• 专利附录
  • 专利说明
  • 权利要求
  • 类似技术
  • 同族专利
  • 引用文献
  • 法律状态
  • 优先权
    • 专利摘要:
    A method of rejecting a grasped object (28) if the distance (30) from the vehicle (1) is greater than a corridor width (40) and relying on the distance (30) provided by the detector (14) if the camera (18) has not recognized any object (42) within a range of distance determined by the detector (14), and / or the camera (18) has recognized an object (38) in a distance interval but which is not in a corridor (36) adjacent to the vehicle (1).
    公开号:FR3047105A1
    申请号:FR1750573
    申请日:2017-01-24
    公开日:2017-07-28
    发明作者:Zoltan Banko；David Losteiner；Tamas Miklos
    申请人:Robert Bosch GmbH；
    IPC主号:
    专利说明:

    Field of the invention
    The present invention relates to a driving assistance method of grasping objects in the environment of the vehicle and issuing a warning if an object is in a dead angle range of the vehicle. The invention also relates to a computer program for applying the method and a driving assistance system for implementing the method.
    State of the art
    In the automotive field various driver assistance systems are used to assist the driver in performing various driving maneuvers. Thus, for example, there are reversing assistants including a camera using a rear view camera providing rearward facing camera image data and providing the driver with a better orientation for better direction. backwards.
    Another driver assistance system warns the driver of the presence of objects in a blind spot.
    For their operation, the driver assistance systems need data relating to the environment of the vehicle; for this purpose, a plurality of sensors or detectors operating in particular by ultrasound are used. Ultrasonic sensors emit ultrasound signals and receive echoes from objects in the environment. From the ultrasound signal travel time and the sound velocity in the air (known velocity) the distance or distance between the detector / sensor and the reflective object is calculated. Such a detector has a field of view for detecting objects in this field of view. Since detectors using ultrasound can only determine a distance, it is necessary, in particular for the warning of objects in the blind spot, to use other information to decide whether or not an object is located in the blind spot. the dead angle.
    DE 10 2012 204 948 A1 discloses a driving assistance method according to which the range of the dead angle is particularly monitored. According to this method, it is intended to detect the objects in the environment of the vehicle using a camera system. It is also possible to use other sensors or detectors, such as, for example, ultrasonic detectors. To avoid erroneous warnings that could be generated by a vehicle distant from several traffic lanes, it is planned to check the plausibility. For this, a recognized object is associated in the angle detected in the range of the blind spot to a traffic corridor using the distance data (distance data).
    DE 10 2012 215 014A1 discloses a blind spot monitoring method. For monitoring purposes, a first environmental input device is used, for example in the form of ultrasonic detectors oriented at a rearward bias. There is also provided another environment capture facility, for example in the form of a camera system to confirm the plausibility of the data provided by the first environment capture facility. For this, we use the data of the first environment capture installation in a module and detects the objects. The relative speed of an object is also determined. Then the plausibility of the data is checked using another module that uses the data provided by the other environment capture facility. Only if an object is in the blind spot is there a warning.
    DE 10 2013 013 082 A1 discloses a blind spot warning system. The system includes a distance measuring unit that measures the distance between the vehicle and a target vehicle in the lateral area or in the rear area. The distance measuring unit is, for example, in the form of an ultrasonic detector. There is also a camera-shaped unit which determines the distance from a line delimiting the width of the traffic corridor. With the aid of the width of the traffic corridor thus obtained, it is determined whether or not the target vehicle is in the neighboring traffic corridor.
    Description and advantages of the invention
    A driving assistance method in which objects are entered into the environment of a vehicle and a warning is issued if an object is in the range of the vehicle's blind spot, comprising: a) grabbing objects in the field of view of a distance sensor fitted to the vehicle and oriented at a rearward angle with respect to the vehicle, and determining the distance of the objects seized with respect to the vehicle, b) recognizing objects in the field of vision of the vehicle, a camera, equipping the vehicle with a rearward orientation with respect to the vehicle and calculating the distance of the distance of a known object from the vehicle and / or associating the recognized objects with a traffic corridor, filter the objects entered by the distance detector using the objects recognized by the camera, and d) issue a warning if a seized object has not been rejected by the filter and if its removal provided by the r the distance detector is less than a substantial distance, wherein i. an object seized is rejected by filtering according to step c) if the distance provided by the distance sensor of the object relative to the vehicle is greater than a width of the traffic corridor and based on the distance provided by the distance detector, if the camera has not recognized any object in a distance interval determined from distance by the distance detector, and / or ii. an object captured by the filtering according to step c) is rejected if the camera has recognized an object in a distance interval determined from the distance provided by the distance sensor, and if this object is not in a traffic corridor next to the vehicle.
