JP6414540B2 - Object detection device - Google Patents

Description

  The present invention relates to an object detection device mounted on a vehicle.

  At intersections and curves (blind curves) where there is no signal and there is a bad view, objects such as cars, motorcycles, bicycles, and pedestrians may suddenly jump out, causing encounter accidents. There seems to be no way to prevent such encounter accidents at first glance, but there are often curved mirrors attached to intersections and blind curves where there is no signal and the prospect is bad, and the driver confirms this Can be expected to reduce accidents.

  However, in general, it is strongly recognized that the driver of the vehicle traveling on the non-priority road should confirm the curve mirror, and the driver of the vehicle traveling on the priority road rarely confirms the curve mirror. Therefore, there is a problem that the curved mirror provided as an accident mitigation tool is not fully utilized.

  In addition, when a pedestrian or a lightless bicycle jumps out at night or the like, it is often difficult for the driver to see these objects even if a curved mirror is provided. Here, it can be considered that the curve mirror is provided because something may come out. Therefore, it is useful for ensuring the safety of the vehicle to install a device that monitors the curve mirror and determines the possibility of collision with an object jumping out from an intersection or the like on the vehicle.

  For example, Patent Document 1 discloses another road from a shift between the frequency of an ultrasonic wave radiated to a curve mirror and the frequency of the ultrasonic wave reflected by the curve mirror and returning to the electric vehicle. Disclosed is a technique for detecting a vehicle traveling on the road.

JP 2012-38136 A

  By the way, depending on the relative positional relationship between the mirror surface of the curved mirror and the own vehicle, the own vehicle may be reflected on the curved mirror. In this case, there is a problem that the own vehicle reflected on the curve mirror is erroneously detected as a detection target. Further, in the case of such erroneous detection, it is useless to determine the possibility of collision with the own vehicle.

  However, in Patent Document 1, since no measures are taken when the vehicle is reflected on the curve mirror, the problem of erroneously detecting the vehicle as a detection target cannot be solved.

  An object of the present invention is to provide an object detection device that can prevent the vehicle from being detected as a detection target even when the vehicle is reflected on a curved mirror.

An object detection device according to an aspect of the present invention includes an image sensor that captures an image in front of the host vehicle,
A mirror detection unit that extracts an image showing a curve mirror from an image captured by the image sensor and detects the direction of the existing curve mirror using the extraction result;
By irradiating the measurement wave in the direction detected by the mirror detection unit and receiving the reflected wave from the object reflected on the curve mirror via the curve mirror, the distance from the own vehicle to the object and A distance sensor for measuring a relative speed of the object with respect to the vehicle;
A collision determination unit that determines the possibility of collision between the host vehicle and the object using the measurement result of the distance sensor;
A vehicle speed sensor for detecting a speed of said vehicle,
When the relative speed measured by the distance sensor is the same speed as the speed detected by the vehicle speed sensor, the vehicle includes a vehicle determination unit that determines that the object is the vehicle.
The collision determination unit prohibits the process of determining the collision possibility when the host vehicle determination unit determines that the object is the host vehicle.

  According to this aspect, the distance sensor receives the reflected wave from the object reflected on the curve mirror via the curve mirror. If the relative speed of the object relative to the vehicle measured by the distance sensor is the same as the speed of the vehicle, the object reflected on the curve mirror is determined as the own vehicle. Therefore, according to this aspect, it is possible to prevent the vehicle from being erroneously detected as a detection target.

  In addition, if the object reflected on the curve mirror is determined to be the host vehicle, the collision possibility determination process is prohibited, preventing unnecessary processing such as determining the possibility of collision with the host vehicle. it can.

In the above aspect, a notification unit that notifies the possibility of collision with the object;
A vehicle control unit that causes the host vehicle to execute collision avoidance control for avoiding a collision with the object;
The notification unit prohibits the notification when the vehicle determination unit determines that the object is the vehicle.
The vehicle control unit may prohibit the collision avoidance control when the vehicle determination unit determines that the object is a vehicle.

  According to this aspect, when the vehicle determination unit determines that the object reflected on the mirror is the vehicle, the collision avoidance control and the collision possibility warning are prohibited, so that the driver's driving is hindered. Can be prevented.

