DE102020205511A1 - Method for determining a type of collision of a vehicle - Google Patents

Abstract

The invention relates to a method for determining a type of collision in a vehicle (1) and to a sensor arrangement (10) with an evaluation and control unit (12) for carrying out the method. The method comprises the following steps: Acquisition of position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y ) in the vehicle (1) at at least two measuring locations (MO1, MO2, MO3, MO4, MO5, MO6, MO7, MO8, MO9) whose projections of their distances to a center of gravity of the vehicle (1) on a central vehicle longitudinal axis (LA) differ Have values, compare at least two of the position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or of at least two processed movement information items (aA, y, aB, y, aC, y) which are based on the at least two position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y ; aMO6, y; aMO7, y; aMO8, y; aMO9, y), generating at least one piece of evaluation information based on the comparison result and determining the type of collision based on the at least one piece of evaluation information.

Description

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

Vehicles are known from the prior art in which frontal collisions of the vehicle with an obstacle are detected by acceleration sensors installed in a central airbag control device. The detection of collisions with partial overlap (so-called "offset crashes") is more difficult than the detection of collisions with full overlap, since only part of the crumple zone is stressed and therefore lower forces and thus lower accelerations occur. The number of crash tests covering the various offset crash scenarios has increased in recent years.

Lateral vehicle collisions can be detected either by pressure sensors in the doors or by acceleration sensors (PAS = Peripheral Acceleration Sensor) attached to the side of the vehicle, which can measure in the lateral or transverse direction of the vehicle. Depending on the required triggering performance, several such “sensor axes” can be provided in the vehicle, for example on pillars of the vehicle body, such as A pillars, B pillars or C pillars. In this case, such a sensor axis can comprise two acceleration sensors attached to the vehicle periphery on the left and right.

From the DE 10 2009 054 473 A1 A method for determining a type of collision of a vehicle is known which is based on the use of two-channel peripheral acceleration sensors, the acceleration sensors being arranged, for example, in the B-pillar of the vehicle and each detecting a first acceleration in the longitudinal direction of the vehicle and a second acceleration in the transverse direction of the vehicle . In this case, signal preprocessing is carried out in which, for example, low-pass filtering of the signals provided by the peripheral acceleration sensors is carried out. A resulting acceleration is then calculated from the two individual channels of the peripheral sensors. The calculation is based on vector addition. The vector addition of the linearly independent sensor signals enables acceleration signals resulting from the linearly independent sensor signals to be calculated. Further signal processing can then be carried out, for example by means of integration or subtraction, and a subsequent algorithm for recognizing the type of collision can be carried out. The algorithm can be a special logic with threshold value comparisons. With the method described, frontal collisions of the vehicle with an obstacle with full coverage can be distinguished from frontal collisions with partial coverage or with an angle.

Disclosure of the invention

The method for determining a type of collision of a vehicle with the features of independent claim 1 and the sensor arrangement with an evaluation and control unit for performing the method with the features of independent claim 12 each have the advantage that a better differentiation between different offset crashes, in particular between those with little lateral movement, such as offset crashes without angles or ODBs (Offset Deformable Barrier), and those with strong lateral movement of the front end, such as offset crashes with a narrow overlap (small overlap). The differentiation takes place on the basis of several lateral accelerations, which can be recorded or measured at measuring locations with different distances from the vehicle's center of gravity, for example in a central airbag control device or on the side of the vehicle. The differentiation allows a control of restraint means that is optimized for the respective crash situation. This includes the deployment time of an airbag of a corresponding restraint system as well as delay times or delay times for “adaptive” restraint devices such as belt force limiters and airbag valves.

Not only so-called ODBs (Offset Deformable Barrier) and 30 ° angle crashes of crashes with full overlap, but also other offset crash scenarios such as an IIHS Small Overlap crash at 64 km / h and 25% overlap, an NHTSA Oblique Crash OMDB (Offset Mobile Deformable Barrier) with a barrier speed of 90 km / h and 35% overlap at a 15 ° angle; a Euro-NCAP MPDB (Moving Progressive Deformable Barrier) with a vehicle speed and a barrier speed of 50 km / h each and a 50% overlap can be recognized and differentiated.

Embodiments of the present invention provide a method for determining a type of collision of a vehicle, which comprises the following steps: Detection of position-dependent, superimposed lateral accelerations in the vehicle at at least two measurement locations whose projections of their distances to a center of gravity of the vehicle on a vehicle longitudinal axis differ Have values. Comparison of at least two of the position-dependent, superimposed lateral accelerations or of at least two processed lateral movement information items which are based on the at least two position-dependent, superimposed lateral accelerations. Generating at least one piece of evaluation information based on the comparison result, and determining the type of collision based on the at least one piece of evaluation information.

The projection of the distance from a measurement location to a center of gravity onto a vehicle longitudinal axis corresponds to a longitudinal distance of the measurement location from a lateral straight line through the center of gravity of the vehicle. This means that the at least two measurement locations have different longitudinal distances from the lateral straight line through the center of gravity.

In addition, a sensor arrangement for a vehicle with at least two sensors, which are each designed to detect a position-dependent, superimposed lateral acceleration at a corresponding measurement location in the vehicle, and an evaluation and control unit is proposed, which is designed to detect the detected position-dependent, superimposed to receive lateral accelerations and to carry out the steps of the method according to the invention for determining a type of collision of the vehicle.

For an algorithmic evaluation of the detected or measured position-dependent, superimposed lateral accelerations in the vehicle, it is advantageous to subject these detected or measured acceleration signals to signal processing, such as low-pass filtering, band-pass filtering, integration, double integration, derivation, etc., and to subject the processed lateral accelerations To use accelerations.