    In other words, the subject of the invention is a method of driving assistance consisting in grasping the objects in the environment of a vehicle and issuing a warning if an object is in a blind spot range . According to the method, the objects are seized in the field of view of a distance detector, fitted to the vehicle and oriented obliquely towards the rear. The distance of the objects seized with respect to the vehicle is thus determined. In addition, we recognize objects that are in the field of view of a camera equipping the vehicle and facing Den, we calculate the distance between the recognized object and the vehicle and / or the objects are associated with a corridor of circulation. Finally, the objects captured by the distance detector are filtered using the objects recognized by the camera. If an object entered is not rejected by the filter and its distance (distance) provided by the distance detector is less than a substantial distance, a warning is issued.
    The sensitive distance whose downward overflow results in the issuance of a warning, is preferably chosen between 3 m and 5 m. In a particularly preferred manner, the sensitive limit is between 4 m and 4.5 m. The sensitive limit is, for example equal to 4.2 m. The higher the sensitive limit, the sooner the warning will be issued. The warning message will be issued as soon as the sensitive limit is high. The orientation of the distance detector towards the back makes it possible to capture most of the range of the blind spot with a single distance detector. At least the covered area of the blind spot range is greater than 50% so that a single monitored dead angle range detector is sufficient. In general, for each of the two sides of the vehicle, a distance detector is sufficient to monitor the associated dead angle range.
    According to the method, an object seized is rejected by filtering if the distance provided by the distance sensor between the object and the vehicle is greater than a width of the traffic corridor and if no object has been recognized by the camera in an interval distance determined from the distance provided by the distance detector. Alternatively or in addition, the objects are rejected by filtering if an object is recognized in a distance interval, determined using the camera according to the distance provided by the distance detector, but if the object is not in a traffic lane adjacent to that of the vehicle. The width of the circulation corridor may for example be fixedly predefined for the method or for example by using a detector such as a camera.
    The blind spot range is the range next to the vehicle and the driver can not see when looking through the interior rearview mirror or one of the two exterior mirrors. Thus, the dead angle range covers at least one range that extends in the longitudinal direction of the vehicle to about 3 m behind the vehicle and arrives, for example, at one of the exterior mirrors of the vehicle. Laterally, relative to the vehicle, the blind spot area extends to at least 3 m apart from the vehicle. Depending on the embodiment of the method, there will be a specific dead angle range or, for example, two dead angle ranges, one of the ranges being adjacent to the left side of the vehicle and the other angle range. dead being adjacent to the right side of the vehicle.
    An object seized is, according to the invention, an object captured by a distance detector; the seized object is in particular another vehicle, such as a truck, a passenger vehicle, a motorcycle or a cyclist.
    The distance detector is, for example, an ultrasonic detector or sensor, a radar detector or an ultrasonic lidar detector, a radar detector or a lidar detector; the ultrasonic detector is preferably selected. An ultrasonic detector thus recognizes exclusively the distance of an object, but it can not give the direction of the object relative to the vehicle. In a variant, it is possible to use a plurality of distance detectors, in particular an array of detectors. Detecting objects according to the invention consists in detecting (recognizing) objects that are in the field of view of a camera. According to the operating algorithm used for the images provided by the cameras, a detected object is for example identical to another vehicle; the detected object may also be, for example, a part of another vehicle. Thus, a detected object will correspond to an object or part of a detected object. In particular, the characteristic elements of a vehicle such as the contours or the number plate will, for example, be recognized as objects in the camera image.
    To judge whether the detected object (object entered) is actually in the blind spot range, the distance provided by the distance detector is not sufficient as a single criterion. In fact, if the determined distance is greater than the width of a traffic corridor, it will not be possible to distinguish between an object that is in the adjacent traffic lane behind the vehicle or an object that is at the height of the vehicle. but is distant from two traffic corridors. This is why it is planned to filter the objects recognized by the camera.
