Driver assistance device

Abstract

Driver assistance device (20) with: an object detection unit (22) which detects one or more objects behind its own vehicle by using a reflected wave resulting from a search wave and identifies in the one or more detected objects an object approaching the own vehicle as an approaching object; a reference unit (21) which refers to a speed of the own vehicle and a relative speed and azimuth of each of the one or more objects detected by the object detection unit in relation to the own vehicle; a phantom detection unit (40) that determines whether the approaching object is a phantom target that is not actually present; and an assistance determination unit (23) which performs a driving assistance of the own vehicle to avoid danger to the own vehicle from a target object and excludes the approaching object, which is determined by the phantom determination unit as the phantom target, from the target object, wherein the reference unit is configured to reference an initial value of the relative velocity of the approaching object in relation to the vehicle itself; the phantom determination unit is configured to: Calculating a second value of the relative speed of the approaching object with respect to the own vehicle according to the speed of the own vehicle and a value of the relative speed of a following object with respect to the own vehicle, wherein the following object is detected by the object detection unit and is located between the own vehicle and the approaching object; Calculating an absolute difference between the first value of the relative velocity of the approaching object and the second value of the relative velocity of the approaching object as a velocity difference; Calculating an absolute difference between the azimuth of the following object and the azimuth of the approaching object, as measured from the reference unit, as an azimuth difference; and Determine whether the approaching object is the phantom target, based on the velocity difference and the azimuth difference.

Description

[Reference to related registrations]

[0001] This application is based on Japanese patent application No. 2018-245839, filed on December 27, 2018; full reference is made here to the disclosure content therein. [Technical field]

[0002] The present disclosure relates to a driver assistance device that performs driver assistance based on object detection information on an object behind a vehicle. [State of the art]

[0003] JP 2014 - 71 012 A describes a system that transmits electromagnetic waves towards the front of the vehicle and receives reflected waves resulting from those electromagnetic waves to detect another vehicle approaching the vehicle and issue a notification. For example, if the vehicle is traveling in an enclosed or covered space, such as under a viaduct or in a tunnel, and reflected waves from another vehicle are further reflected by road structures, in some cases an object (phantom object) will be detected that is not actually present. Patent literature 1 states that if an object detected using reflected waves is determined to be a phantom object, then the phantom object is excluded from being an object for which a notification is required.

[0004] DE 199 47 593 A1 describes a radar device for a vehicle with a radar beam scanning mechanism for scanning an area parallel to a road surface on which the vehicle is traveling, and an object detector for receiving reflected waves to create a detected object image. The object detector is equipped with a ghost echo detection device for determining whether a generated detected object image is a ghost echo in order to eliminate it.

[0005] US 2010 O 214 153 A1 describes a device for assessing ghost target candidates. It determines whether one of two targets is a candidate for a ghost target when the distances of the two targets from a target vehicle are equal and when the difference between the received signal levels of the two targets is not less than a predetermined value, in which case the ghost target data is removed. Conversely, it assesses whether a target is the one traveling alongside the other (i.e., side-by-side targets) based on changes in the angular difference and the left-right position difference between the two targets, corresponding to changes in the distances of the two targets from the vehicle in question. [Summary of the invention]

[0006] Patent literature 1 describes a technique for reliably ruling out a phantom object that is difficult to identify, but does not describe a technique for preventing an actually approaching object from being mistakenly identified as a phantom object. If an actually approaching object is mistakenly identified as a phantom object and no notification is given, the driving safety of the vehicle cannot be guaranteed.

[0007] In view of the foregoing, it is an object of the present invention to provide a driving assistance device that is able to distinguish between an actually approaching object and a phantom object with higher accuracy.

[0008] The present invention comprises: an object detection unit that detects one or more objects behind its own vehicle by using a reflected wave resulting from a search wave and identifies an object approaching the own vehicle among the one or more detected objects; a reference unit that relates a speed of the own vehicle and a relative speed and azimuth of each of the one or more objects detected by the object detection unit with respect to the own vehicle; a phantom determination unit that determines whether the approaching object is a phantom target that is not actually present;and an assistance determination unit that performs a driving assistance function of the own vehicle to avoid hazards to the own vehicle from a target object and excludes the approaching object, which is determined by the phantom determination unit as the phantom target, from the target object, wherein the reference unit is configured to reference a first value of the relative speed of the approaching object with respect to the own vehicle; the phantom determination unit is configured to: calculate a second value of the relative speed of the approaching object with respect to the own vehicle according to the speed of the own vehicle and a value of the relative speed of a following object with respect to the own vehicle, wherein the following object is detected by the object detection unit and is located between the own vehicle and the approaching object;Calculate an absolute difference between the first value of the relative velocity of the approaching object and the second value of the relative velocity of the approaching object as a velocity difference; calculate an absolute difference between an azimuth of the following object and an azimuth of the approaching object, referenced to the unit of reference, as an azimuth difference; and determine whether the approaching object is the phantom target, based on the velocity difference and the azimuth difference.

