Method and device for recording a collision event on a body part of a vehicle

Abstract

The invention describes a method for registering a collision event on a body part of a vehicle using a sensor. According to the invention, deformations of the body part are detected by a sensor designed as a radar sensor and installed in the body part, wherein signals from the radar sensor are evaluated by a control unit using an algorithm to determine a deviation from a predetermined installation position of the radar sensor in order to detect the collision event.

Description

[0001] The present invention relates to a method, a device and a vehicle for registering a collision event on a body part

[0002] Further methods for determining the installation position of a radar sensor on a vehicle component using an alignment algorithm are known from the prior art. For example, EP 4 050 367 A1 discloses a method in which the installation position and associated deviations of a radar sensor are determined exclusively using radar signals.

[0003] With the introduction of highly automated vehicles, all collisions, even minor ones, must be reliably detectable in order to comply with UN Regulation Automated Lane Keeping Systems (ALKS) No. 157.

[0004] DE 10 2020 205 508 A1 discloses a method and device for registering and evaluating a collision event on a vehicle. For this purpose, a mechanical impulse and / or a deforming force is detected using sensors, with an evaluation and control unit configured to register and evaluate a collision event based on the detected contact with the body.

[0005] The object of the present invention is to provide an improved method and an improved device which enable the detection of collisions of a small magnitude which cannot be detected or at least can only be detected unreliably by the sensors known from the prior art.

[0006] The object is achieved by a method having the features of claim 1, a device according to claim 7, and a vehicle according to claim 10. The dependent claims define preferred and advantageous embodiments of the present invention.

[0007] In the method according to the invention, deformations of the body part are determined using a sensor designed as a radar sensor and installed in the body part, wherein signals from the radar sensor are evaluated by a control unit using an algorithm to determine a deviation from a predetermined installation position of the radar sensor in order to detect the collision event. For this purpose, the control unit can have at least one computing unit, at least one memory and at least one interface, which can be designed as hardware and / or software. Using an algorithm known from the prior art, the control unit calculates a deviation from an installation position. The installation position is defined as the position of the radar sensor in the body component at which the radar function, i.e. the detection of objects, is available without restrictions. The radar sensor is preferably calibrated in the installation position.In other words, the installation position is the target position of the radar sensors when the vehicle is new. Deviations from the installation position caused by wind, for example, in the form of twisting or shifting within a specified tolerance are permissible without restriction; deviations from the installation position that exceed the tolerance range lead to a reduction in radar function or even total failure. If the radar function is not available to a sufficient extent, at least information must be given to the vehicle user and, if necessary, the radar function must be switched off. The control unit detects deviations above a specified threshold that are not caused by wind or other normal operating conditions as a collision event, which must be saved in accordance with the ALKS. For this purpose, the installation position of the radar sensor is continuously monitored using the algorithm.In addition, a driving maneuver such as a stop may need to be initiated. Advantageously, the radar sensors designed for object detection can reliably detect deformations caused by collisions. In vehicles already equipped with radar sensors for monitoring traffic ahead and behind, collision detection can be implemented without the need for additional hardware.

[0008] In a preferred embodiment, the deviations determined by the algorithm using the radar signals are determined by changing a pitch, roll, or yaw angle of the radar sensor starting from a predefined installation position. If these values ​​are above a predefined threshold, the control unit detects the deviation as a collision event. The threshold value is determined for each angle through testing and set such that it lies above the deviations caused by normal operating conditions. This allows a deviation from the installation position above the tolerance values ​​to be attributed to a collision with high reliability.

[0009] In a further preferred embodiment, the signals relating to the deviation of the radar sensor from the installation position are fused with signals from other vehicle sensors. By merging the radar sensor signals defining the deviation with other signals, such as radar signals from the radar sensor as well as signals from a camera, an ultrasonic sensor, and / or an acceleration sensor, the type of collision can be determined more precisely and the necessary measures can be determined more specifically. For example, if the radar sensors detect a deformation and the camera detects a cyclist, it can be assumed that the cyclist has fallen as a result of the collision, and the autonomous vehicle must be stopped to provide appropriate assistance.