    According to a first filter variant, for each object entered by a distance detector, it is checked whether the distance interval determined from the distance provided by the distance detector contains an object recognized by the camera. If there is a recognized object in the distance interval, there is a high probability that the recognized object corresponds to the object entered or to a part of the object entered by the distance detector. The object entered by the distance detector is thus in a position that is both captured by the distance detector and by the camera, recognized as an object. Since the camera is facing aft, its field of view is rearward so that there is a strong likelihood that the captured object will be in the adjacent traffic lane behind the vehicle. This justifies issuing a warning because the object is in the blind spot range.
    If, conversely, from the distance range defined from the distance provided by the distance detector, the camera did not recognize any object, but, as the field of view of the camera is oriented towards the camera. aft, there is a high probability that the object entered by the distance detector is at the same height as the vehicle and is not in a neighboring traffic corridor, but at least two traffic corridors at a distance of distance which corresponds to at least two corridors of circulation. In this case, issuing a warning is not justified because the object entered is not in the blind spot range. The object is thus rejected by filtering.
    If, at the detection of objects, an association is made between the recognized objects and a corridor of circulation, the filtering of the seized objects made in addition or in a variant may reject an object seized if the camera has recognized an object within a range of distance defined from the distance provided by the distance detector and if this object is not in a traffic corridor adjacent to the vehicle. In this case also, there is a high probability that the object or a part of the object captured by the distance detector corresponds to the object recognized by the camera and is not in a dead angle range. but at a distance of at least two traffic corridors of the vehicle. In this case, issuing a warning would not be justified.
    According to the proposed filtering method, the proposed filtering method uses a distance interval. To determine the distance interval, the distance provided by the distance detector is calculated by taking into account the distance between the distance detector and the camera to apply the result to the position of the camera. Since this conversion is not necessarily unambiguous, we obtain a distance interval.
    Preferably, the distance interval is obtained using a two-dimensional model and the distance provided by the distance detector. The two-dimensional model does not take into account the "height" that is to say the direction perpendicular to the plane of the roadway. Since the distance detector only defines one direction, the object can be located on an arc defined by the opening angle of the ultrasonic detector or its field of vision and the distance obtained. The conversion of the position of this arc in the model, shifts the distance between the camera and the distance detector, in the X direction and in the Y direction. The lower limit of the interval is then given by the smallest distance of the arc from the position of the camera; the upper limit of the distance interval is given by the greater distance between the arc of the circle and the position of the camera. The defined distance interval can additionally be increased by a predefined value to compensate for measurement errors. For this, we can increase the upper limit by 1 to 20 cm and reduce the lower limit by a distance of 1 to 20 cm.
    If an object entered is not rejected by the filter, it can be assumed that the object entered is actually in a dead angle range and thus a warning is issued. The warning is, for example, in the form of an optical signal, an acoustic signal or a haptic signal; various combinations of warning signals can also be envisaged. For example, to issue a warning, you can activate a warning light, generate an audible signal and / or vibrate the steering wheel.
    The camera used for the application of the method is preferably a reversing camera. This one is in any case oriented towards the rear and equips many vehicles so that the implementation of the process does not require the installation of an additional camera.
    Preferably, objects are recognized in the field of view of the camera by analyzing the optical flow. The images generated by the camera are then analyzed to detect the movements in the images.
    It is thus provided, preferably, to recognize ranges in the field of view of the camera using the optical flow in the direction of flow it is expected to recognize as an object, ranges in the field of view of the camera resulting from the optical flow in the direction of circulation. It can be envisaged that, according to the realization of the algorithm, one recognizes as object a vehicle or that one recognizes as objects several significant areas of the vehicle in the image of the camera.
    For the application of the method as proposed, it does not matter whether the physical object, such as a vehicle, is recognized as a single object or as a set of objects. For the proposed filtering, it is sufficient that the object can be distinguished reliably from the background or with respect to stationary objects such as for example fences or trees; fixed objects have no optical flow in the direction of circulation.
    In addition, for the process it is important to determine the distance (distance) between the recognized object and the vehicle. For example, the distance of a recognized object in the field of view of the camera is calculated using its position in the field of view. For this, it is used, for example, the fact that, thanks to the known orientation of the camera, the images of the camera correspond to a known perspective. Thus and by way of example, the distance of the recognized object relative to the vehicle is even lower than the object was recognized lower in the field of view of the camera. Conversely, the distance will be greater if the object is detected higher in the upper part of the field of view of the camera. Beside the vertical position in the field of view, it is also possible to use the horizontal position in the field of view. For the horizontal position is used, for example, the fact that the distance of a known object is even larger than the object is removed from the middle.