[0009] According to the present invention, the reference unit is configured to obtain a first value of the relative velocity of the approaching object with respect to the vehicle. The phantom detection unit is configured to: calculate a second value of the relative velocity of the approaching object with respect to the vehicle, based on the vehicle's own speed, and a value of the relative velocity of a following object with respect to the vehicle, wherein the following object is detected by the object detection unit and is located between the vehicle and the approaching object; and calculate an absolute difference between the first value of the relative velocity of the approaching object and the second value of the relative velocity of the approaching object as a velocity difference.Furthermore, the phantom detection unit is configured to calculate an absolute difference between the azimuth of the following object and the azimuth of the approaching object, as measured by the reference unit. Finally, the phantom detection unit is configured to determine whether the approaching object is the phantom target, based on the velocity difference and the azimuth difference.The phantom detection unit calculates a velocity difference, an absolute difference between the detected relative velocity of an approaching object and the relative velocity of the approaching object, theoretically obtained under the assumption that the approaching object is a phantom target. It also calculates an azimuth difference, an absolute difference between the azimuth of the detected approaching object and the azimuth of the approaching object, theoretically obtained under the assumption that the approaching object is a phantom target. Based on both the velocity difference and the azimuth difference, the phantom detection unit then determines whether the approaching object is a phantom target. This allows for a more accurate differentiation between a real approaching object and a phantom object. [Brief description of the drawings]

[0010] The previous task and other tasks, features, and advantages of the present disclosure will become clearer from the following detailed description with reference to the accompanying drawings. In the drawings: Fig. Figure 1 shows a driver assistance system according to an exemplary embodiment; Fig. Figure 2 is a diagram showing a velocity condition and an azimuth condition for phantom determination according to the exemplary embodiment; Fig. Figure 3 is a frequency distribution diagram for the relative velocity of a phantom target; Fig. Figure 4 shows an example of an azimuth difference threshold value, which is determined based on a vehicle speed Vs and a distance R between a following vehicle and the vehicle itself; Fig. Figure 5 is a diagram showing the determination of the azimuth difference threshold based on the distance R between the following vehicle and the own vehicle; Fig. Figure 6 shows a relationship between the distance R between the following vehicle and the own vehicle and an azimuth difference between the following vehicle and the phantom target; and Fig. Figure 7 is a flowchart of a driver assistance process according to the exemplary embodiment. [Description of the embodiment examples](First embodiment example)

[0011] As in Fig. Figure 1 shows a driver assistance system 10 according to an embodiment comprising a radar device 11, a vehicle speed sensor 12, an ECU 20 and an alarm device 30.

[0012] The radar device 11 is, for example, a known millimeter-wave radar that transmits a high-frequency signal in a millimeter-wave band. The vehicle can be equipped with a single radar device 11 or with multiple radar devices 11. The radar device 11 is, for example, mounted on a rear end section or the like of the vehicle, defines an area within a predetermined detection angle range as a detection area in which object detection can be carried out, and detects the position of an object within the detection area.

[0013] In particular, the radar device transmits 11 search waves in a predetermined cycle and receives reflected waves resulting from the search waves using a plurality of antennas. From the time at which the search waves are transmitted and the time at which the reflected waves are received, the distance to an object can be calculated.

[0014] Furthermore, a relative velocity is calculated from the frequency of the waves reflected by the object, which is altered by the Doppler effect. Additionally, the object's azimuth can be calculated from the phase difference between the reflected waves received by the majority of antennas. If the object's position and azimuth can be calculated, its relative position to the vehicle can be determined.

[0015] The radar device 11 is an example of an environmental monitoring device that transmits search waves, detects reflected waves received from the search waves reflected by an object around the vehicle, and obtains information about the vehicle's surroundings. As an environmental monitoring device that uses search waves, the driver assistance system 10 may, in addition to the radar device 11, include an ultrasonic sensor, a LIDAR (light detection and ranging / laser imaging detection and ranging) device, and the like.

[0016] A sensor that transmits search waves, such as a millimeter wave radar like radar device 11, a sonar or a LIDAR, sequentially outputs a sampling result as acquisition information to the ECU 20 based on a received signal obtained when the sensor receives reflected waves through an obstacle.

[0017] The vehicle speed sensor 12 is a sensor that detects the vehicle's speed and is not limited to, for example, a wheel speed sensor capable of detecting the rotational speed of a wheel. The wheel speed sensor used as the vehicle speed sensor 12 is, for example, attached to a section of the wheel that is not a tire and outputs a wheel speed signal to the ECU 20 corresponding to the vehicle's wheel speed.

[0018] The driver assistance system 10 can further include various sensors, such as an imaging device, a steering angle sensor, a yaw rate sensor, and a GNSS (global navigation satellite system) receiver, such as a GPS receiver. The ECU 20 can be capable of receiving a detection signal from such sensors.

