second embodiment of the present invention. [Figure 10] FIG. 1 is a diagram showing control of an orientation of a sonar sensor of the collision dangerous object detection apparatus according to the second embodiment of the present invention. [Figure 11] FIG. 1 is a diagram showing a processing flow of the collision dangerous object detection apparatus according to the second embodiment of the present invention. DESCRIPTION OF EMBODIMENTS

[0015] Hereinafter, embodiments of the present invention will be described with reference to FIG. 1 to FIG. 11.

[0016] <First Embodiment> The collision dangerous object detection apparatus 1 according to the present embodiment will be described with reference to FIG. 1 to FIG. 8.

[0017] <Configuration of Collision Dangerous Object Detection Apparatus 1> As shown in Figure 1, the collision hazard object detection device 1 according to this embodiment is configured to include an imaging unit 10, a sonar sensor 20, a road stationary object detection unit 30, a driving speed detection unit 40, a steering angle detection unit 50, a position estimation unit 60, a correction unit 70, a collision hazard object determination unit 80, and a control unit 90.

[0018] The imaging unit 10 is composed of, for example, a wide-angle monocular camera, and captures an image of the area behind the vehicle and transmits that image to the road stationary object detection unit 30, which will be described later. As shown in Figure 2, the imaging unit 10 is, for example, located in the rear center of the vehicle V and captures an image of the area behind the vehicle V. Furthermore, the imaging unit 10 may be configured to start capturing images of the area behind the vehicle when the driver reverses the vehicle V (for example, when shifting gears to the reverse position), and to transmit the captured images to the road stationary object detection unit 30.

[0019] The sonar sensor 20 receives reflected waves from stationary objects on the road and outputs a sensor output value corresponding to the reflected waves. Specifically, the sonar sensor 20 is composed of an ultrasonic sonar sensor and detects reflected waves from stationary objects S on the road within the dashed sector-shaped area shown in Figure 2, and outputs a sensor output value corresponding to the magnitude of the sound pressure of the detected reflected waves. The sonar sensor 20 transmits the detected sensor output value to the correction unit 70. Depending on the performance (detectable range) of the sonar sensor used, multiple sonar sensors may be installed at the rear of the vehicle V.

[0020] The road stationary object detection unit 30 detects road stationary objects S from the image of the area behind the vehicle V transmitted from the imaging unit 10, and calculates the position of the road stationary objects. Here, the road stationary object detection unit 30 detects stationary road stationary objects S, such as blocks or pillars, located behind the vehicle V, from the image transmitted from the imaging unit 10, and calculates the position of the road stationary objects S. Specifically, the road stationary object detection unit 30 detects road stationary objects S within the solid sector-shaped area shown in Figure 2 and calculates the position of the road stationary objects S.

[0021] Here, using Figure 3, we will explain in detail the position of the stationary road object S calculated by the road stationary road object detection unit 30. When the road stationary object detection unit 30 detects a road stationary object S behind the vehicle V, it calculates position information (X0, Y0) indicating the position of the road stationary object S. Specifically, the road stationary object detection unit 30 calculates the position information (X0, Y0) of the road stationary object S from the image transmitted from the imaging unit 10, based on the position where the imaging unit 10 (camera) is installed. More specifically, the road stationary object detection unit 30 calculates the position information (X0, Y0) of the road stationary object S in a plane formed by the camera's central axis (direction of the center of the image being captured) and a camera line perpendicular to the camera's central axis, with the position of the imaging unit 10 as the origin (0,0). Furthermore, the road stationary object detection unit 30 continuously calculates position information (X0, Y0) while detecting a road stationary object S from the image transmitted from the imaging unit 10. Furthermore, when the road stationary object detection unit 30 can no longer detect a road stationary object S that was previously detected in the image transmitted from the imaging unit 10, it transmits information to the position estimation unit 60, which will be described later, indicating that the road stationary object S can no longer be detected.