    According to another preferred feature, in the field of view of the camera is defined at least one range of images to which is associated a traffic lane; the recognized object in the respective image range is associated with the traffic lane associated with the range of images. In this case, we also use the fact that, thanks to the known orientation of the camera, we also know the perspective of the images of the camera. Thus, in the image of the camera, each visible traffic corridor is copied in a determined range of images. Each of the image ranges can be defined as range of own images and each of these ranges of images is associated with a corridor of circulation. If only an object is recognized in such a range of images it can be assumed that the recognized object is on the traffic corridor associated with this range of images. According to other embodiments, it is conceivable to improve the association of a recognized object with a traffic corridor whose recognized position in the camera image hangs by calculation at a point on the bottom of the corridor surface. of circulation.
    To define the ranges of images, it is possible to recognize the markings of the traffic lane appearing in the field of view of the camera and / or to take into account information from a helper of a support system for keeping in the guidance system. already fitted to the vehicle. The invention also relates to a computer program that applies the method described above when the program is executed by a programmable computer installation. The computer program is, for example, a module for implementing a driver assistance system or a subsystem equipping the vehicle or an application for driving assistance functions and which is, for example, executed by a smartphone or tablet. The computer program can be stored in a machine-readable memory port such as a permanent or rewritable memory or in association with a computer installation or on a mobile medium such as CD-ROM, DVD, disk BlueRay or USB key. In addition or alternatively, the computer program may be provided for downloading from a server, for example by a data transmission network such as the Internet or a communication link such as a telephone line or a link radio. The invention also relates to a driving assistance system comprising a camera and at least one distance detector, for implementing the method described. The features developed as part of the method thus apply to the driving assistance system and conversely the characteristics of the driving assistance system apply to the method.
    Preferably, the camera is a reversing camera. In general, the reversing camera is part of another driver assistance system so that this reversing camera can be used in common.
    The distance detector is preferably an ultrasonic detector, a radar detector or a lidar detector, the ultrasonic detector being preferably selected. The distance sensor is oriented so that its field of vision is directed obliquely backwards from the vehicle. In a particularly preferred manner, the driving assistance system includes such a distance detector both on the left side of the vehicle and on its right side.
    Preferably, there is also provided a warning installation that allows to issue a warning to the driver of the vehicle. The warning installation is, for example, an optical display, a speaker or a haptic transmitter. Any combination of such means can also be considered.
    Advantages of the invention
    The method according to the invention makes it possible, by using only a distance detector and a camera, to reliably provide a blind spot warning avoiding erroneous warnings. They must be avoided because they could disturb the driver who could, out of habit, ignore repeated repetition of a justified warning.
    The proposed method applies filtering, advantageously after an object is entered by the distance detector and before generating a warning. For filtering, the recognized objects are used in the field of vision of the camera and in a particularly advantageous manner, a rearview camera that equips the vehicle anyway, which avoids the installation of other systems. The invention proposes an extremely simple filtering method and according to a variant it is checked whether the camera has also recognized an object at the same distance. According to another variant of the filtering method, the objects recognized by the camera are associated with a traffic lane and the object captured by the distance detector is checked if, at the same distance, the camera has recognized an object on the road. next traffic lane. These distance comparisons are simple to apply, so that the filtering method can be applied in a technically simple manner and does not require large computing resources. The filtering process is particularly robust and reliable because of its simplicity.
    According to the method, it is extremely easy to recognize the objects in the camera images, which is particularly advantageous. For the implementation of the filtering method it is not necessary to identify a vehicle in the images of the camera; it is sufficient to identify ranges in the camera images whose optical flow is oriented in the direction of the traffic. That such identified image ranges belong to the same object need not be confirmed so that the exploitation of the camera images is very simple and is done with little computing resources; this corresponds to a robust and reliable exploitation of camera images.
    In summary, the method according to the invention makes it possible, in a simple manner, to generate a reliable warning message if the distance obtained with the distance detector is greater than the width of a traffic corridor, since the method such as proposed, makes it possible to distinguish whether the seized object is in a neighboring traffic corridor or not. Thus, it is not necessary to wait to issue the warning message that the distance obtained is less than a width of the traffic corridor.
    drawings
    The present invention will be described hereinafter in more detail with the aid of examples of driving assistance methods and a driving assistance system shown in the accompanying drawings, in which: FIG. first traffic situation, Figure 2 shows a second traffic situation, Figure 3 shows objects recognized in a camera image, Figure 4 shows a fourth traffic situation and Figure 5 shows the association of objects recognized to a corridor of circulation.
