The invention relates to a driver assistance system for at least partially decelerating a motor vehicle, including a detecting device for detecting an obstacle in an environmental region in direction of travel in front of the motor vehicle and for sensing a distance to the obstacle, i.e. a distance in direction of travel between the obstacle and the motor vehicle. In addition, the driver assistance system includes an electronic control device, which is able to output control signals to a braking system of the motor vehicle depending on the sensed distance and hereby to cause the at least partial deceleration. In addition, the invention relates to a motor vehicle with such a system as well as a corresponding method.

Driver assistance systems for automatically decelerating a motor vehicle are already prior art. In such systems, by means of a detecting device - usually an infrared sensor (IR) - the distance to an obstacle, typically to another vehicle located in front of the motor vehicle, is sensed. Depending on the measured distance, then, for example, the time to collision (TTC) can be calculated, and depending on the TTC, an automatic braking operation of the motor vehicle can be initiated, if applicable. If the TTC falls below a preset threshold, the deceleration is initiated. Therein, it is differentiated between fully automatic systems on the one hand as well as semi-autonomous systems on the other hand. In fully autonomous systems, the braking operation is completely controlled by the driver assistance system without the driver having to intervene. In comparison, in semi- autonomous systems, the motor vehicle is only partially decelerated by the control device, and an optical and/or haptic and/or acoustic warning signal is additionally output to the driver. Then, the driver himself can complete the braking operation and completely brake the motor vehicle by means of the brake pedal or else allow the driver assistance system to reduce the impact energy upon collision with the obstacle. Therein, the present invention can be applied to both types of systems.

By such driver assistance systems, in particular rear-end collisions from behind and thus collisions with a preceding vehicle can be prevented. An infrared braking system for automatically decelerating a motor vehicle is already known for example from the document DE 40 1 1 403 A1 . The configuration of an infrared sensor for detecting preceding vehicles in turn is known from the printed matter EP 0 380 797 B1. The driver assistance systems known from the prior art for automatically decelerating a motor vehicle also have disadvantages: Namely, it has turned out that in some cases the automatic deceleration is initiated by the driver assistance system without a collision with an obstacle impending.

It is an object of the invention to improve the reliability of a driver assistance system of the initially mentioned kind compared to the prior art.

According to the invention, this object is solved by a driver assistance system, by a motor vehicle as well as by a method having the features according to the respective

independent claims. Advantageous implementations of the invention are the subject matter of the dependent claims, of the description and of the figures.

A driver assistance system according to the invention is formed for at least partially decelerating a motor vehicle. The driver assistance system can be a semi-autonomous or a fully autonomous system. The driver assistance system includes a detecting device for detecting an obstacle in an environmental region in direction of travel in front of the motor vehicle. The detecting device senses a distance to the obstacle. The driver assistance system also has an electronic control device, which is able to output control signals to a braking system of the motor vehicle depending on the sensed distance. By means of the control signals, the at least partial deceleration of the motor vehicle is caused. According to the invention, the driver assistance system has a sensing device for providing sensor data with respect to the environmental region as well as verification means, which verify if the obstacle is a predefined object based on the sensor data, and prevent the output of the control signals to the braking system if the obstacle is the predefined object.

The invention is based on several realizations: First, it is based on the realization that the automatic or semi-autonomous braking systems known from the prior art do not correctly function in some situations and for example initiate the automatic braking operation although a collision with an obstacle is not impending. The invention is based on the realization that such situations for example occur at tollbooths or else other entry and exit regions, at which the motor vehicle has to pass a gate or else another barrier. Because the detecting device is typically formed as an infrared sensor and is placed behind the windshield, such a gate is usually interpreted as an obstacle and the motor vehicle is decelerated. Further, the invention is based on the realization that these disadvantages of the prior art can be avoided in that verification means are provided, which make plausible the braking operation depending on sensor data of a sensing device and herein verify whether or not the detected obstacle is a preset object (for example a gate). If it is recognized that the obstacle is the preset object, the automatic braking operation can be prevented. By such an approach, the reliability of the driver assistance system is improved compared to the prior art, and unnecessary braking operations are avoided.