[0022] The vehicle speed detection unit 40 detects the vehicle speed information of the vehicle V and transmits it to the position estimation unit 60. Specifically, the driving speed detection unit 40 detects the driving speed of vehicle V based on, for example, the vehicle speed pulse signal of vehicle V as driving speed information.

[0023] The steering angle detection unit 50 detects the current steering angle of the steering wheel operated by the driver of the vehicle V and transmits the steering angle information to the position estimation unit 60. Specifically, the steering angle detection unit 50 detects the steering angle of the steering wheel based on the output of, for example, a rotation angle sensor installed on the steering wheel.

[0024] When the road stationary object detection unit 30 can no longer detect a road stationary object S due to the vehicle V moving in reverse, the position estimation unit 60 estimates the current position of the road stationary object S based on the position of the last detected road stationary object S in the road stationary object detection unit 30, the vehicle V's driving speed information, and the steering angle information, and calculates the angle θ between the central axis of the sonar sensor 20 and the estimated position of the road stationary object S. The position estimation unit 60 calculates the position information of the stationary object S on the road, with the position of the sonar sensor 20 as the origin, based on the position information of the stationary object S on the road, which is received from the stationary object detection unit 30 and is based on the position of the imaging unit 10. Specifically, as shown in Figure 4, the position estimation unit 60 uses the position of the sonar sensor 20 as the origin (0,0) and calculates the position information (X1, Y1) of a stationary object S on the road in a plane composed of the central axis of the sonar sensor and a sonar sensor line perpendicular to the central axis of the sonar sensor. The position estimation unit 60 calculates position information (X1, Y1) based on the placement position information (relative position information) of the imaging unit 10 and the sonar sensor 20, which are pre-stored in a memory (not shown), and the position information (X0, Y0) received from the road stationary object detection unit 30. Figure 5 illustrates a case where the position of a stationary object S on the road moves from P0 to P1 to P2 as the vehicle V moves in reverse. When the position estimation unit 60 receives information from the stationary object detection unit 30 that the stationary object S on the road can no longer be detected, it calculates the last calculated position information of the stationary object S on the road (X S , Y S Based on the vehicle speed information obtained from the vehicle speed detection unit 40 and the steering angle information of the steering wheel obtained from the steering angle detection unit 50, the current position information (X2, Y2) of the stationary object S on the road is calculated. In other words, the position estimation unit 60 uses the position information (X) of the last detected stationary object S on the road, P1. S , Y S Starting from the point, the position information (X2, Y2) of the current stationary object S on the road P2 is calculated based on the vehicle V's driving speed information and steering angle information. The position estimation unit 60 calculates the angle θ between the central axis of the sonar sensor 20 and the estimated position of the stationary object S on the road from the calculated position information (X2, Y2), and transmits the calculation result to the correction unit 70. Furthermore, if multiple sonar sensors 20 are located at the rear of the vehicle V, the position estimation unit 60 calculates position information (X2, Y2) and angle θ based on the sonar sensor that is closest in distance to the calculated position of the stationary object S on the road. Furthermore, if multiple sonar sensors 20 are installed, the position estimation unit 60 transmits information to the correction unit 70 that allows it to identify the sonar sensor used as a reference when estimating the position of a stationary object S on the road.

[0025] The correction unit 70 corrects the sensor output value of the sonar sensor 20 based on the sensitivity correction amount of the sensor output value at the angle θ calculated by the position estimation unit 60. The correction unit 70 transmits the corrected sensor output value to the collision hazard object determination unit 80. Furthermore, when the road stationary object detection unit 30 has detected a road stationary object S, the correction unit 70 transmits the sensor output value acquired from the sonar sensor 20 to the collision hazard object determination unit 80 without correcting the sensor output value of the sonar sensor 20.