    Description of embodiments
    In the description given hereinafter of exemplary embodiments of the invention, the same components or elements bear the same references and their description will not be repeated except for exceptions. The figures show the object of the invention only schematically.
    Figure 1 shows an extract of a roadway 4 with several corridors; the extract shows three circulation corridors 35, 36, 38 in the same direction of circulation. The traffic lanes 35, 36, 38 each have, for example, the same width 40. The right lane 35 is occupied by a vehicle 1 equipped with a driving assistance system 10. The driving assistance system 10 comprises a control device 12, a distance sensor 14 and a camera 18. The driver assistant system 10 also has a warning device 2 to warn the driver of the vehicle 1 if another vehicle is in a range of blind spot 26. The blind spot area 26 is adjacent in the exemplary embodiment shown at the left side of the vehicle 1 and extends in the longitudinal direction of the vehicle 1 three meters rearward from the rear of the vehicle. rear edge of the vehicle and forwards substantially up to the exterior rearview mirror of the vehicle 1. The width of the blind spot area 26 is comparable to the width of the traffic corridor 40; it is for example equal to 3 meters.
    The driver assistance system 10 provides the driver of the vehicle 1 an indication by the warning device 2 to the extent that an object 28 such as another vehicle 6 is in the blind spot range 26. For monitor the dead angle range 26, the distance sensor 14 is installed on the vehicle 1 so that its field of vision is oriented at a rearward angle, that is to say laterally to the rear. The distance detector 14 also makes it possible to determine the distance (distance) from the grasped object 28. For example, in the case of an ultrasonic detector 16, the distance detector 14 can not indicate the direction in which find the object 28 relative to the vehicle 1. In the example shown in Figure 1 there is shown only a blind spot area 26 on the left side of the vehicle 1; however, preferably, it is also planned to have a distance detector 14 on the right side of the vehicle 1 to allow a dead angle range on the right side of the vehicle to be monitored.
    In addition to the distance detector 14, the vehicle 1 is equipped with a camera 18, preferably in the form of a rearview camera 20 which is thus oriented towards the rear. The field of vision 24 of the rearview camera 20 is thus oriented in the direction opposite to the direction of movement of the vehicle 1. The field of vision 24 of the camera 18 contains a portion of the right circulation corridor 35 on which the vehicle is located. 1 and parts of the adjacent traffic corridor 36 and a remote circulation corridor 38.
    After the entry of the object 28 by the distance detector 14 and after having determined its distance from the vehicle 1, a filtering is applied using the objects recognized by the camera 18. The recognition of objects in the camera image 18 will be described in more detail below with reference to Figure 3.
    In the example shown in FIG. 1, the camera 18 recognizes products whose distance from the vehicle 1 corresponds to the distance of the objects 28 captured by the distance detector 14 with conversion according to the position of the camera 18. It is thus excluded that the seized object 28 is at the same height as the vehicle 1 in the remote circulation corridor 38. Thus, the filtering does not reject the seized object 28.
    In addition or alternatively, the objects recognized by the camera 18 can be associated with a traffic corridor 36, 38. In this case, by filtering, the distance of the grasped object 28 obtained by the distance detector 14 is checked. and converting this distance according to the position of the camera 18, it will be known whether the camera 18 has recognized a subject who is in the neighboring traffic corridor 36. If the camera 18 recognizes a subject in the remote circulation corridor 38, the seized object 28 will be rejected. In the example of FIG. 1, the seized object 28 is in the neighboring traffic corridor 36 and will not be rejected by the filtering.
    Thus, for the traffic situation shown in FIG. 1, the method of the invention finds that the seized object 28 is in the dead angle range 26 and a warning is issued by the warning device 2.
    FIG. 2 shows a second traffic situation in which the vehicle 1 is again in the right circulation corridor 35 of the roadway 4. In contrast to the traffic situation of FIG. 1, the ultrasonic detector 16 captures a object 28 which is not in the dead angle range 26. The seized object 28 is indeed not in the neighboring traffic corridor 36, but in a remote traffic corridor 38. Thus, the only distance provided However, the ultrasonic detector 16 does not allow the driver assistance system 10 to know whether the grasped object 28 is in the remote taxiway 38 substantially at the height of the vehicle 1 or if it is at an angle behind. the vehicle 1 in the neighboring traffic corridor 36. Therefore, by the filtering step it is checked whether the products detected by the ultrasonic detector 16 can be recognized and, after conversion of the distance depending on the position of the reversing camera 20. As in the traffic situation shown in Figure 2, this is not the case, the seized object 28 will be rejected by filtering and there will be no issue of a Warning.
