Method for determining the type of vehicle collision

Abstract

To provide a method for determining a type of collision of a vehicle.SOLUTION: There are provided a method for determining a type of collision of a vehicle (1), and a sensor configuration (10) having evaluation for executing the method and a control unit (12). The method includes the steps of: capturing at least one time profile of at least one motion information (DR, 210, 212 and 213) in the vehicle (1); determining the time profile of the treated rotation motion information from at least the one captured time profile of at least the one motion information (DR, 212 and 213); generating at least one evaluation information by evaluating the time profile of the treated rotation motion information before expiration of at least one predetermined end reference; and determining a type of collision on the basis of at least the one evaluation information.SELECTED DRAWING: Figure 1

Description

[Technical Field]

[0001] The present invention relates to a method for determining the type of vehicle collision, a sensor arrangement with an evaluation and control unit for carrying out the method, and a corresponding computer program product. [Background technology]

[0002] Vehicles are known in the prior art in which a frontal collision between the vehicle and an obstacle is detected by an acceleration sensor installed in a central airbag control device. Recognizing a partial overlap (so-called offset crash) crash is more difficult than a full overlap crash because in the case of a partial overlap, only a portion of the crush zone is loaded, resulting in smaller forces and therefore smaller accelerations. One possibility for improving the detection of an offset crash is a so-called upfront sensor located at the front of the vehicle. Another possibility is the use of a two-axis acceleration sensor at the vehicle's side periphery, for example, at the B-pillar. Peripheral acceleration sensors (PAS) measuring in the lateral direction are usually installed there to detect side crashes, so extending the system to two measurement axes (lateral and longitudinal) does not require much additional cost.

[0003] German Patent Application Publication No. 102009054473 discloses a method for determining the type of vehicle collision. The method is based on the use of a two-channel rim acceleration sensor, which is arranged, for example, on the B-pillar of the vehicle and captures a first translational acceleration in the longitudinal direction of the vehicle and a second translational acceleration in the lateral direction of the vehicle. Signal preprocessing is then performed, for example, low-pass filtering of the signals provided by the rim acceleration sensor. The accelerations obtained from the two individual channels of the rim sensor are then calculated. The calculation is based on vector addition. The vector addition of linearly independent sensor signals allows for the calculation of an acceleration signal obtained from the linearly independent sensor signals. Signal processing can then be performed, for example, by integration or difference calculation, and a subsequent algorithm for recognizing the collision type can be implemented. The algorithm can be a special logic with a threshold comparison. This method allows for distinguishing a full-overlap frontal collision between a vehicle and an obstacle from a partial-overlap or angled frontal collision. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] German Patent Application Publication No. 102009054473 Summary of the Invention [Means for solving the problem]

[0005] The method for determining the type of vehicle crash comprising the features of independent claim 1 and the sensor arrangement with an evaluation and control unit for performing the method comprising the features of independent claim 15 each have the advantage that a better discrimination between different offset crashes is enabled, thereby enabling an optimized control of restraint measures for the respective situation, including the deployment time of the airbags of the corresponding restraint system and the delay times for "adaptive" restraint measures such as belt force limiters and airbag valves.

[0006] It is now possible to recognize and distinguish not only full-overlap crashes, known as Offset Deformable Barrier (ODB) crashes, and 30° angle crashes, but also further offset crash scenarios, such as: IIHS Small Overlap Crash at 64 km / h and 25% overlap, NHTSA Oblique Crash OMDB (Offset Mobile Deformable Barrier) at 90 km / h, 35% overlap, and 15° angle, and Euro-NCAP MPDB (Moving Progressive Deformable Barrier) at 50 km / h vehicle and barrier speeds and 50% overlap, respectively. This means that it is now possible to distinguish between different offset crashes, in particular crashes with small lateral movements (e.g., ODB) and crashes with strong lateral movements of the front of the vehicle (e.g., Small Overlap).

[0007] Different offset crashes can be distinguished by evaluating the time profile of the rotational acceleration. The rotational acceleration, which corresponds to the change in angular velocity around the vehicle's vertical axis, can be reconstructed, for example, by a yaw rate sensor centrally located in the region of the vehicle's center of gravity, or preferably by the difference in longitudinal acceleration measured at the vehicle's side edges. In this case, the edge sensor is preferably used on the B-pillar of the vehicle. Alternatively, sensors can be used on the A- or C-pillar. Advantageously, an embodiment of the method according to the invention allows the use of a single-channel edge acceleration sensor, since only the longitudinal acceleration captured at the left edge of the vehicle and the longitudinal acceleration captured at the right side of the vehicle need to be evaluated to distinguish between different crash types.

[0008] An embodiment of the present invention provides a method for determining a type of collision of a vehicle, the method including the steps of: capturing at least one time profile of at least one motion information of the vehicle; determining a time profile of processed rotational motion information from the at least one captured time profile of the at least one motion information; generating at least one evaluation information by evaluating the time profile of the processed rotational motion information before expiration of at least one predetermined termination criterion; and determining a type of collision based on the at least one evaluation information.