[0019] The alarm device 30 is a device for providing a notification to a driver or the like and can, for example, be a device that provides an audible notification such as a speaker or a buzzer, or a device that provides a visual notification such as a display or indicator located in the vehicle's cabin, but the alarm device 30 is not limited to such a device. The alarm device 30 emits an alarm tone or the like based on a control command from the ECU 20 to notify, for example, the driver that there is a risk of collision with an object.

[0020] The ECU 20 includes a data reference unit 21, an object detection unit 22, an assistance determination unit 23, and a phantom determination unit 40. The ECU 20 contains a CPU, a ROM, a RAM, an I / O interface, and the like, and implements each of the functions of the above units by using the CPU to execute a program installed in the ROM. Thus, based on information obtained from various sensors, such as the radar device 11 and the vehicle speed sensor 12, the ECU 20 generates a control command to the alarm device 30 and outputs the control command, enabling the alarm device to function as a driver assistance device that performs the vehicle's own driver assistance functions.

[0021] The data reference unit 21 obtains a relative velocity and azimuth of an object with respect to the vehicle from the radar device 11. Furthermore, the data reference unit 21 obtains a driving speed of the vehicle from the vehicle speed sensor 12.

[0022] The object detection unit 22 detects an object behind the vehicle using information about reflected waves received by the radar device 11. The type of object is not particularly limited and can include a vehicle, bicycle, motorcycle, pedestrian, animal, structure, and the like, and can be moving or stationary. If the ECU 20 has an environmental monitoring device other than the radar device 11, such as an imaging device, the environmental monitoring device can be used in combination with the radar device 11 to detect an object. Furthermore, the object detection unit 22 detects an approaching object from the detected objects.The object detection unit 22 can determine whether an object is an approaching object by using the relative velocity of the object in relation to its own vehicle.

[0023] The phantom detection unit 40 determines whether an approaching object, detected by the object detection unit 22 as approaching the vehicle, is a phantom target that is not actually present. The phantom detection unit 40 contains a velocity detection unit 41 and an azimuth detection unit 42. The velocity detection unit 41 determines whether a velocity condition for a phantom target is met. The velocity condition is a condition that the difference between a relative velocity of a detected approaching object and a relative velocity of the approaching object, theoretically obtained under the assumption that the approaching object is a phantom target, lies within a predefined range. The azimuth detection unit 42 determines whether an azimuth condition for a phantom target is met.The azimuth condition is a condition that the difference between the azimuth of a detected approaching object and the azimuth of the approaching object, theoretically obtained under the assumption that the approaching object is a phantom target, lies within a predefined range. If an approaching object detected by the object detection unit 22 satisfies both the velocity condition and the azimuth condition for a phantom target, the phantom determination unit 40 determines that the detected approaching object is a phantom target.

[0024] A process carried out by the Phantom Determination Unit 40 is described in relation to Fig. 2 described. As in Fig. As shown in Figure 2, a vehicle 50 and a following vehicle 51 are traveling on a road 60. A lane marking 61 is provided in the middle of the road 60, and a portion of the road 60 to the left of the lane marking 61 is a lane 62, while a portion of the road 60 to the right of the lane marking 61 is a lane 63. A road structure 57 is present between the vehicle 50 and the following vehicle 51. The road structure 57 is, for example, a structure provided on the road, such as a road sign, a metal structure such as a tunnel roof, or a sign. Fig. 2(a) shows the vehicle 50 and the like as seen from the right side, and 2(b) shows the vehicle 50 as seen from above.

[0025] The vehicle 50 travels forward on lane 62 at a vehicle speed Vs, which is the speed (base speed) of the vehicle 50 relative to the road 60. The vehicle speed Vs is detected by the vehicle speed sensor 12 and obtained from the data reference unit 21.

[0026] The following vehicle 51 is a vehicle traveling directly behind the own vehicle 50 and is an example of a following object that follows the own vehicle 50. The following vehicle 51 travels behind the own vehicle 50 in the same direction as the own vehicle 50 on lane 62, in which the own vehicle 50 is traveling. The relative speed of the following vehicle 51 with respect to the own vehicle 50 is expressed as Vrv and is detected by the radar device 11 and obtained by the data reference unit 21.

[0027] If search waves transmitted rearward from the radar device 11 located at a rear end position P50 of the own vehicle 50 are reflected at a detection point P51 located at a front section of the following vehicle 51, and are also further reflected by the road structure 57, the object detection unit 22 of the own vehicle 50 detects an approaching object 55 as a phantom target due to the reflected waves resulting from the aforementioned reflection. The approaching object 55 is detected at a position further behind the following vehicle 51.