In an embodiment, it is provided that the predefined object is a gate. Thus, the verification means verify whether or not the detected obstacle in front of the motor vehicle is a gate. If such a gate is identified in the sensor data of the sensing device, the automatic deceleration is prevented. This embodiment proves - as explained - particularly advantageous for example at tollbooths, at which the motor vehicle typically has to pass a toll gate. Such tollbooths are typically located on highways. For example, a gate can also be present at an entry into a garage or else upon exit from a garage.

However, the invention is not restricted to the identification of a gate as the object, and can also be applied to other objects such as for example a portal or the like. Generally speaking, preferably, a barrier or a fence or a blockade, which is movable between a blocking position and an unblocking position, is understood by the "predetermined object".

Preferably, the detecting device is an infrared sensor. Such sensors for detecting obstacles have already proven themselves. With an infrared sensor, the distance to the obstacle and thus the TTC can be particularly fast measured such that the response time of the driver assistance system is also very low. The infrared sensor can for example be placed behind the windshield of the motor vehicle.

It proves advantageous if the sensing device for providing the sensor data is a sensor device separate from the detecting device. With a sensor device different from the detecting device, namely, it can be detected whether or not the obstacle is the preset object with regard to plausibility check.

Particularly preferably, the sensing device includes a camera, which provides images of the environmental region as sensor data. The verification means can then identify the predefined object in the images. Based on images of a camera, for example, it is possible to find characteristic points or features in the images, and optical flow vectors can be calculated to the characteristic points, which allow tracking the image points over a sequence of images. Optionally, it is possible to perform pattern recognition with regard to the predefined object and thus to reliably identify a certain geometric shape. Such pattern recognition is already used for detection of vehicles or pedestrians. The predetermined object can therefore be particularly reliably and fast detected in the images. By means of a camera, thus, the object can be uniquely detected such that errors in verification can be prevented.

For example, the camera can be a video camera. It can be a CCD camera or a CMOS camera.

As the camera, preferably, a front camera is used, which can be placed behind the windshield and captures the environmental region in front of the motor vehicle. Such cameras are already prior art and are usually employed to automatically control the headlights of the motor vehicle and/or to detect pedestrians or other obstacles, which are located in front of the motor vehicle.

Based on the images, the verification means can determine the current position of the object relative to the motor vehicle and determine whether or not the object is the obstacle, depending on the current relative position of the object as well as depending on the distance to the obstacle measured by means of the detecting device. Based on images of a camera, namely, it can be determined with high accuracy, in which position the identified object is located with respect to the motor vehicle. The object can even be tracked in the images of the camera such that the respectively current position of the object can be correlated with the measured distance to the obstacle.

For the identification of the object in the images, first, characteristic features - such as for example edges and/or corners - can be detected in the images. As the characteristic features, for example, the so-called Harris points can be detected (C. Harris and M.

Stephens„A combined corner and edge detector" In Proceedings of the 4 ^th Alvey Vision Conference, pages 147_151 , 1988). Then, optical flow vectors can be calculated to the characteristic features according to the Lucas Kanade method, and a tracking algorithm can be performed (Bruce D. Lucas and Takeo Kanade„An Iterative Image Registration Technique with an Application to Stereo Vision" pages 674_679, 1981 ). In addition, the movement of the motor vehicle can also be determined, wherein for example operating parameters like the current velocity and/or the current yaw rate and/or the current pitch angle of the motor vehicle can be determined. These operating parameters can for example be picked up on the CAN bus. This vehicle movement can then be transformed into the coordinate system of the camera considering calibration data of the camera. Depending on the optical flow vectors as well as depending on the vehicle movement, then, the current three-dimensional position of the characteristic points can be determined (H.C. Longuet Higgins and K. Prazdny„The Interpretation of a Moving Retinal Image" RoyalP, B - 208:385_397, 1980). This method proves particularly advantageous in particular with static objects such as a gate. Namely, it functions particularly reliably with characteristic points associated with a static and thus stationary object. Therein, moved characteristic points can be filtered out. In this way, it is ensured that the automatic deceleration is only prevented in case of the predetermined object, and the motor vehicle is reliably decelerated in case of other obstacles.