[0009] Furthermore, a sensor arrangement for a vehicle is proposed, comprising at least one sensor for capturing a time profile of at least one piece of movement information in the vehicle, and an evaluation and control unit for receiving the captured time profile of the at least one piece of movement information, wherein the evaluation and control unit is configured to perform the steps of the method for determining the type of collision of a vehicle according to the invention.

[0010] For the algorithmic evaluation of the captured time profile of at least one motion information in the vehicle, it is advantageous to subject the measured motion signals to signal processing such as low-pass filtering, band-pass filtering, integration, double integration, differentiation, etc., and use the processed motion information.

[0011] In this specification, the evaluation and control unit can be understood as a control device, particularly an electrical device such as an airbag control device, that processes or evaluates the acquired sensor signals. The evaluation and control unit can have at least one interface, which can be configured as hardware and / or software. In a hardware configuration, the interface can be, for example, part of a so-called system ASIC that includes various functions of the evaluation and control unit. However, the interface can also be a separate integrated circuit or at least partially comprised of discrete components. In a software configuration, the interface can be, for example, a software module that resides on a microcontroller together with other software modules. A computer program product with program code stored on a machine-readable carrier, such as a semiconductor memory, hard disk memory, or optical memory, and used to perform the evaluation when the program is executed on the evaluation and control unit, is also advantageous.

[0012] The measures and developments set out in the dependent claims make it possible to advantageously improve the method set out in independent claim 1 for determining the type of vehicle collision and to advantageously improve the sensor arrangement for a vehicle set out in independent claim 16.

[0013] It is particularly advantageous to determine a time profile of processed translational movement information from the at least one captured time profile of the at least one movement information, which is further evaluated to generate at least one evaluation information. Here, to generate the at least one evaluation information, a time profile of synthesized movement information can be generated by combining the time profile of the processed rotational movement information and the time profile of the processed translational movement information. The synthesized movement information may preferably correspond to a quotient of the processed rotational movement information divided by the processed translational movement information.

[0014] In an advantageous embodiment of the method, when determining the type of collision, it is possible to recognize whether the collision between the vehicle and the obstacle is a full-overlap frontal collision, an offset collision with a uniform rotation, an offset collision with a strong lateral component, or an offset collision with a strong lateral component and a catch on the obstacle. Thus, it is possible to recognize, for example, on which side of the vehicle the collision occurred, whether the vehicle hit an obstacle such as a barrier or another vehicle obliquely or head-on, whether a full-overlap or partial overlap occurred, and whether a counterclockwise or clockwise rotation occurred.

[0015] In a further advantageous embodiment of the method, the at least one predetermined termination criterion may correspond to a threshold criterion. In this case, the first termination criterion may correspond, for example, to an acceleration criterion that is met when the captured and / or processed acceleration information reaches or exceeds a predetermined first threshold. For example, the second termination criterion may correspond to a time criterion that is met when a predetermined time window is exceeded.

[0016] In a further advantageous embodiment of the method, the first captured motion information may correspond to a rotational velocity about the vehicle's vertical axis captured in the center of the vehicle's center of gravity, and the second captured motion information may correspond to a translational acceleration information captured in the vehicle's longitudinal direction. To this end, the sensor arrangement may include a first sensor positioned in the center of the vehicle's center of gravity and capturing the rotational velocity about the vehicle's vertical axis as the first motion information. To this end, a second sensor positioned in the center of the vehicle's center of gravity may capture translational acceleration information in the vehicle's longitudinal direction as the second motion information. Here, a time profile of the processed rotational motion information may be determined from the first motion information. A time profile of the processed translational motion information may be determined from the captured second motion information or from processed second motion information generated from the second motion information.

[0017] Additionally or alternatively, the third captured motion information may correspond to effective total acceleration information in the vehicle longitudinal direction captured at the left periphery of the vehicle, and the fourth captured motion information may correspond to effective total acceleration information in the vehicle longitudinal direction captured at the right periphery of the vehicle. To this end, the first acceleration sensor may be disposed at the left periphery of the vehicle, and effective total acceleration information in the vehicle longitudinal direction may be captured as the third motion information. The second acceleration sensor may be disposed at the right periphery of the vehicle, and effective total acceleration information in the vehicle longitudinal direction may be captured as the fourth motion information. Here, the time profile of the processed rotational motion information may be determined by combining the third motion information and the fourth motion information, or by combining processed third motion information generated from the third motion information and processed fourth motion information generated from the fourth motion information. The synthesis may correspond to, for example, calculating the difference between the third motion information and the fourth motion information, or between the processed third motion information and the processed fourth motion information, or calculating the ratio between the third motion information and the fourth motion information, or between the processed third motion information and the processed fourth motion information. The time profile of the processed translational motion information may be determined by calculating the sum from the third motion information and the fourth motion information, or from the processed third motion information and the processed fourth motion information.

[0018] In a further advantageous form of the method, when determining the type of collision, the processed rotational motion information or the combined motion information can be compared with at least one threshold value in order to recognize their dynamic behavior and therefore the type of collision.