In the above aspect, the notification unit includes:
An alerting part for displaying an image prompting attention to the curve mirror;
An alarm unit for performing an alarm on a collision with the object using at least one of audio and video,
When the vehicle determination unit determines that the object is the vehicle, the alerting unit displays an image for prompting the alert,
The warning unit may prohibit the warning when the vehicle determination unit determines that the object is a vehicle.

  According to this aspect, when the vehicle determination unit determines that the object reflected in the mirror is the vehicle, an image that prompts attention to the curve mirror is displayed, but an object using sound or video Collision warning is prohibited. Therefore, the driver can be alerted to the object, while the driver can be prevented from being unnecessarily warned. As a result, it is possible to prevent the driver's ability to concentrate on driving.

  In the above aspect, the distance sensor may be a radio wave radar.

  According to this aspect, since the object is detected using the radio wave type radar, the object can be detected with high resolution.

  In the above aspect, the measurement wave radiated by the radio wave radar may have any frequency within a frequency band from a microwave band to a millimeter wave band.

  According to this aspect, since the object is detected using the measurement wave having any frequency within the frequency band from the microwave band to the millimeter wave band, the object can be detected with high resolution.

  In the above aspect, the own vehicle determination unit may determine that the object is the own vehicle when the condition that the intensity of the reflected wave is greater than a predetermined intensity is further satisfied.

  When the vehicle is reflected on the curved mirror, the distance sensor directly receives the reflected wave reflected by the curved mirror. Therefore, compared to when the vehicle is not reflected on the curved mirror, the received intensity of the reflected wave is higher. Significantly larger. In this aspect, in addition to the condition that the speed of the own vehicle is the same as the relative speed, the object is determined to be the own vehicle when the condition that the intensity of the reflected wave is greater than the predetermined intensity is further satisfied. Therefore, according to this aspect, it is possible to accurately determine whether or not the object reflected on the curve mirror is the own vehicle.

  In the above aspect, when the vehicle determination unit further satisfies the condition that the distance sensor detects a first object and a second object that is twice as far as the first object, the object is May be determined.

  When the vehicle is reflected on the curved mirror, the received intensity of the received wave is high. Therefore, the distance sensor also reflects the reflected wave reflected twice by the curved mirror in addition to the reflected wave directly reflected from the curved mirror. It can be detected. Therefore, the distance sensor detects the second object as if the second object is present at a position that is twice as far as the first object.

  In this aspect, when the condition that the distance sensor detects the first object and the second object that is twice the distance from the first object is further satisfied, the object is determined to be the own vehicle, and therefore reflected in the curve mirror. It is possible to accurately determine whether or not the object in the vehicle is the own vehicle.

  ADVANTAGE OF THE INVENTION According to this invention, it can prevent that the own vehicle is misdetected as a detection target object.

It is a figure explaining the outline | summary of the object detection apparatus by embodiment of this invention. It is a side view of the vehicle carrying the object detection apparatus by this Embodiment. It is a block diagram of the object detection apparatus by embodiment of this invention. It is a flowchart which shows the process of the object detection apparatus concerning embodiment of this invention.

  FIG. 1 is a diagram illustrating an outline of an object detection apparatus according to an embodiment of the present invention. The vehicle 11 is equipped with an object detection device according to the present embodiment. As the vehicle 11, for example, a four-wheeled vehicle is employed. The vehicle 11 is approaching the intersection. This intersection is a T-shaped road where the road 16 is orthogonal to the road 12 on which the vehicle 11 travels. A building 15 exists adjacent to the road 12 and the road 16 in front of the road 16 and diagonally to the right of the vehicle 11. Therefore, the road 16 becomes a blind spot of the vehicle 11, and the driver of the vehicle 11 cannot see the bicycle 14 traveling on the road 16.

  Therefore, at this intersection, the curve mirror 13 is provided on the left side of the region 17 where the road 12 and the road 16 intersect. Thereby, the bicycle 14 located in the blind spot is reflected on the curve mirror 13, and the driver of the vehicle 11 can visually observe the bicycle 14 through the curve mirror 13.

  However, if the road 12 is a priority road, the driver of the vehicle 11 looks at the curve mirror 13 from the idea that the person who travels on the road 16 side that is a non-priority road has an obligation to view the curve mirror 13. I may neglect. In addition, the driver of the vehicle 11 may not be able to see the curve mirror 13 at night.