[0026] Here, we will explain the correction of the sensor output value of the sonar sensor 20 in the correction unit 70. The sensor output value of the sonar sensor 20 has directionality. Figure 6 shows an example illustrating how the detection sensitivity of a sonar sensor changes with respect to the angle from the central axis of the sonar sensor, when the detection sensitivity at the central axis of the sonar sensor is set to 0 dB. Specifically, the sonar sensor 20 has a high detection sensitivity at its central axis, resulting in a large sensor output value. As the angle from the central axis of the sonar sensor increases, the detection sensitivity decreases, resulting in a smaller sensor output value. In other words, even if the distance between the sonar sensor 20 and the stationary object S on the road is the same, the magnitude of the sensor output value of the sonar sensor 20 changes depending on the angle between the central axis of the sonar sensor 20 and the position of the stationary object S on the road. Therefore, when the correction unit 70 receives information from the road stationary object detection unit 30 that the road stationary object S that was previously detected can no longer be detected, it corrects the sensor output value acquired from the sonar sensor 20 based on the sensitivity correction amount of the sensor output value at angle θ received from the position estimation unit 60. As shown in Figure 6, for example, if a stationary object S on the road is located 40 degrees away from the central axis of the sonar sensor 20 (angle θ = 40 degrees), the detection sensitivity of the sonar sensor 20 decreases by 6 dB compared to the detection sensitivity at the central axis of the sonar sensor 20. Therefore, the correction unit 70 performs a correction that adds +6dB to the sensor output value acquired from the sonar sensor 20. In this case, if multiple sonar sensors 20 are arranged behind the vehicle V, the correction unit 70 obtains the sensor output value from the sonar sensor that the position estimation unit 60 used as a reference when calculating the angle θ, and corrects the sensor output value.

[0027] The collision hazard object determination unit 80 determines whether or not a stationary object S on the road is a collision hazard object based on the sensor output value corrected by the correction unit 70. The collision hazard object determination unit 80 determines that a stationary object S on the road is a collision hazard object if the sensor output value corrected by the correction unit 70 is greater than a predetermined threshold. The collision hazard object determination unit 80 transmits the determination result to the control unit 90.

[0028] The control unit 90 controls the operation of the entire collision hazard object detection device 1 according to a control program stored in a ROM or the like (not shown). In this embodiment, the control unit 90, for example, when the collision hazard object determination unit 80 determines that a stationary object S on the road is a collision hazard object, outputs a warning sound or displays a warning to the driver of the vehicle V. Here, the control unit 90 may acquire the sensor output value of the sonar sensor 20, calculate the distance between the stationary object S on the road and the sonar sensor 20, and change the type of warning sound, the volume of the warning sound, the content of the warning display, etc., according to that distance. The details of the processing performed by the collision hazard detection device 1 are described below.

[0029] <Processing by collision hazard detection device 1> Figures 7 and 8 will be used to explain the details of the processing performed by the collision hazard object detection device 1.

[0030] It is determined whether or not vehicle V is moving in reverse (step S110). Specifically, for example, the control unit 90 determines whether or not the vehicle V is moving in reverse based on the position information of the shift lever. If the control unit 90 determines that the vehicle V is moving in reverse (YES in step S110), the process proceeds to step S120. On the other hand, if the control unit 90 determines that the vehicle V is not moving in reverse ("NO" in step S110), the process returns to step S110 and transitions to the standby state.

[0031] If it is determined that vehicle V is moving in reverse ("YES" in step S110), the road stationary object detection unit 30 determines whether or not a road stationary object S has been detected (step S120). If the road stationary object detection unit 30 determines that a road stationary object S has been detected (YES in step S120), it proceeds to step S130. On the other hand, if the road stationary object detection unit 30 determines that no road stationary object S has been detected ("NO" in step S120), it returns to step S110 and continues processing.

[0032] If the road stationary object detection unit 30 determines that a road stationary object S has been detected ("YES" in step S120), the position estimation unit 60 calculates the position information (X1, Y1) of the road stationary object S (step S130) and proceeds to step S140. The position estimation unit 60 stores the position information (X1, Y1) calculated in step S120 in a memory (not shown), and updates the position information stored in the memory each time it calculates the position information (X1, Y1) of a stationary object S on the road. In other words, the memory contains the last calculated position information (X S , Y S The value of ) is stored there.