[0019] In a preferred embodiment of this method, if the processed rotational motion information or the resultant motion information does not exceed a predetermined positive first threshold or fall below a predetermined negative second threshold, a full-lap frontal collision can be recognized as the type of collision. Alternatively, if the processed rotational motion information or the resultant motion information exceeds a predetermined positive first threshold, a left-side offset collision with a uniform rotation can be recognized. If the processed rotational motion information or the resultant motion information falls below a predetermined negative second threshold, a right-side offset collision with a uniform rotation can be recognized. If the processed rotational motion information or the resultant motion information first exceeds a predetermined positive third threshold and then falls below a predetermined negative fourth threshold, a left-side offset collision with a strong lateral component can be recognized. If the processed rotational motion information or the resultant motion information first falls below a predetermined negative fifth threshold and then exceeds a predetermined positive sixth threshold, a right-side offset collision with a strong lateral component can be recognized. A left-side offset collision with a strong lateral component and obstacle catch can be recognized when the processed rotational motion information or resultant motion information first exceeds the positive third threshold, then falls below the negative fourth threshold, and then exceeds the predetermined positive seventh threshold.A right-side offset collision with a strong lateral component and obstacle catch can be recognized when the processed rotational motion information or resultant motion information first falls below the negative fifth threshold, then exceeds the positive sixth threshold, and then falls below the predetermined negative eighth threshold.

[0020] Exemplary embodiments of the invention are illustrated in the drawings and explained in more detail in the following description, in which like reference numerals indicate components or elements that perform the same or similar functions. [Brief explanation of the drawings]

[0021] [Figure 1] 3 is a schematic block diagram of a vehicle including an exemplary embodiment of a sensor arrangement according to the invention with an evaluation and control unit for carrying out the method from FIG. 2; [Figure 2] 1 is a schematic flow chart of an exemplary embodiment of a method according to the present invention for determining the type of vehicle crash; [Figure 3] 10A-10C show various graphs of the processed measurement signal for a first crash type; [Figure 4] 10A-10C show various graphs of the processed measurement signal for a second crash type; [Figure 5] 10A-10C show various graphs of the processed measurement signal for a third crash type; [Figure 6] 10A-10C show various graphs of the processed measurement signal for a fourth crash type; DETAILED DESCRIPTION OF THE INVENTION

[0022] As can be seen from Figure 1, the illustrated exemplary embodiment of a sensor arrangement 10 according to the invention for a vehicle 1 comprises at least one sensor 14, 14L, 14R for capturing a time profile of at least one piece of movement information DR, 210, 212, 213 in the vehicle 1, and an evaluation and control unit 12 for receiving the captured time profile of the at least one piece of movement information DR, 210, 212, 213. The evaluation and control unit 12 is configured to perform the steps of a method 100 for determining the type of collision of the vehicle 1, which will be described below with reference to Figures 2 to 6.

[0023] As can be further seen from FIG. 1 , the sensor arrangement 10 in the illustrated exemplary embodiment includes a first sensor 14A, which is arranged centrally in the region of the center of gravity SP of the vehicle 1 and is designed as a yaw rate sensor, capturing the rotational speed around the vehicle vertical axis z as first movement information DR, where the first movement information DR relates to a mathematically positive rotation in the counterclockwise direction or a mathematically negative rotation in the clockwise direction (indicated by the dashed line). Furthermore, a second sensor 14B, which is arranged centrally in the region of the center of gravity SP of the vehicle 1 and is designed as an acceleration sensor, captures translational acceleration information in the vehicle longitudinal direction x as second movement information 210. In the illustrated exemplary embodiment, the two sensors 14A, 14B are arranged in the evaluation and control unit 12. Alternatively, the two sensors 14A, 14B can also be arranged external to the evaluation and control unit 12.

[0024] In addition to or as an alternative to the first sensor 14A designed as a yaw rate sensor and the second sensor 14B designed as an acceleration sensor, peripheral acceleration sensors 14L, 14R, which are sensitive in the longitudinal direction x, are preferably provided to capture motion information 212, 213. The illustrated sensor arrangement 10 thus includes a first acceleration sensor 14L arranged on the left periphery of the vehicle 1 and a second acceleration sensor 14R arranged on the right periphery of the vehicle 1. As can be further seen from FIG. 1 , the two peripheral acceleration sensors 14L, 14R in the illustrated embodiment of the sensor arrangement 10 are designed as two-channel acceleration sensors. Here, the first acceleration sensor 14L captures effective total acceleration information in the vehicle longitudinal direction x as third motion information 212, and also captures effective total acceleration information in the vehicle lateral direction y, which is not further evaluated here and includes a rotational acceleration component 218 in the y-direction on the left side of the vehicle. The second acceleration sensor 14R captures effective total acceleration information in the vehicle longitudinal direction x as fourth movement information 213 and also captures effective total acceleration information in the vehicle lateral direction y, which is not further evaluated here and includes a rotational acceleration component in the y direction on the right side of the vehicle 219. Since in the illustrated exemplary embodiment only the acceleration component in the x direction is used for the evaluation, the two acceleration sensors 14L, 14R can also be designed as single-channel acceleration sensors aligned in the x direction.