  Therefore, an object detection device is mounted on the vehicle 11. The object detection device includes a distance sensor 22 provided on the front surface of the vehicle 11. When the object detection device detects the curve mirror 13, the object detection device causes the distance sensor 22 to emit the measurement wave B <b> 1 toward the curve mirror 13. The measurement wave B1 is reflected by the curve mirror 13. Here, when the bicycle 14 is reflected on the curve mirror 13, the measurement wave B <b> 1 reflected by the curve mirror 13 reaches the bicycle 14. The measurement wave B 1 is reflected by the bicycle 14, and the reflected wave is further reflected by the curve mirror 13 and returns to the distance sensor 22. Therefore, the distance sensor 22 can detect the distance from the vehicle 11 to the bicycle 14 and the relative speed of the bicycle 14 with respect to the vehicle 11.

  FIG. 2 is a side view of the vehicle 11 on which the object detection device according to the present embodiment is mounted. The object detection apparatus includes a distance sensor 22 provided on the front surface of the vehicle 11, for example, and an image sensor 21 provided on the windshield of the vehicle 11, for example. The image sensor 21 has a constant angle of view 221 extending radially from the sensor surface in a side view, and always images the front of the vehicle 11. The object detection device constantly monitors whether or not the image captured by the image sensor 21 includes the curve mirror 13. And if an object detection apparatus detects the curve mirror 13, it will radiate | emit the measurement wave B1 of the distance sensor 21 toward the direction. Then, the object detection device receives the reflected wave from the target reflected on the curve mirror 13 via the curve mirror 13 and detects an object (hereinafter referred to as “target”) that is a detection target located at the blind spot. To detect. When the target is detected, the object detection device notifies the driver of the danger of collision using voice or images, or performs collision avoidance control as necessary, thereby preventing an accident at the intersection.

  Here, the vehicle 11 is reflected on the curve mirror 13 depending on the relative position relationship between the curve mirror 13 and the vehicle 11 traveling on the road 12 because the curve mirror 13 is installed with a deviation from the correct position for some reason. Sometimes. In this case, the distance sensor 22 receives the reflected wave directly reflected from the curve mirror 13. Since the distance sensor 22 cannot determine whether or not the detected target is the vehicle 11, the vehicle 11 is erroneously detected as a target different from the vehicle 11. In this case, if a collision warning or collision avoidance control for the vehicle 11 is performed in spite of erroneous detection, driving of the driver is hindered. The object of the object detection apparatus according to the present embodiment is to solve such a problem.

  Hereinafter, the details of the object detection device according to the present embodiment will be described. FIG. 3 is a block diagram of the object detection apparatus according to the embodiment of the present invention. In addition to the image sensor 21 and the distance sensor 22 shown in FIG. 2, the object detection device includes a vehicle speed sensor 23, a control unit 24, a notification unit 25, a brake unit 26, a PT (power train) unit 27, a headlight 28, and a horn 29. Is provided. In FIG. 3, the image sensor 21, the distance sensor 22, and the control unit 24 may be integrated.

  The image sensor 21 is configured by, for example, a COMS area sensor, a CCD area sensor, or the like, and captures an image in front of the vehicle 11 at a predetermined frame rate (for example, 60 fps). The image sensor 21 may be a monocular camera, a stereo camera capable of measuring a distance, or a depth sensor. In addition, in the vehicle 11, for example, if a stereo camera or a depth sensor is attached to detect the inter-vehicle distance from the vehicle ahead, these existing stereo cameras and depth sensors are employed as the image sensor 21. It only has to be done.

  The distance sensor 22 is configured by, for example, a radio wave radar, irradiates a measurement wave in a direction detected by the mirror detection unit 241, and receives a reflected wave from an object reflected on the curve mirror via the curve mirror. Thus, the distance from the vehicle 11 to the target and the relative speed of the target with respect to the vehicle 11 are measured. As the radio wave radar, a microwave radar that detects a target by emitting microwaves, a millimeter wave radar that detects a target by emitting millimeter waves, or a terahertz radar that detects a target by emitting terahertz waves can be employed. Alternatively, as the distance sensor 22, an infrared rider that detects a target by emitting infrared rays may be employed.

  Here, the distance sensor 22 may detect the time until the measurement wave returns, and calculate the distance to the target using the detected time and the velocity of the measurement wave. Further, the distance sensor 22 may calculate the relative velocity of the target by detecting the Doppler frequency of the reflected wave.