In the present case, the evaluation and control unit can be understood to mean an electrical device, such as a control device, in particular an airbag control device, which processes or evaluates detected sensor signals. The evaluation and control unit can have at least one interface which can be designed in terms of hardware and / or software. In the case of a hardware design, the interfaces can, for example, be part of a so-called system ASIC, which contains a wide variety of functions of the evaluation and control unit. However, it is also possible that the interfaces are separate, integrated circuits or at least partially consist of discrete components. In the case of a software-based design, the interfaces can be software modules that are present, for example, on a microcontroller alongside other software modules. A computer program product with program code which is stored on a machine-readable carrier such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out the evaluation when the program is executed on the evaluation and control unit is also advantageous.

The measures and developments listed in the dependent claims make it possible to improve the method specified in independent claim 1 for determining a type of collision of a vehicle and the sensor arrangement specified in independent claim 12 for a vehicle.

It is particularly advantageous that several lateral accelerations can be recorded along at least two lateral sensor axes which run parallel to a lateral straight line through the center of gravity of the vehicle, the at least two lateral sensor axes having different longitudinal distances from the lateral straight line through the center of gravity of the vehicle . In this case, the individual processed lateral movement information items can each be determined from a combination of at least two lateral accelerations or from at least two processed lateral accelerations of a common lateral sensor axis. The combination can, for example, correspond to an averaging of at least two lateral accelerations or of at least two processed lateral accelerations of a common lateral sensor axis or a selection of one of the lateral accelerations or one of the processed lateral accelerations, which is maximum or minimum in terms of absolute value. For example, the acceleration signals can be combined, which are detected or measured by two acceleration sensors attached to the left and right lateral vehicle peripherals. In addition, an acceleration signal can also be taken into account, which is detected or measured by an acceleration sensor arranged on a central longitudinal axis of the vehicle. Alternatively, the individual processed lateral movement information items can each be determined only from a lateral acceleration or from a processed lateral acceleration, the measurement locations of which are arranged along the central longitudinal axis of the vehicle.

The comparison of at least two of the position-dependent, superimposed lateral accelerations or of at least two processed lateral movement information items can be made, for example, via a difference formation or via one of the two position-dependent lateral accelerations Accelerations or lateral movement information normalized subtraction or a quotient formation of the at least two position-dependent lateral accelerations or lateral movement information can be carried out.

In an advantageous embodiment of the method, when determining the type of collision, the comparison result can be compared in a first further comparison with at least one threshold value in order to identify the type of collision. In addition, when determining the type of collision, at least one of the position-dependent lateral accelerations or lateral movement information can be compared with at least one threshold value in at least one second further comparison. The comparison result of the first further comparison and the comparison result of the at least one second further comparison can be combined with one another in order to identify the type of collision.

In a further advantageous embodiment of the method, when determining the type of collision, time courses of the position-dependent, lateral accelerations or lateral movement information can be evaluated, the first further comparison and / or the second further comparison each being able to be carried out several times at different times. The comparison results of the first further comparisons and the second further comparisons can then be combined with one another in order to identify the type of collision.

In a further advantageous embodiment of the method, when determining the type of collision, it can be recognized whether the collision of the vehicle with an obstacle is a frontal collision with full overlap or an offset collision without angles and predominantly rotational properties or an offset collision with an angle and strong lateral components or an offset collision with an angle and strong lateral components and entanglement with an obstacle. For example, it can be recognized on which side of the vehicle the collision is taking place and whether the vehicle is hitting the obstacle, such as a barrier or another vehicle, at an angle or head-on, and whether there is a complete or partial overlap and whether there is a rotation in the Occurs counterclockwise or clockwise.

Exemplary embodiments of the invention are shown in the drawing and are explained in more detail in the following description. In the drawing, the same reference symbols designate components or elements that perform the same or analogous functions.

Figure list

• 1 FIG. 11 shows a schematic block diagram of a vehicle with a first exemplary embodiment of a sensor arrangement according to the invention with an evaluation and control unit for executing the method from FIG 4th .
• 2 FIG. 11 shows a schematic block diagram of a vehicle with a second exemplary embodiment of a sensor arrangement according to the invention with an evaluation and control unit for executing the method from FIG 4th .
• 3 FIG. 11 shows a schematic block diagram of a vehicle with a third exemplary embodiment of a sensor arrangement according to the invention with an evaluation and control unit for executing the method from FIG 4th .
• 4th shows a schematic flow diagram of an exemplary embodiment of a method according to the invention for determining a type of collision of a vehicle.
• 5 shows a representation of a first type of collision with a corresponding characteristic curve diagram of the processed measurement signals.
• 6th shows a representation of a second type of collision with a corresponding characteristic curve diagram of the processed measurement signals.
• 7th shows a representation of a third type of collision with a corresponding characteristic curve diagram of the processed measurement signals.

Embodiments of the invention

How out 1 until 3 As can be seen, the illustrated embodiments comprise a sensor arrangement according to the invention 10 , 10A , 10B , 10C for a vehicle 1 , 1A , 1B , 1C at least two sensors each 14th , which are each carried out at a corresponding measurement location MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 in the vehicle 1 a position-dependent, superimposed lateral acceleration aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9 to record y, and an evaluation and control unit 12th , 12A , 12B , 12C , which is executed, the detected position-dependent, superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9 to receive y. The respective evaluation and control unit is here 12th , 12A , 12B , 12C formed the steps of the method according to the invention described below 100 to determine a type of collision of the vehicle 1 perform. In 1 until 3 A two-axis coordinate system was introduced in which a vehicle transverse direction or a y-direction runs positively to the right and a vehicle longitudinal direction or x-direction positive upwards.