[0025] The motion of the rigid body or vehicle 1 can be decomposed into a translation of the center of gravity SP of the vehicle 1 and a rotation around the vehicle vertical axis z at the center of gravity SP of the vehicle 1. When limited to in-plane motion, the translation can be characterized by an acceleration vector in the x / y plane (x: longitudinal direction, y: lateral direction). Because the second sensor 14B is located near the vehicle center of gravity SP, it essentially measures the effective translational acceleration as the second motion information 210. The rotation around the center of gravity SP is characterized by a rotation vector or rotation velocity around the vehicle vertical axis z, captured by the first sensor 14A. If the rotation occurs only due to a collision, the angular acceleration, i.e., the time derivative of the rotation vector or rotation velocity around the vehicle vertical axis z, is important. The time derivative of the first motion information DR, captured as the rotation vector or rotation velocity around the vehicle vertical axis z, corresponds to the rotational acceleration, and from the rotational acceleration, processed rotational motion information FeaR can be generated by processing such as filtering. Other than at the point of rotation or center of gravity SP of the measurement location indicated by the vector r, this angular acceleration is represented by a tangentially directed rotational acceleration 214, 215, which in Figure 1 is represented with respect to a mathematically positive rotation of the vehicle 1 counterclockwise about the vehicle vertical axis z, which occurs, for example, in a left-side offset collision with a uniform rotation.

[0026] For the two acceleration sensors 14L, 14R located on the vehicle's side periphery, the rotational accelerations 214, 215 are directed in a direction inclined by an angle α with respect to the longitudinal direction x. Here, α represents the angle between the connecting line of the center of gravity SP to the corresponding acceleration sensor 14L and the dashed line in the lateral direction y passing through the center of gravity SP. It is noteworthy that the rotational acceleration components 216, 217 of the rotational accelerations 210, 215 acting in the x direction are directed in different directions on both sides of the vehicle. In the illustrated crash example, the acceleration component 216 on the left side of the vehicle 1 is directed in the crash direction, while the acceleration component 217 on the right side of the vehicle 1 is directed in the opposite crash direction. Furthermore, the rotational acceleration components 216, 217 acting in the x direction are cos(α) times the amount of the vector rotational accelerations 214, 215. However, if the sensor axis is near the center of gravity SP, e.g., a B-pillar, the angle α is small, and the rotational accelerations 210, 215 are almost entirely in the x direction.

[0027] This rotational movement is superimposed on a translational acceleration 210 of the vehicle 1 in the x direction. The translational acceleration 210 faces rearward during a frontal collision and has the same magnitude at all locations on the vehicle (except the crushable zone). Thus, in the illustrated example, on the left side of the vehicle, the translational acceleration 210 and the rotational acceleration 216 acting in the x direction face rearward and are superimposed and reinforce each other. On the right side of the vehicle, the rotational acceleration 217 acting in the x direction faces forward in the x direction, opposite to the translational acceleration 210. Therefore, the rotational accelerations 216, 217 acting in the x direction can be obtained by calculating the difference according to equation (1) from the total acceleration 212 acting in the x direction measured on the left side of the vehicle or the total acceleration 213 acting in the x direction measured on the right side of the vehicle. 216=-217=(212-213) / 2 (1) On the other hand, the translational acceleration component 210 in the x direction is obtained by calculating the sum according to equation (2). 210=(212+213) / 2 (2)

[0028] As can be further seen from FIGS. 2-6 , the illustrated exemplary embodiment of the method 100 according to the present invention for determining the type of collision of a vehicle 1 includes step S100, in which at least one time profile of at least one motion information DR (210, 212, 213) of the vehicle 1 is captured. In step S110, a time profile of processed rotational motion information FeaR is determined from the at least one captured time profile of the at least one acceleration information DR (212, 213). Here, the at least one motion information DR (212, 213) may be subjected to, for example, low-pass filtering, band-pass filtering, integration, double integration, differentiation, etc. for processing. In step S120, at least one evaluation information is generated by evaluating the time profile of the processed rotational motion information FeaR before at least one predetermined termination criterion expires. In step S130, the type of collision is determined based on the at least one evaluation information.

[0029] In step S130, when determining the type of collision, it is recognized whether the collision between vehicle 1 and obstacle H is a full-lap head-on collision, an offset collision with uniform rotation, an offset collision with a strong lateral component, or an offset collision with a strong lateral component and getting caught on obstacle H.

[0030] In the illustrated exemplary embodiment, in step S110, a time profile of processed translational motion information FeaT is determined from at least one captured time profile of at least one of the motion information 210, 212, 213, and this time profile is further evaluated to generate at least one evaluation information. Then, in step S120, a time profile of composite motion information FeaN generated by combining the time profile of the processed rotational motion information FeaR and the time profile of the processed translational motion information FeaT is evaluated to generate at least one evaluation information. The composite motion information FeaN is preferably generated by dividing the processed rotational motion information FeaR by the processed translational motion information FeaT.