  The distance sensor 22 is configured by a sensor that can narrow the beam diameter of the measurement wave to such an extent that at least one of transmission and reception does not protrude from the curve mirror. The distance sensor 22 includes a sensor that can electronically scan a measurement wave, such as a phased array or digital beam forming, or a sensor that can mechanically scan a measurement wave, such as a mechanical scan.

  The method of narrowing the beam diameter can be realized by increasing the antenna area or increasing the frequency of the measurement wave. For example, the distance sensor 22 includes n (n is an integer of 2 or more) antennas whose beam diameter can be reduced until the curve mirror can be scanned, and normally, for example, n / 2 antennas are driven. When the curved mirror is detected, the beam diameter may be reduced by driving n antennas.

  Further, the distance sensor 22 normally radiates a measurement wave having a frequency smaller than that of the millimeter wave, and when the curve mirror is detected, the distance sensor 22 may narrow the beam diameter by raising the frequency to the millimeter wave to THz band. As a result, the position of the target can be accurately detected.

  The vehicle speed sensor 23 measures the speed of the vehicle 11. Here, the vehicle speed sensor 23 may be configured by an existing vehicle speed sensor mounted on the vehicle 11.

  The control unit 24 includes a computer including a CPU, a ROM, a RAM, and the like, and includes a mirror detection unit 241, a collision determination unit 242, a host vehicle determination unit 243, and a vehicle control unit 244. In addition, the mirror detection part 241-the vehicle control part 244 are implement | achieved when CPU runs the program memorize | stored in ROM, for example.

  The mirror detection unit 241 extracts an image showing a curve mirror from the image captured by the image sensor 21, and detects the direction of the existing curve mirror using the extraction result.

  Here, the mirror detection unit 241 includes a template indicating the characteristic amount of the curve mirror in advance, and the curve mirror may be detected by comparing this template with each image data captured by the image sensor 21. Specifically, a template having the shape of a curve mirror as a feature amount can be employed as the template.

  The shape of the mirror part of the curved mirror is circular when viewed from the front, but in many cases it appears to be elliptical from the vehicle 11. The curved mirror has a shape in which the mirror portion is placed on a single bar extending in the vertical direction. Therefore, as a template, a template having a shape in which an ellipse is placed on a single bar as a feature amount can be employed.

  Then, the mirror detection unit 241 determines whether there is a curve mirror by using an image processing method such as template matching for each image data captured by the image sensor 21. Specifically, the mirror detection unit 241 may calculate the similarity of the curve mirror in each image data, and determine that the curve mirror exists if the similarity is equal to or higher than the reference similarity.

  Further, when the mirror detection unit 241 detects the curve mirror from the image data, the mirror detection unit 241 calculates the direction of the actual curve mirror with respect to the vehicle 11 from the two-dimensional coordinates in the detected image data of the curve mirror. Specifically, the mirror detection unit 241 is a sensor of the image sensor 22 that is set in advance in a three-dimensional vehicle coordinate space with a certain position of the vehicle 11 (for example, the vertex of the angle of view of the image sensor 21) as the origin. The two-dimensional coordinates of the detected curve mirror are plotted in the area corresponding to the surface, and the direction of the straight line connecting the plotted two-dimensional coordinates and the origin may be determined as the direction of the curve mirror in the real space with respect to the vehicle 11. . As the position of the two-dimensional coordinates of the curve mirror, for example, the center of the mirror part can be adopted.

  Then, the mirror detection unit 241 instructs the distance sensor 22 to emit a measurement wave in the direction of the curved mirror in the detected real space. The distance sensor 22 having received this instruction may narrow the beam diameter of the measurement wave and scan the measurement wave over the entire mirror portion, for example.

  The collision determination unit 242 determines the possibility of collision between the vehicle 11 and the target using the measurement result of the distance sensor 22. Here, the collision determination unit 242 first calculates a predicted traveling pattern of the target within a certain period (for example, 30 seconds) from the present time. Here, the collision determination unit 242 may calculate the predicted traveling pattern of the target as follows. First, the target acceleration is calculated from the transition of the relative speed of the target up to the present time, and the predicted position of the target is calculated every predetermined time interval (for example, 1 second) using the calculated acceleration and the current relative speed. A point cloud composed of the calculated predicted positions may be calculated as a predicted traveling pattern. Similarly, the collision determination unit 242 calculates the predicted travel pattern of the vehicle 11 by calculating the predicted position of the vehicle 11 at predetermined time intervals until a predetermined period has elapsed from the current time. Then, the collision determination unit 242 sets an area indicating the size of the vehicle 11 and an area indicating the size of the target at each predicted position of the vehicle 11 and the target, and both areas overlap at the predicted position at the same timing. In such a case, it may be determined that there is a possibility of collision.