How out 1 and 2 can also be seen are the measuring locations MO1 , MO2 , MO3 and the corresponding sensors 14th along a peripheral left side of the vehicle 1A , 1B arranged. In addition, the measuring locations are MO4 , MO5 , MO6 and the corresponding sensors 14th along a peripheral right side of the vehicle 1A , 1B arranged. How out 1 and 3 can also be seen are the measuring locations MO17 MO8 , MO9 and the corresponding sensors 14th along a central longitudinal vehicle axis LA through a center of gravity SP of the vehicle 1A , 1C arranged. The evaluation and control unit 12th , 12A , 12B , 12C corresponds in each case to an airbag control unit in the exemplary embodiments shown and is close to the center of gravity SP of the vehicle 1 , 1A , 1B , 1C arranged. The sensors 14th are in the illustrated embodiments of the sensor arrangement 10 , 10A , 10B , 10C each designed as two-channel acceleration sensors, which each have a lateral acceleration aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y in the vehicle transverse direction y and a longitudinal acceleration aMO1, x; aMO2, x; aMO3, x; aMO4, x; aMO5, x; aMO6, x; aMO7, x; aMO8, x; record or measure aMO9, x. Since only the lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y can be used for evaluation in the transverse direction y of the vehicle, the sensors 14th can also be designed as single-channel acceleration sensors aligned in the transverse direction y of the vehicle.

$$\text{arMO}\left(\text{k}\right)\text{,y}=\text{arMO}\left(\text{k}\right)\text{ sin WMO}\left(\text{k}\right)$$

mit k = (1 bis 9).The movement of a rigid body such as a vehicle 1 , 1A , 1B , 1C can translate his focus SP and in a rotation around its center of gravity SP be disassembled. If one restricts oneself to movements in the plane, the translation can be characterized by an acceleration vector in an xy plane, a longitudinal direction x corresponding to a longitudinal direction of the vehicle and a lateral direction y to a transverse direction of the vehicle. Since the airbag control unit or the evaluation and control unit 12th , 12A , 12B , 12C close to the center of gravity SP of the vehicle 1 , 1A , 1B 1C is arranged, an acceleration sensor arranged there detects 14th essentially the translational acceleration at. The rotation around the center of gravity SP is by a rotation vector DV characterized. If the rotation only occurs as a result of the crash or collision, is the angular acceleration WB essential. Outside the pivot point, which is the center of gravity SP is this angular acceleration WB at one of the measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 , their distance to the pivot point or center of gravity SP each by a radius rA , rB , rC , rD , re , RH determines the cause of the rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9, which is the cross product of the angular acceleration WB and the distance to the pivot point or center of gravity SP representing radius rA , rB , rC , rD , re , RH calculated. This is in 2 and 3 for a mathematically positive counterclockwise rotation. This is supposed to be done by positive values for the rotation vector DV or the angular acceleration WB to be discribed. Designated WMO1 , WMO2 , WMO3 , WMO4 , WMO5 , WMO6 in each case the angle between a straight line shown in dashed lines in the lateral direction through the center of gravity SP and the radius rA , rB , rC , rD , re , RH or location vector of the respective measurement location MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , \ MO9 of the vehicle 1 , 1A , 1B , 1C , then applies to the longitudinal components arMO1, x; arMO2, x; arMO3, x; arMO4, x; arMO5, x; arMO6, x; arMO7, x; arMO8, x; arMO9, x of the rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9 the equation (1). For the lateral components arMO1, y; arMO2, y; arMO3, y; arMO4, y; arMO5, y; arMO6, y; arMO7, y; arMO8, y; arMO9, y of the rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9, equation (2) applies. $$\text{arMO}\left(\text{k}\right)\text{, x}=\text{arMO}\left(\text{k}\right)\text{cos WMO}\left(\text{k}\right)$$

$$\text{arMO}\left(\text{k}\right)\text{, y}=\text{arMO}\left(\text{k}\right)\text{sin WMO}\left(\text{k}\right)$$

with k = (1 to 9).

The different measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 are now through different angles WMO1 until WMO9 and different distances rA , rB , rC , rD , re , RH from the center of gravity SP marked, the angle WMO7 , WMO8 , WMO9 each have a value of 90 °. How out 1 and 2 As can be seen further, two are close to the center of gravity SP lying measuring locations MO2 , MO5 for example on the B-pillars of the vehicle 1A , 1B are arranged by smaller angles WMO2 , WMO5 and a smaller radus rB marked as further from the center of gravity SP remote measuring locations MO3 , MO6 for example on the C-pillars of the vehicle 1A , 1B are arranged. As a result of the larger radius rC and the larger angle WMO3 , WMO6 there are larger lateral parts arMO3, y on the C-pillars; arMO6, y of the rotational acceleration arMO3, arMO6 on than on the B-pillars. At the named measuring locations MO2 , MO3 , MO5 , MO6 is the lateral acceleration arMO2, y; arMO4, y; arMO3, y; arMO6, y at the in 2 mathematical positive rotation shown positive (pointing to the right), since the Measuring locations MO2 , MO4 , MO3 , MO6 in the vehicle longitudinal direction x behind the center of gravity SP are arranged. For the sake of completeness, in 1 and 2 two more measuring locations MO1 , MO4 shown, which in the vehicle longitudinal direction x in front of the center of gravity SP and for example on the A-pillars of the vehicle 1A , 1B are arranged. These measuring locations MO1 , MO4 are by angle WMO1 , WMO4 in the range from 180 ° to 360 ° to the lateral straight line through the center of gravity SP and an equal radius rA characterized. The mathematically positive rotation results in a corresponding negative lateral rotation acceleration arMO1, y; arMO4, y.