[0028] A relative velocity Vg0 (calculated value) of the approaching object 55 with respect to the own vehicle 50 can be calculated, assuming that the approaching object 55 is a phantom target, by the following formula (1) using the relative velocity Vrv of the following vehicle 51 with respect to the own vehicle 50 and the own vehicle velocity Vs. Vg0=Vs+2Vrv

[0029] In contrast, the data reference unit 21 obtains a relative velocity Vg of the approaching object 55, which is detected by the object detection unit 22, with respect to the own vehicle 50 from the radar device 11. The speed determination unit 41 calculates as a velocity difference dVg an absolute difference between the relative velocity Vg (detected value) of the approaching object 55 with respect to the own vehicle and the calculated value Vg0 based on the following formula (2). dVg=|Vg−Vg0|=|Vg−(Vs+2Vrv)|

[0030] As dVg decreases, the relative velocity Vg of the approaching object 55 with respect to the vehicle 50 approaches the calculated value Vg0. The velocity determination unit 41 sets a predefined velocity difference threshold Vc1 and determines whether the velocity difference dVg is less than this threshold. If dVg < Vc1, the velocity determination unit 41 determines that the approaching object 55 meets the velocity condition for a phantom target. If dVg ≥ Vc1, the velocity determination unit 41 determines that the approaching object 55 does not meet the velocity condition for a phantom target.

[0031] The ease of detecting a phantom target has a trade-off with preventing a false detection, where a real approaching object is mistakenly identified as a phantom target. If Vc1 is set to a large value, a phantom target is easily detected, but there is a high probability that a real approaching object (an approaching object that is actually present) will be mistakenly identified as a phantom target. If Vc1 is set to a small value, there is a low probability that a real approaching object will be mistakenly identified as a phantom target, but a phantom target is less easily detected. The value of Vc1 can be set, for example, based on the detection accuracy of the sensors mounted on the vehicle 50, such as the radar device 11 and the vehicle speed sensor 12.

[0032] The velocity difference threshold Vc1 can be set based on an actual measured value. For example, the velocity difference threshold Vc1 can be set to include a predetermined proportion of a set of many relative velocities Vg obtained by the radar device 11 through the data reference unit 21.

[0033] In particular, for example, as in Fig. Figure 3 shows how a frequency distribution diagram is created where the horizontal axis represents the variation of the relative velocity Vg and the vertical axis represents the frequency of the relative velocity Vg. Then, an upper standard limit Vgu and a lower standard limit Vgb are set such that a predetermined proportion (e.g., 95%) of the total set of relative velocities Vg is included, and the median value (a position where the value on the horizontal axis is zero) of the relative velocities Vg is set in the middle. By setting a difference between the upper standard limit Vgu and the lower standard limit Vgb as Vc1, the velocity difference threshold Vc1 can be set to include a predetermined proportion (e.g., 95%) of relative velocities Vg.The percentage can be changed based on the surrounding environment in which the vehicle is driving, the driving condition of the vehicle, or similar factors.

[0034] As in Fig. 2(b) is shown when search waves from the radar device 11, which is provided at the rear end position P50 of the own vehicle 50, are reflected at the detection point P51, which is located at the right front end of the following vehicle 51, the approaching object 55, which is detected as a phantom target, is detected at a position of a detection point P55, which is located on the rear right side of the following vehicle 51.

[0035] The radar device 11 obtains the azimuth θrv of the following vehicle 51 relative to the own vehicle 50 and the azimuth θg of the approaching object 55 relative to the own vehicle 50 via the data reference unit 21. The azimuth θrv is determined as an angle formed by a central axis L1 of the own vehicle, which is a central axis in the lateral direction of the own vehicle 50, and a line segment connecting the rear end position P50 and the detection point P51. The azimuth θg is determined as an angle formed by the central axis L1 of the own vehicle and a line segment connecting the rear end position P50 and the detection point P55.

[0036] The azimuth determination unit 42 calculates as an azimuth difference dθg an absolute difference between the azimuth θrv of the following vehicle 51 in relation to the own vehicle 50 and the azimuth θg of the approaching object 55 in relation to the own vehicle 50 based on the following formula (3). dθg=|θrv−θg|

[0037] If the approaching object 55 is a phantom target, the acquisition point P55 of the approaching object 55 lies on the line segment connecting the rear end position P50 and the acquisition point P51; thus, dθg = 0. The azimuth determination unit 42 sets a predefined azimuth difference threshold θc and determines whether the azimuth difference dθg is less than the azimuth difference threshold θc. If dθg < c, the azimuth determination unit 42 determines that the approaching object 55 meets the azimuth condition for a phantom target. If dθg ≥ θc, the azimuth determination unit 42 determines that the approaching object 55 does not meet the velocity condition for a phantom target.

[0038] If θc is set to a large value, a phantom target will be easily detected, but there is a high probability that a real approaching object will be falsely identified as a phantom target. If θc is set to a small value, there is a low probability that a real approaching object will be falsely identified as a phantom target, but a phantom target will be less easily detected. The value of θc can be set, for example, based on the detection accuracy of the sensors mounted on the vehicle 50, such as the radar device 11 and the vehicle speed sensor 12.