[0031] In the illustrated exemplary embodiment, two threshold criteria are used as predetermined termination criteria in step S120. In this case, the first termination criterion corresponds to an acceleration criterion that is met when the captured and / or processed acceleration information reaches or exceeds a predetermined first threshold. Thus, for example, the captured translational acceleration 210 can be summed in the x-direction. The second termination criterion corresponds to a time criterion that is met when a predetermined time window tF is exceeded. Of course, in alternative exemplary embodiments not shown, other suitable termination criteria, other combinations of termination criteria, or just one termination criterion can be used.

[0032] As already mentioned above, the captured third motion information 212 corresponds to valid total acceleration information captured on the left side of the vehicle 1 in the vehicle longitudinal direction x, and the captured fourth motion information 213 corresponds to valid total acceleration information captured on the right side of the vehicle 1 in the vehicle longitudinal direction x. In the illustrated exemplary embodiment, the processed third motion information FeaLx and the processed fourth motion information FeaRx are generated by processing the third motion information 212 and the fourth motion information 213 using low-pass filtering, band-pass filtering, integration, double integration, differentiation, etc. Here, in the illustrated exemplary embodiment, the time profile of the processed rotational acceleration information FeaR is determined by calculating the difference between the processed third acceleration information FeaLx and the processed fourth acceleration information FeaRx according to equation (3). The time profile of the processed translational acceleration information FeaT is determined by calculating the sum of the processed third acceleration information FeaLx and the processed fourth acceleration information FeaRx according to equation (4). FeaR = FeaLx - FeaRx (3) FeaT = FeaLx + FeaRx (4)

[0033] In an alternative exemplary embodiment not shown, the processed rotational acceleration information FeaR is determined from a first quotient of dividing the processed third motion information FeaLx by the processed fourth motion information FeaRx according to equation (5) or from a second quotient of dividing the processed fourth motion information FeaRx by the processed third motion information FeaLx according to equation (6) and can be used to generate at least one evaluation information. In a crash without rotation, i.e., when FeaLx is approximately equal to FeaRx, the processed rotational acceleration information FeaR takes a value of 1. In a crash with rotation, the processed rotational acceleration information FeaR takes a value significantly above or below 1. FeaR=FeaLx / FeaRx (5) FeaR=FeaRx / FeaLx (6)

[0034] As mentioned above, the captured first movement information DR corresponds to a rotational velocity about the vehicle vertical axis z captured in the center of the region of the center of gravity SP of the vehicle 1. The captured second movement information 210 corresponds to captured translational acceleration information in the vehicle longitudinal direction x. Accordingly, a time profile of the processed rotational movement information FeaR can be determined from the first movement information DR, for example for validating the measurement results. Additionally or alternatively, a time profile of the processed translational acceleration information FeaT can be determined from the captured second movement information 210.

[0035] In the illustrated exemplary embodiment, the composite motion information FeaN is determined from the quotient of the processed rotational motion information FeaR divided by the processed translational motion information FeaT according to equation (7) and is used to generate at least one evaluation information. FeaN=FeaR / FeaT (7)

[0036] In a further exemplary embodiment not shown, a time profile of the processed rotational motion information FeaR and a time profile of the processed translational motion information FeaT are used to generate at least one evaluation information.

[0037] When determining the type of collision, in the illustrated exemplary embodiment, the composite motion information FeaN is compared with at least one threshold value S1, S2, S3, S4, S5, S6, S7, S8 to recognize their dynamic behavior and therefore the type of collision. In an alternative exemplary embodiment not shown, the processed rotational motion information FeaR and the processed translational motion information FeaT are each compared with at least one threshold value to recognize their dynamic behavior and therefore the type of collision.

[0038] The method 100 can be implemented, for example, as software or hardware or as a mixture of software and hardware, for example in the evaluation and control unit 12. Here, the method can be stored as a computer program product with program code on a machine-readable carrier and executed by the evaluation and control unit 12.

[0039] In the following, with reference to Figures 3 to 6, based on the signal characteristics described above, different crash scenarios are distinguished in order to optimize the triggering of the corresponding restraining means.

[0040] FIG. 3 shows a full-lap collision between a vehicle 1 and an obstacle H without any angle. Here, only a weak rotational acceleration component occurs, which is represented by almost identical signal characteristics in the processed third motion information FeaLx and the processed fourth motion information FeaRx, and which almost exclusively indicates the processed translational motion information FeaT. As a result, the processed rotational motion information FeaR and the combined motion information FeaN are very small.

[0041] Such a situation can be recognized, for example, when the composite motion information FeaN or the processed rotational motion information FeaR does not exceed a predetermined positive first threshold S1 or fall below a predetermined negative second threshold S2, where the first threshold S1 and the second threshold S2 may have different amounts. In the illustrated exemplary embodiment, the composite motion information FeaN is compared with two thresholds S1 and S2. Similarly, similar thresholds are considered for the processed rotational motion information FeaR.