How out 1 and 3 As can be seen further, there are three measuring locations MO7 , MO8 , MO9 arranged along the central vehicle longitudinal axis LA. The corresponding acceleration sensors 14th can be mounted along a tunnel. In the above equations (1) and (2) is for such central acceleration sensors 14th , their measuring locations MO7 lie in front of the center of gravity in the vehicle longitudinal direction x, a value of 270 ° for the angle WMO7 to use. For such central acceleration sensors 14th , their measuring locations MO8 , MO9 in the vehicle longitudinal direction x behind the center of gravity SP is for the corresponding angle WMO8 , WMO9 insert a value of 90 ° in each case. This means that the corresponding rotational acceleration arMO7, arMO8, arMO9 at these measurement locations MO7 , MO8 , MO9 act completely in the lateral direction y, so that only the different radii rD , re , RH the different lateral accelerations aMO7, y; aMO8, y; aMO9, y at these measurement locations MO7 , MO8 , MO9 determine. The measuring locations MO8 , MO9 are in the vehicle longitudinal direction x behind the center of gravity SP arranged, therefore the corresponding lateral acceleration arMO8, y; arMO9, y at the in 3 mathematical positive rotation shown positive (pointing to the right). Because the measurement location MO7 in the vehicle longitudinal direction x in front of the center of gravity SP the mathematical positive rotation results in a corresponding negative lateral rotation acceleration arMO7, y.

mit k = (1 bis 9).In addition to the rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7 , arMO8 , arMO9 occurs at the measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 now there is also a translational acceleration at that is outside the crumple zone in the entire vehicle 1 , 1A , 1B , 1C assumes relatively similar values. This means that there is also a longitudinal component at, x and a lateral component at, y of the translational acceleration at for the entire vehicle 1 , 1A , 1B , 1C are the same outside the crumple zone. Depending on the type of collision, this translational acceleration at has no lateral components at, y, for example in the case of a crash without offset and without an angle, only small lateral components at, y, for example in an offset crash without an angle, or significant lateral components at, y, for example at an angle crash or a small overlap crash. The lateral acceleration aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y at the respective measuring location MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 is then given by the superposition of the lateral components of translational and rotational acceleration. This means that the recorded or measured superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y each have a lateral component at, y of the position-independent translational acceleration at and a lateral component arMO1, y; arMO2, y; arMO3, y; arMO4, y; arMO5, y; arMO6, y; arMO7, y; arMO8, y of the position-dependent rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9 and are composed according to equation (3). Here, the angular acceleration WB at measuring locations MO1 MO4 , MO7 , which in the vehicle longitudinal direction x in front of the center of gravity SP of the vehicle 1 , 1A , 1B , 1C and thus an angle WMO (k) in the range from 180 ° to 360 ° to the lateral straight line through the center of gravity SP have a negative sign. $$\text{aMO}\left(\text{k}\right)\text{, y}=\text{at}\text{, y}+\text{arMO}\left(\text{k}\right)\text{, y}=\text{at}\text{, y}+\text{WB r}\left(\text{k}\right)\text{sin WMO}\left(\text{k}\right)$$

with k = (1 to 9).

The in 1 illustrated first embodiment of the sensor arrangement 10A are the measuring locations MO1 , MO4 , MO7 on a first common sensor axis A shown in dash-dotted lines, which is parallel in the vehicle longitudinal direction x at a longitudinal first distance in front of the lateral straight line through the center of gravity SP runs. The measuring locations MO2 , MO5 , MO8 lie on a second common sensor axis B, shown in dash-dotted lines, which are parallel in the vehicle longitudinal direction x behind the lateral straight line through the center of gravity with a longitudinal second distance SP runs. The measuring locations MO3 , MO6 , MO9 lie on a third common sensor axis C, shown in dash-dotted lines, which are parallel in the vehicle longitudinal direction x behind the lateral straight line through the center of gravity with a third longitudinal distance SP runs. Therefore, according to the simplified theory set out above, this results for the measurement locations MO1 , MO4 , MO7 , which are on the common first lateral sensor axis A, for the measurement locations MO2 , MO5 , MO8 , which lie on the common second lateral sensor axis B, and for the measurement locations MO3 , MO6 , MO9 , which lie on the common third sensor axis, each have identical values for the lateral acceleration aMO1, y = aMO4, y = aMO7, y; aMO2, y = aMO5, y = aMO8, y; aMO3, y = aMO6, y = aMO9, y, since the greater distance between the outer measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 to focus SP compared to the vertical distance between the central measuring locations MO7 , MO8 , MO9 to focus SP through the sine component (sinWMO (k)) the angle WMO (k) of the lateral components arMO1, y; arMO2, y; arMO3, y; arMO4, y; arMO5, y; arMO6, y; arMO7, y; arMO8, y; arMO9, y of the rotational acceleration arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9 according to equation (2) of the corresponding measuring location MO1 , MO2, MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 when determining the lateral acceleration aMO1, y; aMO4, y; aMO7, y; aMO2, y; aMO5, y; aMO8, y; aMO3, y; aMO6, y; aMO9, y is compensated.

The in 2 illustrated second embodiment of the sensor arrangement 10B are the measuring locations MO1 , MO4 on the first common sensor axis A shown in dash-dotted lines, which is parallel in the vehicle longitudinal direction x at a longitudinal first distance in front of the lateral straight line through the center of gravity SP runs. The measuring locations MO2 , MO5 lie on the second common sensor axis B, shown in dash-dotted lines, which with a longitudinal second distance parallel in the vehicle longitudinal direction x behind the lateral straight line through the center of gravity SP runs. The measuring locations MO3 , MO6 lie on the third common sensor axis C, shown in dash-dotted lines, which are parallel in the vehicle longitudinal direction x behind the lateral straight line through the center of gravity with a longitudinal third distance SP runs. Therefore result for the measuring locations MO1 , MO4 , which are on the common first lateral sensor axis A, for the measurement locations MO2 , MO5 , which lie on the common second lateral sensor axis B, and for the measurement locations MO3 , MO6 , which lie on the common third sensor axis, each have identical values for the lateral acceleration aMO1, y = aMO4, y; aMO2, y = aMO5, y; aMO3, y = aMO6, y.

In practice, however, deviations between the lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y occur, which indicate an inhomogeneous signal propagation in the vehicle 1A , 1B , local resonances or can be attributed to (very low) centrifugal components.