[0039] The azimuth determination unit 42 can change the azimuth difference threshold θc based on the vehicle's own speed Vs or a distance R between the vehicle 50 and the following vehicle 51. For example, as in Fig. As shown in Figure 4, if the vehicle speed Vs is less than a predetermined vehicle speed threshold Vc2, the azimuth determination unit 42 sets the azimuth difference threshold θc to a predetermined fixed value θc1. If the vehicle speed Vs is greater than or equal to the predetermined vehicle speed threshold Vc2, the azimuth determination unit 42 changes the azimuth difference threshold θc according to the distance R between the vehicle 50 and the following vehicle 51. In particular, the azimuth determination unit 42 sets the azimuth difference threshold θc lower as the distance R between the vehicle 50 and the following vehicle 51 increases. Fig. 4 is a relationship between the distance thresholds Rc1 to Rc3 for the distance R: Rc1 < Rc2 < Rc3. It is also a relationship between predefined fixed values ​​c1 to c5 for the azimuth difference threshold θc: θc1<θc5<θc4<θc3<θc2.

[0040] The distance R can be calculated, for example, from detection information acquired by the radar device 11. In particular, the radar device 11 transmits search waves in a predetermined cycle and receives reflected waves through the plurality of antennas. From the time at which the search waves are transmitted and the time at which the reflected waves are received, a distance to an object can be calculated.

[0041] At the in Fig. In the procedure shown in Figure 4 for setting the azimuth difference threshold c, an approach object velocity Vgr, which is a velocity (base velocity) of the approaching object 55 relative to the road 60, is estimated using the vehicle's own speed Vs, and the probability that the approaching object 55 is a phantom target is estimated based on the approach object velocity Vgr to set the azimuth difference threshold θc. Based on formula (1), the approach object velocity Vgr can be calculated using the following formula (4). Vgr=Vs+Vg0=2Vs+2Vrv

[0042] For simplification, an example is described in which the following vehicle 51 travels at the same base speed as the own vehicle 50 on lane 62, and the relative speed Vrv of the following vehicle 51 with respect to the own vehicle 50 is zero (Vrv = 0). Formula (4) calculates the approaching object speed Vgr to be twice the own vehicle speed Vs (Vgr = 2Vs).

[0043] If the vehicle's own speed Vs is 20 km / h and 40 km / h, the approaching object's speed Vgr is calculated as 40 km / h and 80 km / h, respectively. A speed Vgr of approximately 40 to 80 km / h is a typical speed for a vehicle traveling on a typical road. In other words, the probability that the approaching object 55, traveling at 40 to 80 km / h, is a vehicle can be considered relatively high.

[0044] If, on the other hand, the vehicle's own speed Vs is 60 km / h, 80 km / h, and 100 km / h, the approaching object's speed Vgr is calculated as 120 km / h, 160 km / h, and 200 km / h, respectively. A speed Vgr of approximately 120 to 200 km / h is not a typical speed for a vehicle traveling on a typical road. In other words, the probability that the approaching object 55, traveling at 120 to 200 km / h, is a vehicle can be considered relatively low.

[0045] Based on the previous estimate, the vehicle speed threshold Vc2 is set at a value of approximately 40 to 60 km / h. Fig. 4. If Vs < Vc2, especially if the vehicle's own speed Vs is approximately 20 to 40 km / h, which is low, the probability that a detected approaching object 55 is a vehicle or the like is relatively high, and accordingly, the azimuth difference threshold θc is set small. Specifically, the azimuth difference threshold θc is set to the fixed value θc1, which is set to a small value. This primarily reduces the probability that the approaching object 55 is incorrectly identified as a phantom target, even though the approaching object 55 is a real approaching object.If Vs < Vc2, setting the azimuth difference threshold θc to a small value allows, in particular, a very low probability that, although the approaching object 55 is, for example, a motorcycle approaching the own vehicle 50 while passing through a space between the vehicles, the approaching object 55 will be falsely identified as a phantom target.

[0046] If Vs ≥ Vc2, especially if the vehicle's own speed Vs is approximately 60 to 100 km / h, which is high, the probability that a detected approaching object 55 is a vehicle or the like is relatively low, and accordingly, the azimuth difference threshold θc is set greater than θc1. Thus, priority is given to achieving a simple identification where the approaching object 55 is easily identified as a phantom target.

[0047] Furthermore, if Vs ≥ Vc2, as in Fig. Figure 4 shows that the azimuth difference threshold θc is set smaller when the distance R between the own vehicle 50 and the following vehicle 51 increases, and the azimuth difference threshold θc is set larger when the distance R decreases.

[0048] As in Fig. As shown in Figure 5, with respect to the following vehicle 51, whose distance from the own vehicle 50 is R1, the maximum azimuth θ1 of the following vehicle 51 relative to the own vehicle 50 is expressed as an angle A1-O-B1. Furthermore, with respect to a following vehicle 52, whose distance from the own vehicle 50 is R2, the maximum azimuth θ2 of the following vehicle 52 relative to the own vehicle 50 is expressed as an angle A2-O-B2. The distance R1 is less than the distance R2 (R1 < R2), and the maximum azimuth θ1 is greater than the maximum azimuth θ2 (θ1 > θ2). If a following vehicle is the following vehicle 51, whose distance from the own vehicle 50 is R1, a phantom target with a maximum azimuth θ1 or less is detected. If a following vehicle is the following vehicle 52, whose distance from the own vehicle 50 R2 is, a phantom target with the maximum azimuth θ2 or less is detected.As seen from . Fig. As becomes clear in section 5, a phantom target is detected at a position that deviates more from the position of the own vehicle 50 in the direction towards the side of the own vehicle 50 when the distance R becomes smaller.