[0042] Figure 4 illustrates a left-side offset collision in which the vehicle rotates consistently in a mathematically positive direction counterclockwise around the vehicle's vertical axis z around the impact point. Such behavior can occur, for example, in pole collisions involving an offset deformable barrier (ODB) or offset. As can be seen from Figure 4, in the illustrated left-side offset collision, the processed third motion information FeaLx is greater than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN have values ​​that are clearly different from zero. The processed rotational motion information FeaR typically increases during the collision process, while the normalized resultant motion information FeaN tends to remain constant because the denominator, the processed translational motion information FeaT, also increases.

[0043] Such a situation can be recognized, for example, when the resultant motion information FeaN or the processed rotational motion information FeaR exceeds a predetermined positive first threshold S1. In the illustrated exemplary embodiment, the resultant motion information FeaN is compared to the first threshold S1. A further condition for recognizing this collision type may require that the sign of the resultant motion information FeaN remain unchanged. In some cases, a short-term sign change is permitted as long as it does not fall below any of thresholds S4, S5, or S8, which will be introduced further below. In a right-side collision, the roles are reversed: the processed fourth motion information FeaRx is greater than the processed third motion information FeaLx. Therefore, the processed rotational motion information FeaR and the resultant motion information FeaN have negative signs. In the illustrated exemplary embodiment, such a right-side offset collision with a uniform rotation is properly recognized when the resultant motion information FeaN falls below a negative second threshold S2.

[0044] FIG. 5 illustrates a left-side offset or angular collision with a strong lateral component. Such behavior may occur, for example, during a 30° collision. The left-side collision initially results in a mathematically positive counterclockwise rotation about the vehicle's vertical axis z until a first time point t1. During this rotation, the processed third motion information FeaLx is greater than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN are greater than zero. However, if the front of the vehicle is further pushed to the right by an obstacle H on the left side, a mathematically negative clockwise rotation about the vehicle's vertical axis z occurs after the first time point t1. During this rotation, the processed third motion information FeaLx is less than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN are less than zero. The illustration in Figure 5 shows vehicle 1 immediately after the first time point t1, where the processed third motion information FeaLx in vehicle 1 is equal to the processed fourth motion information FeaRx, or the processed rotational motion information FeaR and composite motion information FeaN in vehicle 1 are each equal to zero or have changed signs.

[0045] For example, such a situation can be recognized when the resultant motion information FeaN or the processed rotational motion information FeaR first exceeds a predetermined positive third threshold S3 and then falls below a negative fourth threshold S4, where the third threshold S3 and the fourth threshold S4 may have different magnitudes. In the illustrated exemplary embodiment, the resultant motion information FeaN is compared with two thresholds S3 and S4. As a special case, the value "zero" can also be selected for the fourth threshold S4. In this case, this collision type is recognized based on the sign change of the resultant motion information FeaN after previously exceeding the third positive threshold S3. In the illustrated exemplary embodiment, such a right-side offset collision with a uniform rotation is properly recognized when the resultant motion information FeaN first falls below a predetermined negative fifth threshold S5 and then exceeds a positive sixth threshold S6, where the fifth threshold S5 and the sixth threshold S6 may have different magnitudes.

[0046] Finally, FIG. 6 shows a left-side offset or angular crash with a strong lateral component, which causes a rotation around obstacle H during the further crash process, for example, due to "hooking" of obstacle H. Such a movement can occur, for example, in a 30° angle crash or a small overlap test. The left-side impact initially causes a mathematically positive counterclockwise rotation about the vehicle's vertical axis z until a first time point t1. During this rotation, the processed third motion information FeaLx is greater than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN are greater than zero. However, if the front of the vehicle is further pushed to the right by the obstacle H on the left, then after the first time point t1, a mathematically negative rotation in the clockwise direction about the vehicle vertical axis z occurs until the second time point t2, at which time the processed third motion information FeaLx is smaller than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN are smaller than zero or have changed signs again.Then, due to the obstacle H getting caught, a mathematically positive rotation in the counterclockwise direction about the vehicle vertical axis z occurs again, at which time after the second time point t2, the processed third motion information FeaLx is larger than the processed fourth motion information FeaRx, and the processed rotational motion information FeaR and the resultant motion information FeaN are larger than zero or have changed signs again. The solid lines in Fig. 6 show the vehicle 1 immediately after the first time point t1, when the processed third motion information FeaLx for the vehicle 1 is equal to the processed fourth motion information FeaRx, or when the processed rotational motion information FeaR and the composite motion information FeaN for the vehicle 1 are each equal to zero or change signs. The dashed lines in Fig. 6 show the vehicle 1 immediately after the first time point t1, when the processed third motion information FeaLx for the vehicle 1 is again equal to the processed fourth motion information FeaRx, or when the processed rotational motion information FeaR and the composite motion information FeaN for the vehicle 1 are each equal to zero or change signs again.