[0033] The collision hazard object determination unit 80 determines whether the sensor output value received from the correction unit 70 is greater than a predetermined threshold (step S140). If the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is greater than a predetermined threshold (YES in step S140), it proceeds to step S150. On the other hand, if the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is smaller than a predetermined threshold ("NO" in step S140), it proceeds to step S160.

[0034] If the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is greater than a predetermined threshold ("YES" in step S140), the control unit 90 determines that the stationary object S on the road is a collision hazard object for the vehicle V, issues a warning to the driver of the vehicle V (step S150), and proceeds to step S160.

[0035] If the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is smaller than a predetermined threshold ("NO" in step S140), the road stationary object detection unit 30 determines whether or not a road stationary object S has been detected (step S160). In other words, the system determines whether the stationary road object S detected by the road stationary object detection unit 30 has moved outside the imaging area of ​​the imaging unit 10 due to the vehicle V moving in reverse. If the road stationary object detection unit 30 determines that a road stationary object S has been detected ("YES" in step S160), it returns to step S130 and continues processing. On the other hand, if the road stationary object detection unit 30 determines that no road stationary object S has been detected ("NO" in step S160), it proceeds to step S170.

[0036] If the road stationary object detection unit 30 determines that no road stationary object S has been detected ("NO" in step S160), the position estimation unit 60 calculates the current position information (X2, Y2) of the road stationary object S and calculates the angle θ based on that position information (step S170). In other words, when a stationary object S on the road, which was detected by the stationary object detection unit 30, moves outside the imaging area of ​​the imaging unit 10 due to the vehicle V moving in reverse, the position estimation unit 60 estimates the position information of the stationary object S on the road and calculates the angle θ. Furthermore, when the position estimation unit 60 receives information from the road stationary object detection unit 30 indicating that road stationary object S can no longer be detected, it stores in memory the position information (X) of the last detected road stationary object S in step S130. S , Y S Based on this, the current position information (X2, Y2) of the stationary object S on the road is calculated, and the angle θ is calculated from the position information (X2, Y2).

[0037] The correction unit 70 corrects the sensor output value received from the sonar sensor 20 based on the angle θ calculated in step S160 (step S180), and then proceeds to step S190. Specifically, the correction unit 70 corrects the sensor output value acquired from the sonar sensor 20 based on the sensitivity correction amount of the sensor output value of the sonar sensor at angle θ received from the position estimation unit 60.

[0038] The collision hazard object determination unit 80 determines whether the sensor output value received from the correction unit 70 is greater than a predetermined threshold (step S190). If the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is greater than a predetermined threshold (YES in step S190), it proceeds to step S200. On the other hand, if the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is smaller than a predetermined threshold ("NO" in step S190), it proceeds to step S210.

[0039] If the collision hazard object determination unit 80 determines that the sensor output value received from the correction unit 70 is greater than a predetermined threshold ("YES" in step S190), the control unit 90 will output a warning sound or display a warning about the risk of collision to the driver of the vehicle V (step S200), and then proceed to step S210.

[0040] The control unit 90 determines whether or not the vehicle V is moving in reverse (step S210). If the control unit 90 determines that the vehicle V is moving in reverse (YES in step S210), the process returns to step S170 and continues. On the other hand, if the control unit 90 determines that the vehicle V is not continuing to move in reverse ("NO" in step S210), the process is terminated.