[0049] In relation to Fig. 2. If the distance R between the vehicle 50 and the following vehicle 51 is small, the approaching object 55 is located further to the right of lane 62 and is detected at a position closer to lane 63. Therefore, the azimuth difference threshold θc is set to a large value.

[0050] If the distance R between the vehicle 50 and the following vehicle 51 is large, the approaching object 55 will be detected at a position relatively close to the lane 62. Therefore, the azimuth difference threshold θc is set low, thus prioritizing the reduction of the probability that, although the approaching object 55 is a real approaching object, it will be falsely identified as a phantom target.

[0051] The vehicle speed threshold Vc2, the distance thresholds Rc1 to Rc3, the specified fixed values ​​θc1 to θc5 for the in Fig. The azimuth difference threshold θc shown in Figure 4 can be changed according to the lane width, the basic speed of the following vehicle 51, or the like.

[0052] The distance thresholds Rc1 to Rc3 and the predetermined fixed values ​​θc1 to θc5 for the azimuth difference threshold θc can be determined based on actual measured values. For example, with respect to actual measured values ​​obtained from the radar device 11 by the data reference unit 21, as in Fig. Figure 6 shows a distribution diagram created in which the horizontal axis represents the distance R between the own vehicle 50 and the following vehicle 51, and the vertical axis represents the azimuth difference by plotting actual measured values ​​of the azimuth difference between the following vehicle 51 and a phantom target, which is detected when the distance between the own vehicle 50 and the following vehicle 51 is R. Based on the distribution diagram, the azimuth difference threshold θc can be set to include a predetermined proportion (e.g., 95%) of the total set of azimuth differences between the following vehicle and the phantom target.

[0053] In Fig. For example, curve 71 and curve 72 will be obtained, such that all actually measured values ​​between curves 71 and 72 are included. Then the azimuth difference threshold θc can be set such that θc is reduced with increasing R according to curves 71 and 72. Alternatively, in Fig. 6. A stepped rectangular line 73 and a stepped rectangular line 74 are obtained such that a predetermined proportion (e.g., 95%) of actually measured values ​​is contained between the stepped lines 73 and 74. Then the azimuth difference threshold θc can be set such that θc is gradually reduced with increasing R according to the rectangular lines 73 and 74. The proportion of actually measured values ​​can be changed based on the surrounding environment in which the vehicle 50 is driving, the driving condition of the vehicle 50, or the like.

[0054] The previous threshold values ​​can be stored in the ECU 20 as they are or in the form of a mathematical formula, a data table, or the like.

[0055] The assistance determination unit 23 performs a driving assistance function for the vehicle 50 to prevent a collision with a target object and excludes an approaching object, identified by the phantom determination unit 40 as a phantom target, from the target object. Even if an approaching object 55 is detected by the object detection unit 22, and the approaching object 55 is identified by the phantom determination unit 40 as a phantom target, the approaching object 55 does not impair the driving safety of the vehicle 50. Therefore, the assistance determination unit 23 does not perform any driving assistance functions, such as notification, collision avoidance, or damage reduction, for the approaching object 55.If an approaching object 55 is detected by the object detection unit 22 and the phantom determination unit 40 determines that the approaching object 55 is not a phantom target, the approaching object 55 could impair the driving safety of the vehicle 50. Therefore, the assistance determination unit 23 performs a driving assistance action, such as issuing a notification, avoiding a collision, or reducing damage to the approaching object 55. For example, a control command to emit an alarm tone is sent to the alarm device 30.

[0056] A driver assistance process carried out by the ECU 20 is described in relation to Fig. 7 described. The ECU 20 repeatedly executes the command described in Fig. The driver assistance process shown in 7 is carried out in a predefined cycle while the vehicle is driving.

[0057] First, as shown in step S101, the ECU 20 obtains detection information such as the vehicle's own speed Vs and a relative speed and azimuth of an object behind the vehicle from various sensors such as the radar device 11 and the vehicle speed sensor 12. Then the control system proceeds to step S102.

[0058] In step S102, based on the detection information obtained from radar device 11, an object located behind the vehicle is detected. The type of object is not particularly restricted and can include a vehicle, bicycle, motorcycle, pedestrian, animal, structure, and the like. The control system then proceeds to step S103.

[0059] Step S103 determines whether the objects detected in step S102 include an approaching object near the vehicle. Specifically, if, for example, an object's relative speed to the vehicle is greater than or equal to a predefined approach speed threshold, the object is determined to be approaching. If an approaching object is present (i.e., if an approaching object is detected), the control system proceeds to step S104. If no approaching object is present (i.e., if no approaching object is detected), the process terminates.