[0047] Such a situation can be recognized, for example, when the resultant motion information FeaN or the processed rotational motion information FeaR first exceeds a predetermined positive third threshold S3, then falls below a negative fourth threshold S4, and then exceeds a positive seventh threshold S7. Here, the third threshold S3, the fourth threshold S4, and the seventh threshold S7 can have different amounts. In the illustrated exemplary embodiment, the resultant motion information FeaN is compared with the thresholds S3, S4, and S7. Such a right-side offset or angular collision with a strong lateral component, in which a further crash process causes a rotation around the obstacle H, for example, due to a "hook" on the obstacle H, is appropriately recognized in the illustrated exemplary embodiment when the resultant motion information FeaN first falls below a predetermined negative fifth threshold S5, then exceeds a positive sixth threshold S6, and then falls below a negative eighth threshold S8, which can be equal in amount to the positive seventh threshold S7. The fifth threshold S5, the sixth threshold S6, and the eighth threshold S8 may have different amounts.

[0048] All the above thresholds S1, S2, S3, S4, S5, S6, S7, S8 may change constantly or during the collision process, for example as a function of time or speed reduction.

[0049] A combination of feature evaluations based on the processed rotational motion information FeaR and / or the composite motion information FeaN can also be performed. The scenario from Fig. 5 can also be recognized, for example, by a logical AND operation between the processed rotational motion information FeaR and the composite motion information FeaN. The scenario from Fig. 6 can also be recognized, for example, by a logical OR operation between the processed rotational motion information FeaR and the composite motion information FeaN. Furthermore, threshold tests can also be linked to time conditions, so that, for example, corresponding threshold conditions need to follow one another at a specific time interval.

[0050] The first quotient obtained by dividing the processed third motion information FeaLx by the processed fourth motion information FeaRx, or the second quotient obtained by dividing the processed fourth motion information FeaRx by the processed third motion information FeaLx, can also be evaluated. This is not explicitly stated in the above example. However, it is clear that in the non-offset case where the processed third motion information FeaLx is approximately equal to the processed fourth motion information, these quotients have values ​​close to 1. During rotation, these quotients have values ​​above or below 1 depending on the direction. Again, it is easy to set a corresponding set of thresholds.

[0051] The above-mentioned scenarios differ in terms of occupant dynamics, and therefore different restraint strategies may be advantageous for different scenarios. Thus, for example, trigger thresholds for restraint system components can be adapted to the recognized crash type to optimize the triggering time of restraint measures such as belt tensioners and airbags. Furthermore, the control of adaptive restraint measures such as belt force limiters and adaptive airbag valves can be adapted to the recognized crash type. Furthermore, during offset, side head airbags can be deployed on the impact side and / or the opposite side of the vehicle depending on the recognized crash type.

Claims

1. A method (100) for determining the type of collision of a vehicle (1), comprising: capturing at least one time profile of at least one movement information (DR, 210, 212, 213) of said vehicle (1); - determining a time profile of processed rotational movement information (FeaR) from said at least one captured time profile of said at least one movement information (DR, 212, 213); generating at least one evaluation information by evaluating a time profile of said processed rotational motion information (FeaR) before the expiration of at least one predetermined termination criterion; determining a type of the collision based on the at least one evaluation information; Including, A method (100) in which a time profile of processed translational motion information (FeaT) is determined from the at least one captured time profile of the at least one motion information (210, 212, 213) and further evaluated to generate the at least one evaluation information.

2. 2. The method (100) of claim 1, wherein to generate the at least one evaluation information, a time profile of combined motion information (FeaN) is generated by combining the time profile of the processed rotational motion information (FeaR) and the time profile of the processed translational motion information (FeaT).

3. 3. The method (100) of claim 2, wherein the resultant motion information (FeaN) corresponds to the quotient of the processed rotational motion information (FeaT) divided by the processed translational motion information (FeaR).

4. 4. The method according to claim 1, wherein when determining the type of collision, it is recognized whether the collision between the vehicle (1) and the obstacle (H) is a full-lap frontal collision, an offset collision with a uniform rotation, an offset collision with a strong lateral component, or an offset collision with a strong lateral component and a catch on the obstacle (H).

5. 5. The method (100) of any one of claims 1 to 4, wherein said at least one predetermined termination criterion corresponds to a threshold criterion.

6. 6. The method (100) of claim 5, wherein the first termination criterion corresponds to an acceleration criterion that is met when the captured and / or processed acceleration information reaches or exceeds a predetermined first threshold.

7. 7. The method (100) according to claim 5 or 6, wherein the second termination criterion corresponds to a time criterion that is met when a predetermined time window is exceeded.

8. the first captured motion information (DR) corresponds to a rotational velocity of the vehicle (1) about a vertical axis (z) captured in the center of a region of the center of gravity (SP) of the vehicle (1); The second captured motion information (210) corresponds to captured translational acceleration information in the vehicle longitudinal direction (x). A method (100) according to any of claims 1 to 7.