[0041] <Effects and Actions> As described above, the collision hazard object detection device 1 according to this embodiment includes: a road stationary object detection unit 30 that detects a stationary object S on the road from an image of the rear of the vehicle V captured by the imaging unit 10 and calculates the position of the road stationary object S; a sonar sensor 20 that receives reflected waves from the road stationary object S and outputs a sensor output value corresponding to the reflected waves; a position estimation unit 60 that, when the road stationary object S can no longer be detected by the road stationary object detection unit 30 due to the vehicle V moving in reverse, estimates the position of the road stationary object S based on the position of the last road stationary object S detected by the road stationary object detection unit 30, the vehicle V's driving speed information and steering angle information, and calculates the angle θ between the central axis of the sonar sensor 20 and the estimated position of the road stationary object S; a correction unit 70 that corrects the decrease in sensitivity of the sensor output value based on the angle θ; a collision hazard object determination unit 80 that determines whether or not the road stationary object S is a collision hazard object based on the sensor output value corrected by the correction unit 70; and a control unit 90. In other words, when the correction unit 70 receives information from the road stationary object detection unit 30 that a road stationary object S that was previously detected can no longer be detected, it corrects the sensor output value acquired from the sonar sensor 20 based on the sensitivity correction amount of the sensor output value at angle θ received from the position estimation unit 60. As a result, the collision hazard object determination unit 80 determines whether or not a stationary object S on the road is a collision hazard object based on the sensor output value corrected by the correction unit 70. Therefore, even if a stationary object S on the road is outside the imaging area of ​​the imaging unit 10, it can be accurately determined whether or not it is a collision hazard object, thereby improving the collision safety of the vehicle. Furthermore, since the correction unit 70 corrects the sensor output value of the sonar sensor 20, the collision hazard object determination unit 80 can determine whether or not a stationary object S on the road is a collision hazard object using the same predetermined threshold, regardless of the position of the stationary object S on the road.

[0042] Furthermore, when the collision hazard object determination unit 80 determines that a stationary object S on the road is a collision hazard object, the control unit 90 outputs a warning sound and displays a warning to the driver about the risk of collision, thereby reducing the possibility of the vehicle V colliding with the stationary object S on the road.

[0043] Furthermore, even if there is a stationary object S on the road at the edge of the FOV (field of view) of the sonar sensor 20, the correction unit 70 corrects the sensor output value of the sonar sensor 20 based on the sensitivity correction amount at angle θ, so that the stationary object S on the road can be detected without being missed. This allows for an expanded detection range for a single sonar sensor, reducing the number of sonar sensors required for installation in vehicle V, thereby lowering the cost of the collision hazard object detection device 1. Furthermore, even when using an inexpensive sonar sensor with a narrow field of view, the sensor output value of the sonar sensor 20 can be corrected based on the angle θ, thereby achieving detection performance equivalent to that of a sonar sensor with a wide field of view. This allows the collision hazard object detection device 1 to be constructed using an inexpensive sonar sensor with a narrow field of view, thereby reducing the cost of the collision hazard object detection device 1.

[0044] <Second Embodiment> The collision-hazard object detection device 1A according to this embodiment will be described using Figures 9 to 11. Note that components that are denoted by the same reference numerals as those in the first embodiment have the same function, and therefore, a detailed explanation of them will be omitted. <Configuration of collision hazard object detection device 1A>

[0045] As shown in Figure 9, the collision hazard object detection device 1A according to this embodiment is configured to include an imaging unit 10, a sonar sensor 20, a road stationary object detection unit 30, a driving speed detection unit 40, a steering angle detection unit 50, a position estimation unit 60, a correction unit 70A, a collision hazard object determination unit 80, a control unit 90, and a direction control unit 100.

[0046] The correction unit 70A outputs a control amount to the direction control unit 100 (described later) so that the orientation of the central axis of the sonar sensor 20 becomes an angle θ, thereby correcting the decrease in sensitivity of the sensor output value. Specifically, the correction unit 70A sends an instruction to the direction control unit 100 to change the orientation of the sonar sensor 20 so that the orientation of the central axis of the sonar sensor 20 is at an angle θ. After the orientation of the central axis of the sonar sensor 20 is changed, the direction control unit 100 acquires the sensor output value from the sonar sensor 20. The correction unit 70A transmits the acquired sensor output value to the collision hazard object determination unit 80 as a corrected sensor output value. If multiple sonar sensors are installed, the correction unit 70A transmits identification information of the sonar sensor whose direction is to be changed to the direction control unit 100.