[0060] Step S104 determines whether a speed condition is met. In particular, by using the relative speed Vg and the relative speed Vrv obtained from the radar device 11, and the vehicle speed Vs obtained from the vehicle speed sensor 12, it is determined whether a speed condition shown in the following formula (5) is met. The speed difference threshold Vc1 is predefined and set in the ECU 20, for example, by using the frequency distribution diagram in Fig. 3. If it is determined that the speed condition is met, the control system advances to step S105. If it is determined that the speed condition is not met, the control system advances to step S110. |Vg−(Vs+2Vrv)| <Vc1

[0061] In step S105, it is determined whether an azimuth condition is met. In particular, based on the azimuth θrv of the following vehicle and the azimuth θg of the approaching object, which are obtained from the radar device 11, it is determined whether an azimuth condition shown in the following formula (6) is met. The in Fig. The data table shown in Figure 4 is pre-stored in ECU 20. Using the vehicle's own speed Vs, obtained in step S101, and the distance R between the vehicle and the following vehicle, the azimuth difference threshold θc is calculated from the data table stored in ECU 20. If it is determined that the azimuth condition is met, the control system proceeds to step S106. If it is determined that the azimuth condition is not met, the control system proceeds to step S110. |θrv−θg|<θc

[0062] In step S106, it is determined whether the approaching object near the vehicle detected in step S103 is a phantom. If both the speed condition shown in step S104 and the azimuth condition shown in step S105 are met, the approaching object is determined to be a phantom. Consequently, the approaching object detected in step S103 is excluded from being an object for which driver assistance to avoid hazards to the vehicle is required, and no notification command is issued to the alarm device 30.

[0063] In step S110, it is determined that the approaching object, which is near the vehicle detected in step S103, is not a phantom. If at least one of the velocity conditions shown in step S104 and the azimuth condition shown in step S105 is not met, it is determined that the approaching object is not a phantom. The control system then proceeds to step S111.

[0064] In step S111, the approaching object detected in step S103 is identified as requiring driver assistance to prevent a hazard to the vehicle, and it is determined whether a notification should be issued. This determination might involve, for example, checking whether the approaching object has entered a notification zone defined behind the vehicle. If it is determined that the approaching object has entered the notification zone, a notification is required, and the control system proceeds to step S112. In step S112, a notification command is sent to the alarm device 30, and the process terminates. Conversely, if step S111 determines that no notification is required, the process terminates at that point.

[0065] As previously described, according to the present embodiment, an approaching object is determined to be a phantom target based on both the velocity and azimuth conditions. This makes it possible to distinguish between an actual approaching object and a phantom object with higher accuracy.

[0066] The embodiment described above can achieve the following advantageous effects.

[0067] The ECU 20 comprises the object detection unit 22, the data reference unit 21, the phantom detection unit 40, and the assistance detection unit 23. The object detection unit 22 detects an object behind the own vehicle 50 by using reflected waves resulting from search waves transmitted by the radar device 11 or the like, and identifies an object approaching the own vehicle 50 as an approaching object 55. The data reference unit 21 obtains the own vehicle speed Vs and a relative speed and azimuth of the object detected by the object detection unit 22 with respect to the own vehicle 50. The phantom detection unit 40 determines whether the approaching object 55 is a phantom target that is not actually present.The assistance determination unit 23 performs a driving assistance of the own vehicle 50 to avoid danger to the own vehicle 50 from a target object and excludes the approaching object 55, which is determined by the phantom determination unit 40 as a phantom target, from the target object.

[0068] The phantom detection unit 40 comprises the speed detection unit 41 and the azimuth detection unit 42. Based on the vehicle's own speed Vs and the relative speed Vrv of the following vehicle 51, which is located between the vehicle 50 and the approaching object 55, with respect to the vehicle 50, the speed detection unit 41 calculates the calculated value Vg0 of the relative speed of the approaching object 55 with respect to the vehicle 50. Furthermore, the speed detection unit 41 calculates the absolute difference dVg between the detected value Vg of the relative speed of the approaching object 55 with respect to the vehicle 50, obtained from the data reference unit 21, and the calculated value Vg0.Based on the azimuth θrv of the following vehicle 51 and the azimuth θg of the approaching object 55, which are referenced by the data reference unit 21, the azimuth determination unit 42 calculates the azimuth difference dθg, an absolute difference between the azimuth of the following vehicle 51 relative to the own vehicle 50 and the azimuth of the approaching object 55 relative to the own vehicle 50. Then, based on both the velocity difference dVg and the azimuth difference dθg, the phantom determination unit 40 determines whether the approaching object 55 is a phantom target.The phantom detection unit 40 can determine whether the approaching object 55 is a phantom target based on the velocity difference dVg and the azimuth difference dθg, which are each calculated using the relative velocity Vg and azimuth θg of the detected approaching object 55, and the relative velocity Vg0 and azimuth θrv of the approaching object 55, which are theoretically obtained by assuming that the approaching object 55 is a phantom target. Thus, it is possible to distinguish between an actual approaching object and a phantom object with higher accuracy.