    FIG. 3 shows, by way of example, the image of the rear view camera 20. The ultrasonic detector 16 supplied the seized object 28 and determined the distance between the grasped object 28 and the vehicle 1. A FIG. 3 shows the distance converted as a function of the position of the rearview camera 20 relative to the grasped object 28; this distance is indicated by an arrow with the reference 30. The grasped object 28 is another vehicle 6 which is slanted behind the vehicle 1, but nevertheless outside the dead angle range 26.
    By analyzing the optical flow of the images provided by the rearview camera 20, there are recognizable marking areas as products 32 in the image of the other vehicle 6. In order to apply the proposed method, it does not matter that the other vehicle 6 is identified as a set because it is sufficient just to recognize as products 32, image ranges, marking associated with the other vehicle 6 because, in the preferred method, it is checked only if the recognized product 32 is in the Distance interval that has been defined based on the distance determined by the distance detector 14. This considerably reduces the computing power necessary for image processing.
    FIG. 4 shows a third traffic situation which corresponds practically to the second traffic situation according to FIG. 2. Unlike this second traffic situation of FIG. 2, the other vehicle 6 is now both in the field of 22 of the ultrasonic detector 16 and also in the field of view 24 of the reversing camera 20. In these conditions, check whether the distance obtained by conversion according to the position of the reversing camera from what has been determined by the ultrasonic detector 16 would give a positive result. This is why, for filtering, the combination of the recognized products 32 with a circulation corridor 36, 38 is used in addition or alternatively. In the case of the third driving situation shown in FIG. that the known objects 32 at the distance 30 provided by the ultrasonic detector 16, we see that the products 32 recognized are associated with the remote circulation corridor 36 so that in this third traffic situation, the object 28 captured by the detector ultrasound 16 will be rejected by filtering. There will be no warning issued.
    FIG. 5 shows the age of the reversing camera 20 which corresponds very largely to the image described in relation with FIG. 3. The image of the rearview camera 20 again shows the other vehicle 6 which has been determined As an object grasped by the ultrasound detector 16 and whose distance 30 has been determined. By the analysis of the optical flux, several recognized products 32 have also been determined.
    In addition, however, it is intended to define at least one image range 34 associated with the remote traffic corridor 38 in the embodiment of FIG. 5. The recognized products 32 which are in the image range 34 are thus also associated with the remote taxiway 38. Where appropriate, to improve the accuracy of this association, it is intended to convert the position of the recognized product 32 to a point on the surface of the traffic corridor and to verify whether this point is In other embodiments, other image ranges are defined, for example, also a range of images corresponding to the neighboring traffic corridor 36, and it is associated with this range of images. neighboring traffic corridor 36. It is also possible to associate a range of images to each traffic corridor.
    NOMENCLATURE OF MAIN ELEMENTS 2 Warning installation 4 Multi-lane pavement 10 Driving assistance system 12 Control unit 14 Distance detector 18 Camera 20 Reversing camera 22 Field of view of distance detector 24 Camera viewing range 26 Blind spot 28 Item 32 Product 35, 36, 38 Traffic lane 40 Width of a traffic lane
    权利要求:
    Claims (10)
    [1" id="c-fr-0001]
    CLAIMS 1 °) Driving assistance method according to which the objects (28) are seized in the environment of a vehicle (1) and a warning is issued if an object (28) is in the range of the angle mortise (26) of the vehicle (1), comprising: a) grasping objects (28) in the field of view (22) of a distance sensor (14) fitted to the vehicle (1) and biased towards the rear relative to the vehicle (1), and determining the distance of the objects seized (28) from the vehicle (1), b) recognizing objects (32) in the field of view (24) of a camera ( 18), equipping the vehicle (1) with a rearward orientation with respect to the vehicle (1) and calculating the distance of the distance of a known object (32) from the vehicle (1) and / or associating recognizable objects (32) at a traffic lane (36, 38), c) filtering objects (28) captured by the distance detector (14) using recognized objects (32) by the camera (18), and d) issuing a warning if a seized object (28) has not been rejected by the filter and its distance provided by the distance detector (14) is less than a substantial distance, according to which i. an object seized (28) is rejected by filtering according to step c) if the distance (30) provided by the distance sensor (14) from the object (28) relative to the vehicle (1) is greater than one circulation lane width (40) and relying on the distance (30) provided by the distance detector (14), if the camera (18) has not recognized any object (42) within a range of distance determined from the distance (30) by the distance detector (14), and / or ii. an object seized (28) is rejected by the filtering according to step c) if the camera (18) has recognized an object (38) within a distance interval determined from the distance (30) provided by the sensor distance (14), and if this object is not in a traffic corridor (36) adjacent to the vehicle (1).