[0047] The direction control unit 100 changes the direction of the central axis of the sonar sensor 20. Specifically, the direction control unit 100 changes the orientation of the sonar sensor 20 based on the angle θ received from the correction unit 70A, so that the direction of the central axis of the sonar sensor 20 and the direction of the stationary object S on the road are the same. More specifically, as shown in Figure 10, the direction control unit 100 controls the orientation of the sonar sensor 20 so that the angle between the reference direction of the central axis of the sonar sensor 20 and the central axis of the sonar sensor 20 is θ. If multiple sonar sensors are installed, the direction control unit 100 changes the orientation of the sonar sensor as instructed by the correction unit 70A.

[0048] <Processing by collision hazard detection device 1A> Figure 11 will be used to explain the details of the processing performed by the collision hazard object detection device 1A. In addition, only step S300, which is a change from the processing of the collision-hazard object detection device 1 according to the first embodiment, will be described below.

[0049] The correction unit 70A sends an instruction to the direction control unit 100 to change the orientation of the central axis of the sonar sensor 20 to an angle θ. The direction control unit 100 then acquires the sensor output value of the sonar sensor 20 after the orientation of the central axis of the sonar sensor 20 has been changed (step S300).

[0050] <Effects and Effects> As described above, the collision hazard object detection device 1A according to this embodiment includes: a road stationary object detection unit 30 that detects a stationary object S on the road from an image of the rear of the vehicle V captured by the imaging unit 10 and calculates the position of the road stationary object S; a sonar sensor 20 that receives reflected waves from the road stationary object S and outputs a sensor output value corresponding to the reflected waves; and when the road stationary object detection unit 30 can no longer detect the road stationary object S due to the vehicle V moving in reverse, the position of the last road stationary object S detected by the road stationary object detection unit 30 and the movement of the vehicle V. The system includes a position estimation unit 60 that estimates the position of a stationary object S on the road based on speed information and steering angle information, and calculates the angle θ between the central axis of the sonar sensor 20 and the estimated position of the stationary object S on the road; a correction unit 70A that corrects the decrease in sensitivity of the sensor output value based on the angle θ; a collision hazard object determination unit 80 that determines whether or not the stationary object S on the road is a collision hazard object based on the sensor output value corrected by the correction unit 70A; a control unit 90; and a direction control unit 100 that controls the direction of the central axis of the sonar sensor 20. The correction unit 70A outputs a control amount to the direction control unit 100 so that the orientation of the central axis of the sonar sensor 20 becomes an angle θ, thereby correcting the decrease in sensitivity of the sensor output value. In other words, the correction unit 70A changes the direction of the central axis of the sonar sensor 20 to the direction of the position of the stationary object S on the road estimated by the position estimation unit 60, and acquires the sensor output value of the sonar sensor 20. Therefore, the correction unit 70A can acquire the sensor output value of the sonar sensor 20 at an angle in which the detection sensitivity of the sensor output value of the sonar sensor 20 does not decrease. As a result, the collision hazard object determination unit 80 determines whether or not a stationary object S on the road is a collision hazard object based on the sensor output value corrected by the correction unit 70A. Therefore, even if a stationary object S on the road is outside the imaging area of ​​the imaging unit 10, it can be accurately determined whether or not it is a collision hazard object, thereby improving the collision safety of the vehicle. Furthermore, the correction unit 70A controls the orientation of the sonar sensor 20 and acquires the sensor output value of the sonar sensor 20. Therefore, the collision hazard object determination unit 80 can determine whether or not a stationary object S on the road is a collision hazard object using the same predetermined threshold, regardless of the position of the stationary object S on the road.

[0051] The correction unit 70A acquires the sensor output value of the sonar sensor 20 only after the orientation of the central axis of the sonar sensor 20 has been changed to the direction of the assumed stationary object S on the road, even if there is a stationary object S on the road at the edge of the FOV (field of view) of the sonar sensor 20. This expands the detection range of a single sonar sensor and reduces the number of sonar sensors to be placed on the vehicle V. This will allow for a reduction in the cost of the collision hazard object detection device 1A. Furthermore, since the correction unit 70A acquires the sensor output value of the sonar sensor 20 only after the orientation of the central axis of the sonar sensor 20 has been changed to the direction of the assumed stationary object S on the road, the collision hazard object detection device 1A can be configured using an inexpensive sonar sensor with a narrow field of view. This will allow for a reduction in the cost of the collision hazard object detection device 1A.