[0069] If it is determined that the velocity difference dVg satisfies the velocity condition and the azimuth difference dθg satisfies the azimuth condition, the phantom detection unit 40 can determine that the approaching object 55 is a phantom target. In particular, the velocity detection unit 41 can set the velocity condition to be a condition that the velocity difference dVg is less than the predetermined velocity difference threshold Vc1. Furthermore, the azimuth detection unit 42 can set the azimuth condition to be a condition that the azimuth difference dθg is less than the predetermined azimuth difference threshold θc.Then, if the velocity determination unit 41 makes a positive determination that the velocity condition is met, and the azimuth determination unit 42 makes a positive determination that the azimuth condition is met, the phantom determination unit 40 can determine that the approaching object 55 is a phantom target. By setting the velocity difference threshold Vc1 and the azimuth difference threshold θc according to the situation of the vehicle 50 and its surroundings, an accurate determination can be easily carried out.

[0070] The azimuth determination unit 42 can determine whether the vehicle's own speed Vs is less than the predefined vehicle speed threshold Vc2. If the vehicle's own speed Vs is less than the predefined vehicle speed threshold Vc2, the azimuth determination unit 42 can set the azimuth difference threshold θc to the predefined fixed value θc1. If the vehicle's own speed Vs is greater than or equal to the predefined vehicle speed threshold Vc2, the azimuth determination unit 42 can change the azimuth difference threshold θc within a range greater than the predefined fixed value θc1, based on the distance R between the vehicle 50 and the following vehicle 51, for example, as in the Fig. 4 and Fig.6 shown. The velocity condition and the azimuth condition can be adjusted based on the estimated basic velocity Vgr and the position of the approaching object 55, thus it is possible to successfully adjust the trade-off relationship between the ease of detecting a phantom target and the prevention of an erroneous determination, in which a real approaching object is mistakenly identified as a phantom target.

[0071] The present disclosure has been described according to the embodiments shown, but it is understood that the present disclosure is not limited to these embodiments or structures. The present disclosure includes various modifications and variations in an equivalent area. In addition, the scope of protection and spirit of the present disclosure include various combinations or forms and other combinations or forms comprising only one element, one or more elements, or one or fewer elements.

Claims

[1] Driving assistance device (20) with: an object detection unit (22) that detects one or more objects behind a host vehicle by using a reflected wave resulting from a search wave and recognizes an object approaching the host vehicle in the one or more detected objects as an approaching object; a reference unit (21) that obtains a speed of the own vehicle and a relative speed and an azimuth of each of the one or more objects detected by the object detection unit with respect to the own vehicle; a phantom determination unit (40) which determines whether the approaching object is a phantom target that is not actually present; and an assistance determination unit (23) which performs driving assistance of the own vehicle to avoid danger to the own vehicle from a target object and excludes the approaching object determined by the phantom determination unit as the phantom target from the target object, wherein the reference unit is configured to obtain a first value of the relative speed of the approaching object with respect to the own vehicle; the phantom determination unit is configured to: Calculating a second value of the relative speed of the approaching object with respect to the own vehicle according to the speed of the own vehicle and a value of the relative speed of a following object with respect to the own vehicle, the following object being detected by the object detection unit and being located between the own vehicle and the approaching object; Calculating an absolute difference between the first value of the relative speed of the approaching object and the second value of the relative speed of the approaching object as a speed difference; Calculating an absolute difference between an azimuth of the following object and an azimuth of the approaching object obtained from the reference unit as an azimuth difference; and Determine whether the approaching object is the phantom target based on the speed difference and the azimuth difference. [2] Driving assistance device according to claim 1, wherein: the phantom determination unit has: a speed determination unit (41) which determines whether the speed difference is less than a predetermined speed difference threshold value; and an azimuth determination unit (42) which determines whether the azimuth difference is smaller than a predetermined azimuth difference threshold value, wherein the phantom determination unit is configured to determine that the approaching object is the phantom target in response to both: the speed difference is determined to be less than the specified speed difference threshold; and the azimuth difference is determined to be smaller than the specified azimuth difference threshold. [3] The driving assistance device according to claim 2, wherein the azimuth determination unit is configured to: Determining whether the speed of the own vehicle is less than a predetermined own vehicle speed threshold; Setting the azimuth difference threshold to a predetermined fixed value in response to determining that the speed of the own vehicle is less than the predetermined own vehicle speed threshold; and Changing the azimuth difference threshold based on a distance between the own vehicle and the following object within a range greater than the predetermined fixed value in response to determining that the speed of the own vehicle is greater than or equal to the predetermined own vehicle speed threshold. [4] The driving assistance device according to claim 3, wherein the azimuth determination unit is configured to: Changing the azimuth difference threshold to a smaller value as the distance between the own vehicle and the following object increases in response to determining that the own vehicle speed is greater than or equal to the specified own vehicle speed threshold.

Images