A method for detecting guardrails based on images of a camera is already prior art and for example described in the document US 2012/0069185 A1.

Additionally or alternatively, the sensing device can have at least one distance sensor, in particular at least one ultrasonic sensor and/or at least one radar sensor and/or at least one lidar sensor (for example a laser scanner), for providing the sensor data. Based on distance values of a distance sensor too, namely, it can be determined with sufficient accuracy if the detected obstacle is the preset object such as for example a gate or the like. Namely, the mentioned distance sensors are typically disposed on the front bumper of the motor vehicle, such that in absence of measured values or in absence of detection it can be inferred with large probability that the obstacle is a gate disposed on a certain level above the ground. If the motor vehicle moves towards a gate, this gate is exclusively detected by the detecting device, while the at least one distance sensor is not able to sense the gate and thus does not provide any distance values. By means of this plausibility check, thus, the presence of a gate in front of the motor vehicle can be reliably inferred.

In an embodiment, it is provided that the verification means interpret the obstacle as the predefined object, in particular as a gate, and prevent the output of the control signals to the braking system if the at least one distance sensor does not detect any object and thus does not provide measured values after detection of the obstacle by the detecting device. This embodiment can also be combined with the object recognition by means of the camera, in order to once again make plausible the decision, whether or not the obstacle indeed is the predefined object. Therein, the advantage of a distance sensor is in that already present distance sensors can be used, such as for example the ultrasonic sensors of a parking assistance usually present anyway. In addition, the invention relates to a motor vehicle with a driver assistance system according to the invention.

A method according to the invention serves for operating a driver assistance system for at least partially decelerating a motor vehicle, wherein an obstacle in an environmental region in direction of travel in front of the motor vehicle is detected by means of a detecting device and a distance to the obstacle is sensed. By means of a control device, control signals are output to a braking system of the motor vehicle depending on the sensed distance, and by means of the control signals, the at least partial deceleration of the motor vehicle is caused. By means of a sensing device, sensor data with respect to the environmental region is provided, wherein verification means verify if the obstacle is a predefined object based on the sensor data, and prevent the output of the control signals to the braking system if the obstacle is the predefined object.

The preferred embodiments presented with respect to the driver assistance system according to the invention and the advantages thereof correspondingly apply to the motor vehicle according to the invention as well as to the method according to the invention.

Further features of the invention are apparent from the claims, the figures and the description of figures. All of the features and feature combinations mentioned above in the description as well as the features and feature combinations mentioned below in the description of figures and/or shown in the figures alone are usable not only in the respectively specified combination, but also in other combinations or else alone.

Now, the invention is explained in more detail based on a preferred embodiment as well as with reference to the attached drawings.

There show:

Fig. 1 in schematic illustration a motor vehicle with a driver assistance system according to an embodiment of the invention;

Fig. 2 in schematic illustration a block diagram of the driver assistance system;

Fig. 3 an exemplary image of a camera; and

Fig. 4 a flow diagram of a method according to an embodiment of the invention. A motor vehicle 1 illustrated in Fig. 1 is for example a passenger car. The motor vehicle 1 includes a driver assistance system 2, which is an automatic or semi-autonomous brake assist system. The driver assistance system 2 serves for at least partially decelerating the motor vehicle 1 . Therein, a control device 3 is electrically coupled to a braking system 4 of the motor vehicle 1. The control device 3 can output control signals 5 to the braking system 4, by means of which the automatic braking operation of the motor vehicle 1 is caused. Due to the control signals 5, namely, the braking system 4 can at least partially decelerate the motor vehicle 1.

In addition, a detecting device formed as an infrared sensor 6 in the embodiment is associated with the driver assistance system 2. The infrared sensor 6 is placed behind a windshield 7 and captures an environmental region 9 in direction of travel 8 or in vehicle longitudinal direction in front of the motor vehicle 1 . The infrared sensor 6 can measure a distance 10 to an obstacle 1 1 located in the environmental region 9 in front of the motor vehicle 1 . For example, the obstacle 1 1 can be a preceding vehicle. The infrared sensor 6 communicates the current distance 10 to the control device 3, which determines the time to collision TTC depending on the distance 10 and on the relative speed between the obstacle 1 1 and the vehicle 1 and generates the control signals 5 depending on the TTC. If the TTC for example falls below a preset threshold, the automatic deceleration is initiated.