[0010] In a further additional or alternative embodiment, the control unit determines the severity of the collision based on the magnitude of the deviations of the radar sensors from the installation position and / or based on signals from other vehicle sensors. Advantageously, based on a correlation between the deviation and the severity of the collision, preferably in comparison with other sensor signals, a corresponding classification of the collision can be performed and stored for further processing.

[0011] In a further preferred embodiment, the driving strategy of the preferably autonomously driving vehicle is adapted depending on the severity of the collision. Depending on the determined severity, i.e. the extent of the collision, which is determined by means of the radar sensors, the driving strategy is adapted, such as a change in the trajectory, for example in order to move away from a colliding object or the initiation of a stopping maneuver. The driving strategy is preferably determined based on the deviation determined by the radar sensors and signals from other sensors. A pre-trained artificial intelligence model can determine a suitable driving strategy depending on the signals of the deviation determined by the radar sensors, the objects detected by the radar sensors and / or the signals from other sensors such as a camera for recording environmental conditions.

[0012] In a further preferred embodiment, the temporal progression of the deformation is evaluated to determine the collision. By evaluating the temporal progression, error detection can be minimized. For example, a steep gradient of the deformation measured with the pitch, roll, or yaw angles suggests a collision; a flat gradient can more easily rule out a collision. Here, too, threshold values ​​are provided, indicating the deformation gradient at which a collision is registered.

[0013] The device according to the invention comprises a radar sensor that detects deformations of the body part. A control unit evaluates signals from the radar sensor using an algorithm to determine a deviation from a predefined installation position in order to detect the collision event. The control unit uses the algorithm to detect the deviations and, based on the deviation, determines whether or not a collision between the body part and an object has occurred, preferably by comparing them with predefined threshold values. To determine the deviations, the pitch, roll, and / or yaw angles of the radar sensor, or their changes relative to the installation position, are preferably measured.

[0014] In a preferred embodiment, the device comprises, in addition to the radar sensor, several radar sensors installed on one or more body parts. Several radar sensors installed on a bumper, for example, on the right and left sides of the vehicle, enable not only an improved original object detection function but also improved detection of collision events. To detect collisions with objects from different directions, one or more radar sensors are installed, for example, in the front and rear bumpers. To enable all-round collision detection, radar sensors are also installed in the doors, side panels, or trunk lid.

[0015] The vehicle according to the invention has the device according to the invention described above and is thus configured to carry out automated driving operation in accordance with the ALKS.

[0016] Further advantages, features, and details will become apparent from the following description, in which at least one exemplary embodiment is described in detail—possibly with reference to the drawings. Described and / or illustrated features may form the subject matter of the invention alone or in any meaningful combination, possibly independently of the claims, and may, in particular, also be the subject of one or more separate applications. Identical, similar, and / or functionally equivalent parts are provided with the same reference numerals.

[0017] They show: Fig. 1 a schematic plan view of a vehicle and Fig. 2 a bumper of the vehicle Fig. 1 with a radar sensor

[0018] In Fig. 1 shows a plan view of a vehicle 2 having two radar sensors 1 on each of its bumpers 3. In addition to their object detection function, the radar sensors 1 are used to detect collisions with an object. For this purpose, the radar sensors detect signals corresponding to deformations of a body part, which in this case is designed as a bumper. The radar sensor signals are evaluated by a control unit 4 using an algorithm to determine a deviation from a predetermined installation position of the radar sensor in order to detect the collision event. The radar sensors are preferably arranged on or at least in the region of the outer end of the bumpers 3, so that deformations caused by collisions can be detected over the entire area of ​​the bumpers 3.