[0052] <Example 1> The above-described road stationary object detection unit 30 is an example of a configuration that detects road stationary objects S from images captured by a wide-angle monocular camera. However, since it is sufficient to detect road stationary objects S and determine their location, road stationary objects S may also be detected using, for example, a stereo camera, Lidar, millimeter-wave radar, etc. This reduces the influence of weather conditions, ambient light, etc., when detecting the position of stationary objects S on the road, thereby improving the accuracy of the position of stationary objects S on the road. Furthermore, when the road stationary object detection unit 30 can no longer detect a road stationary object S, the position estimation unit 60 can calculate the current position information of the road stationary object S based on the improved position information of the road stationary object S, thereby improving the accuracy of the position of the road stationary object S.

[0053] <Modification 2> The collision hazard object detection devices 1 and 1A described above were devices that determined whether a stationary object S on the road was a collision hazard object when the vehicle V was moving in reverse. However, the imaging unit 10 and sonar sensor 20 may be provided at the front of the vehicle V, and the determination of whether a stationary object S is a collision hazard object may be made when the vehicle V is moving forward. This allows for the detection of collision hazards that are not visible to the driver, such as low-profile blocks, thereby improving the vehicle's collision safety.

[0054] <Variation 3> If the position of the stationary object S on the road assumed by the position estimation unit 60 is outside the range of angles that can be controlled by the direction control unit 100, the correction unit 70A may perform a correction by changing the direction of the central axis of the sonar sensor 20 and a correction of the sensor output value based on the sensitivity correction amount of the sensor output value at the angle θ between the central axis of the sonar sensor 20 and the assumed position of the stationary object S on the road, thereby correcting the decrease in sensitivity of the sensor output value. Specifically, the correction unit 70A sets the direction of the central axis of the sonar sensor 20 to the maximum angle that can be controlled by the direction control unit 100, and acquires the sensor output value of the sonar sensor 20. The correction unit 70A then calculates the angle θ1 between the central axis of the sonar sensor 20 in that state and the estimated position of the stationary object S on the road, and corrects the acquired sensor output value based on the sensitivity correction amount of the sensor output value at angle θ1. In other words, even if the position of the stationary object S on the road assumed by the position estimation unit 60 is outside the range of angles that can be controlled by the direction control unit 100, the correction unit 70A can correct the decrease in sensitivity of the sensor output value. As a result, the collision hazard object determination unit 80 determines whether or not a stationary object S on the road is a collision hazard object based on the sensor output value corrected by the correction unit 70A, thereby enabling accurate determination of whether or not a stationary object S on the road is a collision hazard object and improving the collision safety of the vehicle.

[0055] While embodiments of this invention have been described in detail above with reference to the drawings, the specific configuration is not limited to these embodiments and includes designs and the like that do not depart from the spirit of this invention. [Explanation of Symbols]

[0056] 1; Collision hazard detection device 1A; Collision Hazard Detection Device 10; Imaging Unit 20; Sonar sensor 30; Roadside stationary object detection unit 40; Driving speed detection unit 50; Rudder angle detection unit 60;Position estimation part 70; Correction section 70A; Correction section 80; Collision Hazard Object Determination Unit 90; Control Unit 100; Direction control unit V; Vehicle

Claims

1. A road stationary object detection unit detects stationary objects on the road from an image taken of the area behind the vehicle and calculates the position of the stationary objects on the road. A sonar sensor that receives reflected waves from a stationary object on the road and outputs a sensor output value corresponding to the reflected waves, When the road stationary object detection unit can no longer detect the road stationary object due to the vehicle moving in reverse, the position estimation unit estimates the position of the road stationary object based on the last detected position of the road stationary object in the road stationary object detection unit, the vehicle's travel speed information, and steering angle information, and calculates the angle between the central axis of the sonar sensor and the estimated position of the road stationary object. A correction unit that corrects the decrease in sensitivity of the sensor output value based on the calculated angle, A collision hazard determination unit determines whether or not the stationary object on the road is a collision hazard object based on the sensor output value corrected by the correction unit, A collision-hazard object detection device characterized by comprising the following features.