  In addition, when the own vehicle determination unit 243 determines that the target is the vehicle 11, the collision determination unit 242 prohibits the above-described processing for determining the possibility of collision. Thereby, generation | occurrence | production of the useless process of determining the collision possibility with the vehicle 11 can be prevented.

  The own vehicle determination unit 243 determines that the target is the vehicle 11 when the relative speed measured by the distance sensor 22 is the same as the speed detected by the vehicle speed sensor 23. When the vehicle 11 is reflected on the curve mirror, the measurement wave radiated from the distance sensor 22 is directly reflected by the curve mirror and returns to the vehicle 11, so that the distance sensor 22 has the same speed as the vehicle 11. A target approaching 11 is detected. Thereby, it can be determined whether or not the target is the vehicle.

  When the vehicle determination unit 243 determines that the target is the vehicle 11, the vehicle control unit 244 prohibits the brake unit 26 and the PT unit 27 from executing collision avoidance control for avoiding a collision with the target. To do.

  The notification unit 25 notifies the possibility of collision between the target and the vehicle 11. In this Embodiment, the alerting | reporting part 25 prohibits the alerting | reporting of collision possibility, when the own vehicle determination part 243 determines with the target being the vehicle 11. FIG.

  Specifically, the notification unit 25 includes an alarm unit 251 and an alerting unit 252. The alarm unit 251 includes, for example, a speaker, a head-up display, an alarm lamp, and the like, and performs an alarm regarding a collision with the target using audio and video. As an aspect of the alarm, there are an aspect in which an alarm sound or a sound such as “collision with an oncoming vehicle” is output from the speaker, an aspect in which an annotation mark is displayed on the head-up display, and an aspect in which the alarm lamp blinks.

  The alarm unit 251 prohibits the alarm when the vehicle determination unit 243 determines that the target is the vehicle 11.

  The alerting unit 252 is configured with, for example, a head-up display, and displays an image that prompts alerting the curve mirror. As an alerting mode, an image in which a curved mirror is marked is displayed on a head-up display, or an image in which a target is virtually displayed on an obstacle such as a building is displayed on the head-up display. Alternatively, if a headlight is employed as the alerting unit 252, a mode in which the headlight is directed toward the curve mirror may be employed.

  Then, when the own vehicle determination unit 243 determines that the target is the vehicle 11, the alerting unit 252 displays an image for prompting alerting or lights the curve mirror on the headlight.

  The brake unit 26 includes, for example, a brake that brakes the vehicle 11 and an automatic braking mechanism that automatically sets an operation amount of the brake under the control of the vehicle control unit 244. As the automatic braking mechanism, a mechanism having a DSC (Dynamic Stability Control System) control function capable of preventing a side slip at the time of braking of the vehicle 11 can be adopted.

  The PT unit 27 includes a steering unit 271 and an engine unit 272. For example, the steering unit 271 includes an automatic steering mechanism that automatically sets the steering operation amount and automatically changes the steering angle of the tire under the control of the vehicle control unit 244.

  The engine unit 272 includes, for example, an automatic setting mechanism that automatically sets the number of gear stages of the transmission and the number of rotations of the engine under the control of the engine, the transmission, and the vehicle control unit 244.

  The headlight 28 is mounted on the front surface of the vehicle 11 and irradiates the front of the vehicle 11. The horn 29 is mounted on the vehicle 11 and emits an alarm sound for urging attention to an oncoming vehicle or the like.

  Next, processing of the object detection device in the present embodiment will be described. FIG. 4 is a flowchart showing processing of the object detection apparatus according to the embodiment of the present invention.

  First, the mirror detection unit 241 determines whether or not the image data captured by the image sensor 21 includes a curve mirror using a curve mirror template (S401). When the curve mirror can be detected (YES in S401), the mirror detection unit 241 calculates the direction of the curve mirror with respect to the vehicle 11 in real space from the two-dimensional coordinates in the image data of the curve mirror (S402).