The in 3 illustrated third embodiment of the sensor arrangement 10C are the three measuring locations MO7 , MO8 , MO9 arranged along the central vehicle longitudinal axis LA, the measuring locations MO7 on the first sensor axis A, the measurement location MO8 on the second sensor axis B and the measuring location MO9 lies on the third sensor axis C.

Mixtures of the exemplary embodiments shown are also conceivable. So can in a vehicle 1 for example, only one central sensor on the second sensor axis B. 14th at the measuring location MO8 be present, on the third sensor axis C, on the other hand, two peripheral sensors can be present 14th at the measuring locations MO3 , MO6 to be available.

How out 4th As can be seen, the illustrated embodiment comprises a method according to the invention 100 to determine a type of collision of a vehicle 1 , 1A , 1B , 1C one step S100 , in which position-dependent, superimposed lateral Accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y in the vehicle 1 at at least two measuring locations MO1, MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 are recorded, their projections of their distances rA , rB , rC , rD , re , RH to a focus SP of the vehicle 1 have different values on the central longitudinal vehicle axis LA. In step S110 are in the illustrated embodiment of the method 100 at least two processed lateral movement information items aA, y, aB, y, aC, y compared with one another, which are based on the at least two position-dependent, superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y based. In step S120 at least one piece of evaluation information is generated based on the comparison result, and in step S130 the type of collision is determined based on the at least one piece of evaluation information.

The in 1 illustrated first embodiment of the sensor arrangement 10A prepares the procedure 100 those in the crotch S100 detected lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y each through a filtering or an integration before the comparison in step S110 on. In addition, the process generates 100 in step S110 before comparing a combination of those at the measurement locations MO1 , MO4 , MO7 recorded, processed lateral accelerations aMO1, y; aMO4, y; aMO7, y a processed first lateral movement information aA, y for the first common lateral sensor axis A. From a combination of the at the measurement locations MO2 , MO5 , \ MO8 recorded, processed lateral accelerations aMO2, y; aMO5, y; aMO8, y processed second lateral movement information aB, y for the second common lateral sensor axis B is generated. From a combination of those at the measuring locations MO3 , MO6 , MO9 recorded, processed lateral accelerations aMO3, y; aMO6, y; aMO9, y becomes processed third lateral movement information aC, y generated for the third common lateral sensor axis C. In the exemplary embodiment shown, the combination corresponds to averaging the processed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y of a common lateral sensor axis A, B, C. Therefore, the processed first lateral movement information aA, y of the first sensor axis A corresponds to the mean value of the processed lateral accelerations aMO1, y; aMO4, y; aMO7, y. The processed second lateral movement information aB, y of the second sensor axis B corresponds to the mean value from the processed lateral accelerations aMO2, y; aMO5, y; aMO8, y. The processed third lateral movement information aC, y of the third sensor axis C corresponds to the mean value from the processed lateral accelerations aMO3, y; aMO6, y; aMO9, y.

The in 2 illustrated second embodiment of the sensor arrangement 10B prepares the procedure 100 those in the crotch S100 detected lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y in each case by a filtering or an integration before the comparison in step S110 on. In addition, the process generates 100 in step S110 before the comparison from a combination of the at the measuring locations MO1, MO4 recorded, processed lateral accelerations aMO1, y; aMO4, y the first processed movement information aA, y for the first common lateral sensor axis A. From a combination of the at the measurement locations MO2 , MO5 recorded, processed lateral accelerations aMO2, y; aMO5, y, the second processed movement information aB, y for the second common lateral sensor axis B is generated. From a combination of those at the measuring locations MO3 , MO6 detected, processed lateral acceleration aMO3, y; aMO6, y, a third prepared movement information item aC, y is generated for the third common lateral sensor axis C. In the exemplary embodiment shown, the combination corresponds to averaging the processed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y of a common lateral sensor axis A, B, C. Therefore, the processed first lateral movement information aA, y of the first sensor axis A corresponds to the mean value of the processed lateral accelerations aMO1, y; aMO4, y. The processed second lateral movement information aB, y of the second sensor axis B corresponds to the mean value from the processed lateral accelerations aMO2, y; aMO5, y. The processed third lateral movement information aC, y of the third sensor axis C corresponds to the mean value from the processed lateral accelerations aMO3, y; aMO6, y.

In alternative exemplary embodiments, not shown, the combination of at least two lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y for determining the processed lateral movement information aA, y, aB, y, aC, y of a selection of one of the lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y of a common sensor axis A, B, C or one of the processed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y correspond to a common sensor axis A, B, C. For example, a maximum or minimum lateral acceleration aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y or an absolute maximum or minimum processed lateral acceleration aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y of a common sensor axis A, B, C can be selected as corresponding processed lateral movement information aA, y, aB, y, aC, y.

The in 3 illustrated third embodiment of the sensor arrangement 10C prepares the procedure 100 those in the crotch S100 detected lateral accelerations aMO7, y; aMO8, y; aMO9 each by filtering or integrating before the comparison in step S110 on. In addition, the procedure uses 100 in step S110 those at the measurement site MO7 recorded processed lateral acceleration aMO7 as first processed movement information aA, y for the first common lateral sensor axis A. The one at the measurement locations MO8 detected, processed lateral accelerations aMO8, y is used as second processed movement information aB, y for the second common lateral sensor axis B. The ones at the measuring sites MO9 detected, processed lateral acceleration aMO9, y is used as third processed movement information aC, y for the third common lateral sensor axis C.