    [0002]
    2) Method according to claim 1, characterized in that one chooses a sensitive distance of the order of 3 m to 5 m.
    [0003]
    3) Method according to claim 1 or 2, characterized in that in step b) is recognized as objects (32) in the field of view (24) of the camera (14) by analyzing the optical flow.
    [0004]
    4) Method according to claim 3, characterized in that ranges of the field of view (24) of the camera (18) with an optical flow in the direction of flow are recognized as object (32).
    [0005]
    Method according to one of Claims 1 or 4, characterized in that the distance of an object (32) recognized in the field of view (24) of the camera (18) is calculated with the aid of its position in the field of vision (24).
    [0006]
    Method according to one of claims 1 or 5, characterized in that at least one image area (34) is defined in the field of view (24) of the camera (18), and a method is associated with corridor (36, 38) at this image range (34), and associating an object (32) recognized in the respective image range (34) with the traffic corridor (36, 38) associated with this range. of images (34).
    [0007]
    7 °) computer program executing the method according to one of claims 1 to 6, when the program is applied by a computer.
    [0008]
    8 °) driving assistance system (10) comprising a camera (14) and at least one distance detector (14), characterized in that it executes the method according to one of claims 1 to 6.
    [0009]
    9 °) driving assistance system (10) according to claim 8, characterized in that the camera (18) is a rear view camera (20).
    [0010]
    Driving assistance system (10) according to claim 8 or 9, characterized in that the distance sensor (14) is an ultrasonic detector (16).
    类似技术:
    公开号 | 公开日 | 专利标题
    JP5959581B2|2016-08-02|Vehicle periphery monitoring device
    FR2883826A1|2006-10-06|APPARATUS FOR ASSISTING THE DRIVING OF A VEHICLE
    FR3047105A1|2017-07-28|METHOD AND DEVICE FOR DRIVING ASSISTANCE
    FR2863398A1|2005-06-10|DRIVING ASSISTANCE APPARATUS FOR PREVENTING COLLISION OF VEHICLES
    FR2731515A1|1996-09-13|DEVICE FOR DETERMINING THE DISTANCE OF A VEHICLE IN RELATION TO A SIDE MARKING OF THE PAVEMENT
    EP2061006A1|2009-05-20|Method of detecting a phenomenon that can reduce visibility for a vehicle
    FR2976246B1|2019-07-05|DRIVING ASSISTANCE SYSTEM WITH DETECTION OF OBJECTS
    FR2961774A1|2011-12-30|METHOD FOR WARNING OF COLLISION RISK DRIVER
    JP2010249668A|2010-11-04|Object detecting device for vehicle
    FR2987452A1|2013-08-30|METHOD FOR DETECTING A SURROUNDING FIELD BY A DRIVING ASSISTANCE SYSTEM
    FR2941423A1|2010-07-30|METHOD AND DEVICE FOR PREVENTING THE COLLISION BETWEEN A VEHICLE AND AN OBJECT
    FR2969098A1|2012-06-22|METHOD AND DEVICE FOR ASSISTING THE DRIVER OF A VEHICLE
    FR3012784A1|2015-05-08|DEVICE FOR DETECTING THE LATERAL POSITION OF A PIETON IN RELATION TO THE TRACK OF THE VEHICLE
    FR3035364A1|2016-10-28|METHOD FOR MANAGING AN OBJECT RECOGNITION SYSTEM FOR INCLINABLE VEHICLES AND SYSTEM FOR IMPLEMENTING THE METHOD
    EP2043044B1|2010-05-26|Method and device for automobile parking assistance
    EP1785966B1|2008-12-10|Evaluation process, by a motor vehicle, of the characteristics of a frontal element
    FR2993994A1|2014-01-31|METHOD FOR MANAGING A SYSTEM FOR ENTERING THE ENVIRONMENT OF A VEHICLE WITH AT LEAST TWO TRANSMITTING / RECEIVING UNITS AND SYSTEM FOR IMPLEMENTING THE METHOD
    FR3095176A1|2020-10-23|Parking aid device and parking aid device control method