  On the other hand, if the curve mirror cannot be detected (NO in S401), the process returns to S401.

  In S403, the mirror detection unit 241 instructs the distance sensor 22 to emit a measurement wave in the direction of the curve mirror. Next, the distance sensor 22 radiates a measurement wave in the direction of the curve mirror (S404).

  Next, the distance sensor 22 detects the relative velocity of the target by detecting the Doppler frequency of the reflected wave (S405).

  Next, the own vehicle determination unit 243 determines whether or not the target is an approaching target approaching the vehicle 11. For example, if the relative speed is positive, the target is approaching the vehicle 11, and if the relative speed is negative, the target is away from the vehicle 11, the own vehicle determination unit 243 If the relative speed is positive, it may be determined that there is an approaching target.

  When the approaching target is detected (YES in S406), the own vehicle determination unit 243 determines whether or not the relative speed is equal to the speed of the vehicle 11 detected by the vehicle speed sensor 23 (S407). On the other hand, when the approaching target is not detected (NO in S406), the own vehicle determination unit 243 returns the process to S401 because there is no possibility of a collision with the target.

  If the relative speed of the target is equal to the speed of the vehicle 11 (YES in S407), there is a high possibility that the target is the vehicle 11. Therefore, the own vehicle determination unit 243 determines that the intensity of the received wave received by the distance sensor 22 is the reference intensity. It is determined whether it is larger (S408). On the other hand, if the relative speed of the target is not equal to the speed of the vehicle 11 (NO in S407), there is a low possibility that the target is the vehicle 11, and therefore the own vehicle determination unit 243 advances the process to S412.

  When the target is not the vehicle 11, the measurement wave passes through the path of the vehicle 11 -curve mirror-target-curve mirror-vehicle 11. On the other hand, when the target is the vehicle 11, the measurement wave passes through the route of the vehicle 11 -curve mirror-vehicle 11. Here, since the measurement wave is a radio wave, it attenuates in proportion to the fourth power of the distance. The measurement wave is greatly attenuated when reflected.

  The path of the measurement wave when the target is not the vehicle 11 is longer than that when the target is the vehicle 11 by the distance reciprocating between the curve mirror and the target section. In addition, the number of reflections when the target is not the vehicle 11 is 3 times in total, 2 times with the curve mirror and 1 time with the target, but the number of reflections when the target is the vehicle 11 is 1 time with the curve mirror. . Therefore, when the target is not the vehicle 11, the measurement wave is greatly attenuated as compared with the case where the target is the vehicle 11. Therefore, in the present embodiment, the process of S408 is provided to improve the determination accuracy as to whether or not the target is the vehicle 11. Here, as the reference intensity, for example, values such as 50%, 60%, 80%, and 90% of the standard intensity of the measurement wave measured when the target is the vehicle 11 can be adopted.

  If the intensity of the received wave is greater than the reference intensity (YES in S408), the vehicle determination unit 243 has the ID (identifier) included in the received wave as the ID of the measurement wave transmitted from the distance sensor 22 mounted on the vehicle 11. It is determined whether or not (S409). On the other hand, if the intensity of the received wave is not greater than the reference intensity (NO in S408), the process proceeds to S412.

  Here, the ID is an ID included in the transmission format of the measurement wave, and for example, an identifier of the distance sensor 22 or the vehicle 11 can be adopted. For example, the distance sensor 22 may receive a measurement wave transmitted from a vehicle different from the vehicle 11. In this case, since the distance sensor 22 directly receives the measurement wave, the distance sensor 22 receives a strong measurement wave. Therefore, the process of S409 is provided to determine whether or not the received measurement wave is transmitted by the distance sensor 22, and the target is determined to be the vehicle 11 based on the measurement wave transmitted from another vehicle. Is prevented.

  In S409, if the ID included in the received measurement wave is the ID of the measurement wave transmitted from the distance sensor 22 (YES in S409), the process proceeds to S410. On the other hand, if the ID included in the received measurement wave is not the ID of the measurement wave transmitted from the distance sensor 22 (NO in S409), the process proceeds to S412.