When comparing in step S110 of the procedure 100 In the exemplary embodiment shown, a difference is formed from at least two of the lateral movement information items aA, y, aB, y, aC, y. In addition, the difference formed can be normalized to one of the two lateral movement information items aA, y, aB, y, aC, y. As a further alternative, a quotient can be formed from at least two of the lateral movement information items aA, y, aB, y, aC, y. To generate the evaluation information, the comparison result can be used in step S120 be compared in a first further comparison with at least one threshold value in order to identify the type of collision. In addition, in step S120 the lateral movement information aA, y, aB, y, aC, y, which is for the comparison in step S110 are used, are compared with at least one threshold value in at least a second further comparison. The comparison result of the first further comparison and the comparison result of the at least one second further comparison are then combined with one another in order to identify the type of collision. In the exemplary embodiment shown, a Difference (aC, y − aB, y) between the third lateral movement information aC, y and the second lateral movement information aB, y, which is normalized according to the third lateral movement information aC, y, with a corresponding threshold value SD compared. In addition, the second lateral movement information is aB, y with a further threshold value SB2 compared. Depending on the comparison results, the different types of collisions can then be distinguished, as described below.

In the exemplary embodiment shown, when determining the type of collision, temporal profiles of the lateral movement information aA, y, aB, y, aC, y are evaluated. For this purpose, the first further comparison and the second further comparison are each carried out several times at different points in time and the comparison results of the first further comparisons and the second further comparisons are combined with one another in order to identify the type of collision.

In an alternative embodiment of the method (not shown) 100 be in step S110 at least two of the position-dependent, superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y compared to each other. A difference between at least two position-dependent lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y are formed, which lie on different sensor axes A, B, C. In addition, the difference formed can be applied to one of the two position-dependent lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y can be normalized. As a further alternative, at least two of the lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y, which lie on different sensor axes A, B, C, a quotient is formed. To generate the evaluation information, the comparison result can be used in step S120 be compared in a first further comparison with at least one threshold value in order to identify the type of collision. In addition, in step S120 the position-dependent lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y, which for the comparison in step S110 are used, are compared with at least one threshold value in at least a second further comparison. The comparison result of the first further comparison and the comparison result of the at least one second further comparison are then combined with one another in order to identify the type of collision. In addition, when determining the type of collision, time courses of the position-dependent lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y are evaluated. For this purpose, the first further comparison and the second further comparison are each carried out several times at different points in time and the comparison results of the first further comparisons and the second further comparisons are combined with one another in order to identify the type of collision.

When determining the type of collision in step S130 it is recognized whether there is a collision of the vehicle 1 with an obstacle H is a frontal collision with full overlap or an offset collision without angles and predominantly rotational properties or an offset collision with an angle and strong lateral components or an offset collision with an angle and strong lateral components and a hooking with an obstacle H. .

The procedure 100 can for example in software or hardware or in a mixed form of software and hardware, for example in the evaluation and control unit 12th , 12A , 12B , 12C be implemented. Here the procedure 100 be stored as a computer program product with program code on a machine-readable carrier and from the evaluation and control unit 12th , 12A , 12B , 12C are executed.

Referring to FIG 5 until 7th various collision scenarios are described, each of which is characterized by a characteristic interplay of translational and rotational lateral signals.

In the case of a first type of collision, not shown, in which the vehicle 1 collides with an obstacle H with full overlap and without an angle, neither significant translational accelerations at, y in the lateral direction nor significant rotational accelerations arMO1, arMO2, arMO3, arMO4, arMO5, arMO6, arMO7, arMO8, arMO9 occur at the measurement locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 on. Hence the superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y at the measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 as well as the processed lateral movement information aA, y, aB, y, aC, y of the sensor axes A, B, C, very small apart from vibrations, so that equation (4) applies. $$\text{aC}\text{, y}\approx \text{away}\text{, y}\approx \text{aA}\text{, y}\approx 0$$

Such a situation can be recognized, for example, if none of the superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y at the measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 and / or none of the processed lateral movement information aA, y, aB, y, aC, y of the sensor axes A, B, C exceed predetermined threshold values. This means that the above-mentioned difference (aC, y-aB, y) normalized according to the third lateral movement information aC, y between the third lateral movement information aC, y and the second lateral movement information aB, y is smaller than the corresponding threshold value SD1, and the second lateral movement information aB, y is smaller than the further threshold value SB2_1 is.

5 Fig. 13 shows a second type of collision, which corresponds to a left-hand offset collision in which the vehicle 1 rotates consistently around the point of impact in a mathematically positive direction counterclockwise around the vertical axis of the vehicle. Such behavior can occur, for example, in the case of an ODB (offset deformable barrier) or a pole impact with offset. Only slight translational accelerations at, y occur here in the lateral direction y, since there are no forces on the vehicle in the lateral direction y 1 act. On the other hand, a rotation develops in such collisions, which leads to clearly superimposed lateral accelerations aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y at the measuring locations MO1 , MO2 , MO3 , MO4 , MO5 , MO6 , MO7 , MO8 , MO9 to lead.

Using the example of the illustrated left-side crash with rotation in the mathematically negative sense, based on equations (3) and (4) by neglecting the lateral components at, y of the translational acceleration at, positive lateral movement information (pointing to the right) aB, y; aC, y along the second and third sensor axes B, C in the direction of travel behind the center of gravity SP and negative lateral movement information aA, y (pointing to the left) in the vehicle longitudinal direction x in front of the center of gravity SP , where the amount of lateral movement information aA, y; aB, y; aC, y with increasing distance from the center of gravity SP increases. In summary, inequality (5) applies. $$\text{aC}\text{, y}>\text{away}\text{, y}>0>\text{aA}\text{, y}$$

How out 5 It can also be seen that the acting translational accelerations at, at, y and rotational lateral accelerations arA, y; arB, y; arC, y on the vehicle 1 clarified, wherein the rotary signals along the sensor axes A, B, C are drawn starting from the vehicle longitudinal axis LA but also act on the left and right side of the vehicle according to the above explanations. A possible time course of the resulting processed lateral movement information aA, y, aB, y, aC, y of the sensor axes A, B, C is shown in the corresponding characteristic diagram as processed measurement signals AMS shown. This means that the above-mentioned difference (aC, y-aB, y) normalized according to the third lateral movement information aC, y between the third lateral movement information aC, y and the second lateral movement information aB, y is greater than the corresponding threshold value SD2 is, and the second lateral movement information aB, y is also greater than the further threshold value SB2_2 is.