    EP1528409B1|2006-06-21|Method and system for measuring the speed of a vehicle on a curved trajectory
    WO2018041978A1|2018-03-08|Device for determining a speed limit, on-board system comprising such a device, and method for determining a speed limit
    FR2973320B1|2019-06-14|METHOD AND DEVICE FOR ASSISTING A PARKING MANEUVER OF A VEHICLE
    JP5651649B2|2015-01-14|Object position detection device
    FR3082485A1|2019-12-20|DISPLAY DEVICE FOR ASSISTING THE DRIVING OF A DRIVER OF A VEHICLE
    FR3056530A1|2018-03-30|OBSTRUCTION DETECTION BY FUSION OF OBJECTS FOR MOTOR VEHICLE
    FR2873480A1|2006-01-27|Vehicle e.g. heavy vehicle, speed monitoring system, has Doppler radar calculating speed using received signals, and category selection unit selecting vehicle category based on vehicle profile derived from distance between vehicle and radar
    同族专利:
    公开号 | 公开日
    GB201701167D0|2017-03-08|
    GB2548465A|2017-09-20|
    DE102016201070A1|2017-07-27|
    FR3047105B1|2020-12-18|
    引用文献:
    公开号 | 申请日 | 公开日 | 申请人 | 专利标题
    CN109703556A|2018-12-20|2019-05-03|斑马网络技术有限公司|Driving assistance method and equipment|
    US10578727B2|2016-09-22|2020-03-03|Robert Bosch Gmbh|Method and processing unit for detecting a wet or damp roadway and for object detection|ES2158827B1|2000-02-18|2002-03-16|Fico Mirrors Sa|DEVICE FOR DETECTION OF PRESENCE OF OBJECTS.|
    DE10305935A1|2003-02-13|2004-08-26|Valeo Schalter Und Sensoren Gmbh|Device for detecting objects in the environment of a motor vehicle|
    US7893819B2|2005-03-03|2011-02-22|Continetntal Teves AG & Co, oHG|Method and device for avoiding a collision in a lane change maneuver of a vehicle|
    DE102005027653A1|2005-06-15|2006-12-21|Robert Bosch Gmbh|Blind spot`s object detecting device, has comparison unit ascertaining, based on preset correlations between detection signals whether vehicle following host vehicle is present, and blocking output unit from outputting warning signal|
    JP4420011B2|2006-11-16|2010-02-24|株式会社日立製作所|Object detection device|
    DE102008061357A1|2008-12-10|2010-06-17|Valeo Schalter Und Sensoren Gmbh|Monitoring device and method for monitoring blind spot areas of a vehicle|
    DE102011018159A1|2011-04-19|2012-10-25|GM Global Technology Operations LLC |Device and method for driver assistance|
    DE102012204948A1|2012-03-28|2013-10-02|Robert Bosch Gmbh|Method for assisting driver when driving vehicle at changing lanes, involves assigning vehicle and detected objects to tracks of roadway, and generating warning signal when detected object regarding vehicle is located on relevant track|
    KR20140019571A|2012-08-06|2014-02-17|주식회사 만도|Blind spot warning system and method|
    DE102012215014A1|2012-08-23|2014-02-27|Robert Bosch Gmbh|Method for monitoring blind angle of vehicle, involves detecting surrounding laterally adjacent to vehicle, where objects in detected surrounding and differential speeds of objects to vehicle are determined|DE102017119036A1|2017-08-21|2019-02-21|Valeo Schalter Und Sensoren Gmbh|Avoidance of false alarms during blind spot monitoring|
    法律状态:
    2018-01-24| PLFP| Fee payment|Year of fee payment: 2 |
    2018-11-16| PLSC| Publication of the preliminary search report|Effective date: 20181116 |
    2020-01-23| PLFP| Fee payment|Year of fee payment: 4 |
    2021-01-20| PLFP| Fee payment|Year of fee payment: 5 |
    2022-01-18| PLFP| Fee payment|Year of fee payment: 6 |
    优先权:
    申请号 | 申请日 | 专利标题
    DE102016201070.0A|DE102016201070A1|2016-01-26|2016-01-26|Method and device for driver assistance|
    [返回顶部]