  In S410, it is determined whether the own vehicle determination unit 243 has detected two targets. When the target is the vehicle 11, since the attenuation of the measurement wave is small, the measurement wave directly reflected by the curve mirror is further reflected by the vehicle 11, and the reflected measurement wave has a certain strength through the curve mirror. There is a high possibility of returning to 11. In this case, the distance sensor 22 can detect the second target at a position twice as long as the first target, in addition to the first target. Therefore, in S410, when the distance sensor 22 detects a second target located twice as long as the first target, the vehicle determination unit 243 determines that two targets have been detected (in S410). YES), the process returns to S401. That is, in the case of YES in S410, the own vehicle determination unit 243 determines that the target is the own vehicle. Therefore, in order to prohibit the collision possibility determination process, collision avoidance control, warning, and alerting, the process is performed. The process returns to S401. In FIG. 4, in the case of YES in S <b> 410, alerting is not performed, but alerting may be performed by the alerting unit 252.

  On the other hand, when two targets are not detected (NO in S410), the process proceeds to S411.

  In S <b> 412, the collision determination unit 242 detects the target distance by acquiring the target distance from the distance sensor 22. Next, the collision determination unit 242 determines the possibility of collision between the target and the vehicle 11 by calculating the predicted traveling pattern of the target and the predicted traveling pattern of the vehicle 11 (S413).

  If the collision determination unit 242 determines that there is a collision possibility (YES in S414), the process proceeds to S415. If the collision determination unit 242 determines that there is no collision possibility (NO in S414), the process proceeds to S411.

  In S415, the collision determination unit 242 determines the possibility of collision between the target and the vehicle 11 when the collision avoidance control is performed.

  In this case, the collision determination unit 242 calculates a collision avoidance travel pattern of the vehicle 11 using the current target predicted travel pattern and the current speed and acceleration of the vehicle 11, and uses the collision avoidance travel pattern to calculate the vehicle. What is necessary is just to determine whether the vehicle 11 and a target collide, when it is assumed that 11 was made to drive | work.

  Here, the predicted travel pattern calculated in S413 can be adopted as the target predicted travel pattern.

  Further, the collision avoidance travel pattern includes a collision avoidance travel pattern by braking of the vehicle 11 and a collision avoidance travel pattern by acceleration of the vehicle 11. For example, the collision determination unit 242 may calculate a collision avoidance travel pattern based on acceleration when the collision avoidance travel pattern based on braking is calculated and it is determined that the collision cannot be avoided. If the collision determination unit 242 determines that the collision cannot be avoided even in the collision avoidance travel pattern by acceleration, the collision determination unit 242 may determine YES in S416.

  As the collision avoidance travel pattern by braking, for example, a pattern in which the braking distance is the shortest from the current speed of the vehicle 11 can be adopted. The collision avoidance traveling pattern by acceleration is, for example, before the target reaches the predicted collision position of the target and the vehicle 11 calculated when the target and the vehicle 11 travel at the current speed or acceleration. A pattern in which the vehicle 11 travels at an acceleration at which the vehicle 11 can pass through the predicted collision position can be employed. In this case, as an acceleration that can pass through the predicted collision position, for example, an acceleration that is assumed when the number of gear stages of the transmission is shifted down may be employed.

  Here, if the road width of the road on which the vehicle 11 is traveling is larger than the reference width, the collision determination unit 242 may calculate a collision avoidance traveling pattern by further combining automatic steering. In this case, for example, if the vehicle 11 includes a navigation device, the collision determination unit 242 may acquire road width information from the navigation device. Further, as the reference width, for example, a value obtained by multiplying the vehicle width of the vehicle 11 by a certain coefficient (for example, 1.2 times, 1.5 times, 2.0 times the vehicle width) can be adopted.

  As a pattern that combines the collision avoidance traveling pattern by braking and automatic steering, for example, in the predicted stop position of the vehicle 11 when the vehicle 11 is braked at the shortest braking distance, an area indicating the vehicle 11 and an area indicating the target If they overlap, a pattern can be adopted in which the steering angle of the vehicle 11 is automatically adjusted so that the two regions do not overlap and the vehicle 11 is braked.

  Further, as a pattern that combines the collision avoidance traveling pattern by acceleration and the automatic steering, for example, if the area of the vehicle 11 and the target area overlap at the predicted collision position, the predicted collision position has a curvature that does not cause skidding. A pattern in which the steering angle is automatically adjusted so that the vehicle 11 detours can be employed.