6th shows a third type of collision, which corresponds to a left-sided offset or angle collision with strong lateral components. Such behavior can occur, for example, in the event of a 30 ° angle collision. In such a left-sided collision, pushing the front end of the vehicle away from the obstacle, which corresponds to a positive lateral translational acceleration at, y to the right, after a brief mathematical positive counterclockwise rotation around the vehicle's vertical axis leads to a mathematical negative clockwise rotation around the vehicle's vertical axis. This can be expressed in equations (3) and (4) by using negative values for the rotation vector DV and the angular acceleration WB be written.

In 6th a point in time t1 of the collision is shown, which is in the phase of the mathematically negative clockwise rotation about the vertical axis of the vehicle. How out 6th As can also be seen, the lateral translational acceleration at, y and the lateral rotational acceleration arA, y are superimposed along the first sensor axis A in front of the center of gravity SP constructive. Behind the center of gravity SP the lateral rotational acceleration arB, y run along the second sensor axis B and the lateral rotational acceleration arC, y run along the third sensor axis C in the opposite direction to the lateral translational acceleration at, y. This results in lower values for the resulting processed lateral movement information aB, y, aC, y, which with increasing distance from the center of gravity SP continue to decrease. In summary, inequality (6) applies at time t1. $$\text{aA}\text{, y}>\text{away}\text{, y}>\text{aC}\text{, y}$$

How out 6th It can also be seen that the acting translational accelerations at, at, y and rotational lateral accelerations arA, y; arB, y; arC, y on the vehicle 1 clarified, wherein the rotary signals along the sensor axes A, B, C are drawn starting from the vehicle longitudinal axis LA but also act on the left and right side of the vehicle according to the above explanations. A possible time course of the resulting processed lateral Movement information aA, y, aB, y, aC, y of the sensor axes A, B, C are in the corresponding characteristic diagram as processed measurement signals AMS shown. This means that the above-mentioned difference (aC, y-aB, y) normalized according to the third lateral movement information aC, y between the third lateral movement information aC, y and the second lateral movement information aB, y is smaller than the corresponding threshold value SD3, where the value of SD3 is negative, and the second lateral movement information aB, y, however, greater than the further threshold value SB2_3 is.

7th shows a fourth type of collision, which corresponds to a left-sided offset or angular collision with strong lateral components, in which a rotation around the obstacle H occurs in the further course of the crash, for example due to "hooking" on the obstacle H. Such behavior can occur, for example, in a 30 ° angle crash or in a small overlap test. As a result of the impact on the left, a mathematical positive rotation in the counterclockwise direction about the vertical axis z of the vehicle initially occurs. If the front vehicle is pushed away to the right by the obstacle H on the left side, a mathematical negative rotation in clockwise direction about the vertical axis z of the vehicle then occurs. By hooking on the obstacle H, a mathematical positive rotation in the counterclockwise direction about the vertical axis z of the vehicle then occurs again.

In 7th a point in time t2 of the collision is shown, which is in the phase of the mathematically positive counterclockwise rotation after the negative clockwise rotation about the vertical axis of the vehicle. How out 7th As can also be seen, the lateral rotational acceleration arA, y runs along the first sensor axis A in front of the center of gravity SP in the opposite direction to the lateral translational acceleration at, y, so that the resulting lateral movement information aA, y B is smaller than the lateral translational acceleration at, y. Behind the center of gravity SP the lateral rotational acceleration arB, y along the second sensor axis B and the lateral rotational acceleration arC, y along the third sensor axis C constructively overlap with the lateral translational acceleration at, y. This results in higher values for the resulting processed lateral movement information aB, y, aC, y, which increases with the distance from the center of gravity SP continue to increase. In summary, inequality (7) applies at time t2. $$\text{aC}\text{, y}>\text{away}\text{, y}>\text{aA}\text{, y}$$

How out 7th It can also be seen that the acting translational accelerations at, at, y and rotational lateral accelerations arA, y; arB, y; arC, y on the vehicle 1, the rotary signals along the sensor axes A, B, C starting from the vehicle longitudinal axis LA but also acting on the left and right sides of the vehicle according to the above explanations. A possible time course of the resulting processed lateral movement information aA, y, aB, y, aC, y of the sensor axes A, B, C is shown in the corresponding characteristic diagram as processed measurement signals AMS shown. This means that the above-mentioned difference (aC, y-aB, y) normalized according to the third lateral movement information aC, y between the third lateral movement information aC, y and the second lateral movement information aB, y is greater than the corresponding threshold value SD4 is, and the second lateral movement information aB, y is also greater than the further threshold value SB2_4 is.

In the event of a right-hand collision, the signs of all lateral accelerations and the directions of rotation are reversed. The corresponding ratios must therefore be corrected or interpreted absolutely.

The in 6th The third collision type shown, in which the second prepared lateral movement information aB, y along the second sensor axis B is higher than the third prepared lateral movement information aC, y along the third sensor axis C, is recognized by the fact that the difference in the third lateral movement information aC, y for the second lateral movement information aB, y falls below a predetermined negative threshold value SD3. If this difference is normalized to one of the two lateral movement information items aB, y, aC, y before the threshold value comparison, an evaluation with the correct sign for left and right collisions results automatically. The evaluation using a quotient of the two lateral movement information items aB, y, aC, y also includes left and right collisions equally. Since the normalized difference or the quotient can already assume large values for very small signals due to the “relative” type of query, a link with a minimum signal level is preferably carried out for one of the two lateral movement information items aB, y, aC, y. This means that the second lateral movement information aB, y has a corresponding threshold value SB2 or the third lateral movement information aC, y with a corresponding threshold value SB3 is compared. This will make the in 6th The collision type shown is only recognized if one of the two lateral movement information items aB, y, aC, y also exceeds the corresponding threshold value SB2 , SB3 exceeds.