[0019] In the Fig. 2 is the bumper 3 from Fig. 1 as a single part. The radar sensor 1 is typically in a Fig. 1 is installed in the areas 8. In an enlarged view, the radar sensor 1 is shown in an xyz axis system as it is arranged and aligned in the assembly in one of the areas 8. Furthermore, the arrow 7 shows a force acting on the bumper 3 caused by a collision with an object such as a road user. The force causes a deformation of the bumper 3, which in turn causes a deflection of the radar sensor 1 installed in the area 8. The installation position of the radar sensor is defined by the xyz axes, i.e. the radar sensor 1 is in a target installation position required for the radar function for object detection. In other words, the axis system preferably assigned to the main axes of the radar sensor installed in the areas 8 corresponds to the xyz axes of the installation position.The exemplary deflection of radar sensor 7 caused by the force exerted results in a rotation angle about the y-axis, indicated by arrow 5 and referred to as pitch, and a rotation angle about the z-axis, indicated by arrow 6 and referred to as yaw. Due to these rotations, the x-axis of the installation position is deflected in the x' direction and the y-axis in the y' direction. The rotation or deflection of the axes is determined by control unit 4 using a prior art algorithm for determining a deviation from a predetermined installation position defined by the xyz axes. Depending on the point of application of the force, a rotation around the x-axis (not shown) or a deflection of the z-axis can also occur, referred to as rolling.

[0020] The control unit 4 monitors the twist angles and as soon as one of the twist angles 5,6 exceeds a predetermined threshold value, the control unit 4 detects a collision and initiates the measures required for ALKS and / or measures to change the driving strategy.

[0021] In addition to the signals provided by the radar sensors, signals from other sensors, for example those in Fig. The camera 9 shown in Figure 1 is evaluated by the control unit 4, so that in addition to detecting a collision, an associated object causing the collision can be determined. This allows subsequent measures to be initiated following a collision, taking the colliding object into account. QUOTES CONTAINED IN THE DESCRIPTION

[0000] This list of documents submitted by the applicant was generated automatically and is included solely for the convenience of the reader. This list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions. Cited patent literature

[0000] EP 4 050 367 A1

[0002] DE 10 2020 205 508 A1

[0004] Cited non-patent literature

[0000] Introduction of highly automated vehicles must comply with UN Regulation Automated Lane Keeping Systems (ALKS) No. 157

[0003]

Claims

[1] Method for registering a collision event on a body part (3) of a vehicle (2) with a sensor, characterized by that deformations of the body part (3) are determined using a sensor designed as a radar sensor (1) installed in the body part (3), wherein signals from the radar sensor (1) are evaluated by a control unit (4) using an algorithm to determine a deviation from a predetermined installation position of the radar sensor (1) for detecting the collision event. [2] Method according to claim 1, characterized by that the deviations of a pitch, roll and / or yaw angle of the radar sensor (1) from a predetermined installation position above a predetermined threshold value, determined by the algorithm using the radar signals, are detected by the control unit (4) as a collision event. [3] Method according to claim 1 or 2, characterized bythat the signals for the deviation of the radar sensor from the installation position are fused with signals from other vehicle sensors (9). [4] Method according to one of claims 1 to 3, characterized by that the severity of the collision is derived by means of the control unit (4) based on the extent of the deviations of the radar sensors from the installation position and / or signals from other vehicle sensors. [5] Method according to claim 4, characterized by that the driving strategy of the preferred autonomous vehicle (2) is adapted depending on the severity of the collision. [6] Method according to one of claims 1 to 5, characterized by that the temporal course of the deformation is evaluated to determine the collision. [7] Device for registering and evaluating a collision event on a body part of a vehicle (2) with a sensor according to the method according to one of claims 1 to 6, characterized bythat a radar sensor (1) detects deformations of the body part, wherein a control unit (4) evaluates signals from the radar sensor (1) with an algorithm for determining a deviation from a predetermined installation position in order to detect the collision event. [8] Device according to claim 7, characterized by that in order to determine the deformation of the body part, further radar sensors are arranged on the body part (3) in addition to the radar sensor (1). [9] Device according to claim 8, characterized by that the body part (3) includes the bumper, door and / or trunk lid. [10] Vehicle with a device according to one of claims 7 to 9.

Images