Besides the infrared sensor 6, the driver assistance system 2 additionally includes a sensing device, which provides sensor data with respect to the environmental region 9. A camera 12 and/or distance sensors 13 are associated with the sensing device. In the embodiment, preferably, at least the camera 12 is employed, while the distance sensors 13 can only optionally be used.

The camera 12 is a front camera placed behind the windshield 7 and providing images of the environmental region 9. These images are communicated to verification means 14, which optionally also receive the distance values of the distance sensors 13. The verification means 14 and the control device 3 can optionally be integrated in a common component, in particular a common microprocessor 15.

The distance sensors 13 can for example include ultrasonic sensors and/or lidar sensors and/or radar sensors. An exemplary block diagram of the driver assistance system 2 is illustrated in Fig. 2. The infrared sensor 6 (IR) communicates its information, namely the measured distance 10 to the obstacle 1 1 , both to the verification means 14 and to the control device 3. The verification means 14 additionally receive the images of the camera 12 (K) as well as the distance values of the distance sensors 13. The verification means 14 verify whether or not the detected obstacle 1 1 is a predefined object. If it is recognized that the obstacle 1 1 is the preset object, preventing signals 16 are output to the control device 3, which prevent the output of the control signals 5 to the braking system 4 (B).

For example, the predefined object is a gate 18 as it is presented in the exemplary camera image 17 according to Fig. 3. This gate 18 is detected in the images of the camera 12 with the aid of the verification means 14, and it is verified whether or not the obstacle 1 1 is the gate 18. If this is affirmed, the preventing signals 16 are output.

For detecting the gate 18, the images of the camera 12 can be subjected to pattern recognition with regard to characteristic features. Then, optical flow vectors can be calculated to the characteristic features, and the flow vectors can be combined to a common object in a manner known per se. The flow vectors of a common object can then be tracked in the sequence of images such that the respectively current position of the gate 18 with respect to the motor vehicle 1 is known. For determining the current position of the gate 18, operating parameters of the motor vehicle 1 can also be used, as in particular the current velocity and/or the yaw rate and/or the pitch angle. Calibration data of the camera 12 can also be used.

If the gate 18 is detected in the images, in the further step, it is verified if the gate 18 indeed is the obstacle 1 1 . For this purpose, the current position of the gate 18 is compared to the distance 10 to the obstacle 1 1. Only if it is determined that the obstacle 1 1 is the gate 18, the preventing signals 16 are generated.

For making plausible this decision, the distance values of the distance sensors 13 can also be used. Herein, the verification means 14 can verify whether or not the distance sensors 13 have detected an object. In case of a gate 18 located outside of the capturing ranges of the distance sensors 13 disposed on the bumper, measured values are not provided by the distance sensors 13. This represents a unique indication that the obstacle 1 1 is a gate 18. A flow diagram of a method according to an embodiment of the invention is illustrated in Fig. 4. The method starts in a first step S1. In a second step S2, the infrared sensor 6 detects the obstacle 1 1. In a subsequent step S3, then, the current distance 10 to the obstacle 1 1 is determined. In a still further step S4, the TTC is calculated from the distance 10 and from a relative speed between the obstacle 1 1 and the vehicle 1 . Parallel, in a step S5, images 17 are captured by means of the camera 12. In a following step S6, the predetermined object 18 is identified in the images 17. In a further step S7, the current position of the object 18 is determined and the object 18 is tracked. In a step S8, then, it is verified whether or not the object 18 and the obstacle 1 1 are one and the same item. If the object 18 is the obstacle 1 1 , in a step S9, the preventing signals 16 are output to the control device 3. If preventing signals 16 are not output, in a step S10, the output of the control signals 5 to the braking system 4 and thus deceleration of the motor vehicle 1 is effected if the TTC is smaller than a preset threshold value.

Optionally, in the step S8, it can also be verified whether or not the distance sensors 13 have detected an object. The preventing signals 16 can be output only if the distance sensors 13 have not detected any object and the obstacle 1 1 is therefore a gate 18 located above the ground.