2. The collision hazard object detection device according to claim 1, characterized in that the correction unit corrects the sensor output value based on the sensitivity correction amount of the sensor output value at the angle calculated by the position estimation unit.

3. The sonar sensor is equipped with a direction control unit that changes the orientation of the central axis, The collision hazard object detection device according to claim 1, characterized in that the correction unit outputs a control amount to the direction control unit so that the orientation of the central axis of the sonar sensor becomes the angle, thereby correcting the decrease in sensitivity of the sensor output value.

4. The collision hazard object detection device according to claim 2 or 3, characterized in that the collision hazard object determination unit determines that the stationary object on the road is a collision hazard object when the sensor output value corrected by the correction unit is greater than a predetermined threshold.

5. A program for causing a computer to execute a method for detecting a collision hazard in a collision hazard detection device comprising a road stationary object detection unit, a sonar sensor, a position estimation unit, a correction unit, and a collision hazard object determination unit, The road stationary object detection unit performs a first step of detecting a road stationary object from an image taken of the area behind the vehicle and calculating the position of the road stationary object. The sonar sensor receives reflected waves from stationary objects on the road and outputs a sensor output value corresponding to the reflected waves in a second step, A third step in which, when the road stationary object detection unit can no longer detect the road stationary object due to the vehicle moving in reverse, the position estimation unit estimates the position of the road stationary object based on the position of the road stationary object last detected by the road stationary object detection unit, the vehicle's travel speed information, and steering angle information, and calculates the angle between the central axis of the sonar sensor and the estimated position of the road stationary object, The correction unit performs a fourth step of correcting the decrease in sensitivity of the sensor output value based on the calculated angle, A fifth step in which the collision hazard object determination unit determines whether or not the stationary object on the road is a collision hazard object based on the sensor output value corrected by the correction unit, A program that causes a computer to execute a method for detecting collision-hazardous objects, which includes the following features.

6. A non-transient recording medium readable by a computer, which contains a program for causing a computer to execute a method for detecting a collision hazard in a collision hazard object detection device comprising a road stationary object detection unit, a sonar sensor, a position estimation unit, a correction unit, and a collision hazard object determination unit, The road stationary object detection unit performs a first step of detecting a road stationary object from an image taken of the area behind the vehicle and calculating the position of the road stationary object. The sonar sensor receives reflected waves from stationary objects on the road and outputs a sensor output value corresponding to the reflected waves in a second step, A third step in which, when the road stationary object detection unit can no longer detect the road stationary object due to the vehicle moving in reverse, the position estimation unit estimates the position of the road stationary object based on the position of the road stationary object last detected by the road stationary object detection unit, the vehicle's travel speed information, and steering angle information, and calculates the angle between the central axis of the sonar sensor and the estimated position of the road stationary object, The correction unit performs a fourth step of correcting the decrease in sensitivity of the sensor output value based on the calculated angle, A fifth step in which the collision hazard object determination unit determines whether or not the stationary object on the road is a collision hazard object based on the sensor output value corrected by the correction unit, A computer-readable, non-transient recording medium containing a program for causing a computer to execute a method for detecting collision-hazardous objects, which includes the following features.

Citation Information

Patent Citations

  • Method of detecting shift amount of detecting center axis

    JP2008122292A

  • Parking assist system

    JP2015105070A

  • Driving assistance device

    JP2017182300A

  • Object detection device

    JP2018072105A

  • Object detection device, object detection method, and program

    JP2019159380A