9. the time profile of the processed rotational motion information (FeaR) is determined from first motion information (DR); The time profile of the processed translational motion information (FeaT) is determined from the captured second motion information (210) or from the processed second motion information (210) generated from the second motion information (210).

9. The method (100) of claim 8.

10. the third captured motion information (212) corresponds to total available acceleration information captured at the left periphery of the vehicle (1) in the longitudinal direction (x) of the vehicle; The fourth captured motion information (213) corresponds to the total available acceleration information captured at the right periphery of the vehicle (1) in the longitudinal direction (x) of the vehicle.

8. The method (100) according to any one of claims 1 to 7.

11. The time profile of the processed rotational motion information (FeaR) is determined by combining the third motion information (212) and the fourth motion information (213), or by combining the processed third motion information (FeaLx) generated from the third motion information (212) and the processed fourth motion information (FeaRx) generated from the fourth motion information (213), The time profile of the processed translational motion information (FeaT) is determined by sum calculation from the third motion information (212) and the fourth motion information (213) or from the processed third motion information (FeaLx) and the processed fourth motion information (FeaRx).

11. The method (100) of claim 10.

12. The method (100) according to claim 11, characterized in that the synthesis corresponds to calculating the difference or the ratio between the third motion information (212) and the fourth motion information (213), or between the processed third motion information (FeaLx) and the processed fourth motion information (FeaRx).

13. 13. The method (100) according to any one of claims 1 to 12, characterized in that when determining the type of the collision, the processed rotational motion information (FeaR) or the combined motion information (FeaN) are compared with at least one threshold value (S1, S2, S3, S4, S5, S6, S7, S8) in order to recognize their dynamic behavior and therefore the type of the collision.

14. When the processed rotational motion information (FeaR) or the composite motion information (FeaN) does not exceed a predetermined positive first threshold (S1) and does not fall below a predetermined negative second threshold (S2), a full-lap front collision is recognized as the type of collision; or A left-side offset collision with a uniform rotation is recognized when the processed rotational motion information (FeaR) or the resultant motion information (FeaN) does not exceed the predetermined positive first threshold (S1); or A right-side offset collision with uniform rotation is recognized when the processed rotational motion information (FeaR) or the resultant motion information (FeaN) is not below the predetermined negative second threshold (S2); or A left-side offset collision with a strong lateral component is recognized when the processed rotational motion information (FeaR) or the resultant motion information (FeaN) first exceeds a positive third threshold (S3) and then falls below a predetermined negative fourth threshold (S4); or A right-side offset collision with a strong lateral component is recognized when the processed rotational motion information (FeaR) or the resultant motion information (FeaN) first falls below a predetermined negative fifth threshold (S5) and then exceeds a predetermined positive sixth threshold (S6); or A left offset collision with a strong lateral component and a snag with an obstacle (H) is recognized when the processed rotational motion information (FeaR) or the resultant motion information (FeaN) first exceeds the third positive threshold (S3), then falls below the fourth negative threshold (S4), and then exceeds a predetermined seventh positive threshold (S7); or When the processed rotational motion information (FeaR) or the resultant motion information (FeaN) first falls below the negative fifth threshold (S5), then exceeds the positive sixth threshold (S6), and then falls below a predetermined negative eighth threshold (S8), a right-side offset collision with a strong lateral component and a snag with an obstacle (H) is recognized.

14. The method of claim 13.

15. A sensor arrangement (10) for a vehicle (1), comprising: At least one sensor (14A, 14B, 14L, 14R) designed to capture a time profile of at least one piece of motion information (DR, 210, 212, 213) in the vehicle (1); an evaluation and control unit (12) designed to receive the captured time profile of the at least one movement information (DR, 210, 212, 213), 15. A sensor arrangement (10) characterized in that the evaluation and control unit (12) is configured to carry out the steps of the method (100) for determining the type of collision of a vehicle (1) according to any one of claims 1 to 14.

16. The sensor configuration (10) according to claim 15, characterized in that the first sensor (14A) is arranged and configured in the center of the area of ​​the center of gravity (SP) of the vehicle (1) and captures the rotation speed around the vehicle vertical axis (z) as the first motion information (DR).

17. 17. The sensor arrangement (10) according to claim 16, characterized in that a second sensor (14B) arranged centrally in the region of the center of gravity (SP) of the vehicle (1) is designed to capture translational acceleration information in the vehicle longitudinal direction (x) as second movement information (210).

18. a first acceleration sensor (14L) disposed on the left periphery of the vehicle (1) and designed to capture effective total acceleration information in the vehicle longitudinal direction (x) as third movement information (212); A second acceleration sensor (14R) is arranged on the right peripheral edge of the vehicle (1) and is designed to capture effective total acceleration information in the vehicle longitudinal direction (x) as fourth movement information (213).

16. The sensor arrangement (10) according to claim 15.

19. 15. A computer program comprising a program code stored on a machine-readable carrier for carrying out the method according to any one of claims 1 to 14, when the program is executed in an evaluation and control unit (12).

Images