  In S416, when the collision determination unit 242 determines that the collision between the vehicle 11 and the target cannot be avoided even if the collision avoidance control is performed (YES in S416), the collision control and passing control by the vehicle control unit 244, and an alarm An alarm by the unit 251 is performed. Here, the collision avoidance control is performed, for example, by the vehicle control unit 244 controlling the brake unit 26 and the PT unit 27 so that the vehicle 11 travels in the collision avoidance travel pattern calculated by the collision determination unit 242. That's fine. The passing control may be, for example, a control in which the vehicle control unit 244 passes the headlight 28 with a high beam and sounds the horn 29 within a legally restricted range. Here, the degree of warning by the warning unit 251 may be increased as the relative speed of the target increases, for example.

  On the other hand, when it is determined in S416 that a collision can be avoided (NO in S416), a collision avoidance control by the vehicle control unit 244 and a warning by the notification unit 25 are performed (S418).

  In S411, since there is a high possibility that the target is an oncoming vehicle, an alarm by the alarm unit 251 and an alert by the alerting unit 252 are performed. When the processes of S411, S417, and S418 are completed, the process returns to S401.

  Thus, according to the present embodiment, when the target is determined to be the vehicle 11 (YES in S410), the collision possibility determination process is prohibited, so that the collision possibility with the vehicle 11 is determined. It is possible to prevent unnecessary processing from occurring. Further, when the relative speed of the target is the same as the speed of the vehicle 11 (YES in S407), the reception intensity of the measurement wave is larger than the reference intensity (YES in S408), and two targets can be detected ( YES at S410) The target is determined to be the vehicle 11. Therefore, it can be accurately determined whether or not the target is the vehicle 11.

DESCRIPTION OF SYMBOLS 11 Vehicle 13 Curve mirror 21 Image sensor 22 Distance sensor 23 Vehicle speed sensor 24 Control part 25 Notification part 26 Brake unit 27 PT unit 28 Headlight 29 horn 241 Mirror detection part 242 Collision judgment part 243 Own vehicle judgment part 244 Vehicle control part 251 Alarm Part 252 Alerting part 271 Steering unit 272 Engine unit

Claims (7)

An image sensor that captures an image in front of the vehicle;
A mirror detection unit that extracts an image showing a curve mirror from an image captured by the image sensor and detects the direction of the existing curve mirror using the extraction result;
By irradiating the measurement wave in the direction detected by the mirror detection unit and receiving the reflected wave from the object reflected on the curve mirror via the curve mirror, the distance from the own vehicle to the object and A distance sensor for measuring a relative speed of the object with respect to the vehicle;
A collision determination unit that determines the possibility of collision between the host vehicle and the object using the measurement result of the distance sensor;
A vehicle speed sensor for detecting a speed of said vehicle,
When the relative speed measured by the distance sensor is the same speed as the speed detected by the vehicle speed sensor, the vehicle includes a vehicle determination unit that determines that the object is the vehicle.
The said collision determination part is an object detection apparatus which prohibits the process which determines the said collision possibility, when the said vehicle determination part determines that the said object is a own vehicle. A notification unit for reporting the possibility of collision with the object;
A vehicle control unit that causes the host vehicle to execute collision avoidance control for avoiding a collision with the object;
The notification unit prohibits the notification when the vehicle determination unit determines that the object is the vehicle.
The object detection device according to claim 1, wherein the vehicle control unit prohibits the collision avoidance control when the vehicle determination unit determines that the object is a vehicle. The notification unit
An alerting part for displaying an image prompting attention to the curve mirror;
An alarm unit for performing an alarm on a collision with the object using at least one of audio and video,
When the vehicle determination unit determines that the object is the vehicle, the alerting unit displays an image for prompting the alert,
The object detection device according to claim 2, wherein the warning unit prohibits the warning when the vehicle determination unit determines that the object is a vehicle.   The object detection apparatus according to claim 1, wherein the distance sensor is a radio wave radar.   The object detection apparatus according to claim 4, wherein the measurement wave radiated from the radio wave type radar has any frequency within a frequency band from a microwave band to a millimeter wave band.   The object detection device according to claim 1, wherein the vehicle determination unit determines that the object is a vehicle when the condition that the intensity of the reflected wave is greater than a predetermined intensity is further satisfied.   The vehicle determination unit determines that the object is a vehicle when the distance sensor further satisfies a condition that the distance sensor detects a first object and a second object that is twice the distance from the first object. Item 7. The object detection device according to any one of Items 1 to 6.

Images