The in 5 second type of collision shown and the one in 7th illustrated fourth Collision types in which the third processed lateral movement information aC, y along the third sensor axis C is higher than the second processed lateral movement information aB, y along the second sensor axis B, are recognized by the fact that the respective difference between the third lateral movement information aC, y and the second lateral movement information aB, y each have a predetermined threshold value SD2 , SD4 exceeds, whereby the threshold value is used to distinguish it SD4 for the fourth type of collision less than the threshold value SD2 is specified for the second type of collision. If the differences are normalized to one of the two lateral movement information items aB, y, aC, y before the threshold value comparison, an evaluation with the correct sign for left and right collisions results automatically. The evaluation using a quotient of the two lateral movement information items aB, y, aC, y also includes left and right collisions equally. Since the normalized difference or the quotient can already assume large values for very small signals due to the “relative” type of query, a link with a minimum signal level for one of the two lateral movement information items aB, y, aC, y is also carried out here.

The second type of collision and the fourth type of collision are qualitatively similar. As a result, a distinction is very difficult and can be made possible, for example, by different threshold levels. Alternatively, the time course of the lateral movement information aB, y, aC, y can be evaluated. As shown in the characteristic diagram 5 As can be seen, the temporal course of the lateral movement information aB, y, aC, y in the second collision type 2 is uniform, while the temporal course of the lateral movement information aB, y, aC, y in the fourth collision type has different phases with different characteristics, as shown in FIG Characteristic diagram 7th can be seen. With the second collision type, the difference (aC, y - aB, y) of the third lateral movement information aC, y and the second lateral movement information aB, y is always positive or the quotient (aC, y / aB, y) from the third lateral Movement information aC, y and the second lateral movement information aB, y are always greater than 1. This can be recognized by corresponding threshold value queries SD2 are met several times, or corresponding further threshold values SD2 * not be undercut. In the fourth type of collision, the difference (aC, y - aB, y) of the third lateral movement information aC, y and the second lateral movement information aB, y meanwhile become negative or the quotient (aC, y / aB, y) from the third lateral movement information aC, y and the second lateral movement information aB, y falls under 1. This can be recognized by the fact that negative threshold values SD4 * are exceeded in the meantime, before then in the in 7th phase shown around the time t2 the positive threshold value SD4 is exceeded.

By the front of the focus SP arranged first sensor axis A, the second collision type and the fourth collision type can be differentiated by further comparisons and links. For example, the respective difference (aB, y − aA, y) of the second lateral movement information aB, y and the first lateral movement information aA, y can each have predetermined threshold values SD2_2 , SD4_2 be compared. The threshold value is used here to differentiate SD4_2 for the fourth type of collision less than the threshold value SD2_2 specified for the second type of collision. In addition, a link is carried out with a minimum signal level for the two lateral movement information items aA, y, aB, y. This means that the second lateral movement information aB, y has a corresponding threshold value SB2 and the first lateral movement information aA, y with a corresponding threshold value SB1 is compared.

Of course, other suitable comparisons and links can also be carried out to differentiate between the various types of collision.

Since the scenarios described differ with regard to their occupant kinematics, a different restraint strategy can be advantageous for the different scenarios. For example, triggering thresholds for components of the restraint system can be adapted to the type of collision recognized in order to optimize the point in time at which restraint means, such as seat belt tensioners, airbags, etc. are triggered. In addition, the control of adaptive restraint devices, such as belt force limiters and adaptive airbag valves, can be adapted to the type of collision recognized. In addition, in the event of an offset, side head airbags can be deployed either on the side of the collision and / or on the side of the vehicle facing away from the collision, depending on the type of collision recognized.

QUOTES INCLUDED IN THE DESCRIPTION

This list of the documents listed by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent literature cited

• DE 102009054473 A1 [0004]

Claims (15)

Method (100) for determining a type of collision of a vehicle (1), comprising the following steps: Acquisition of position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) in the vehicle (1) at at least two measuring locations (MO1, MO2, MO3, MO4, MO5, MO6, MO7, MO8, MO9), their projections of their distances (rA, rB, rC, rD, rE, rF) have different values for a center of gravity (SP) of the vehicle (1) on a central vehicle longitudinal axis (LA), Compare at least two of the position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or of at least two processed lateral movement information items (aA, y, aB, y, aC, y) which are based on the at least two position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) based, Generating at least one piece of evaluation information based on the comparison result, and Determining the type of collision based on the at least one piece of evaluation information. Method (100) according to Claim 1 , characterized in that several lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) along at least two lateral Sensor axes (A, B, C) are detected, which run parallel to a lateral straight line through the center of gravity (SP) of the vehicle (1), the at least two lateral sensor axes (A, B, C) having different longitudinal distances from the lateral straight line have through the center of gravity (SP) of the vehicle (1). Method (100) according to Claim 2 , characterized in that the individual processed lateral movement information (aA, y, aB, y, aC, y) each consist of a combination of at least two lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5 , y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or of at least two processed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y ; aMO7, y; aMO8, y; aMO9, y) of a common lateral sensor axis (A, B, C). Method (100) according to Claim 3 , characterized in that the combination of averaging of at least two lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y ) or of at least two processed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) of a common lateral sensor axis ( A, B, C) or a selection of one of the lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or one of the processed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) corresponds to the maximum or minimum amount is. Method (100) according to Claim 1 , characterized in that the individual processed lateral movement information (aA, y, aB, y, aC, y) each come from a lateral acceleration (aMO7, y; aMO8, y; aMO9, y) or from a processed lateral acceleration (aMO7, y; aMO8, y; aMO9, y) whose measurement locations (MO7, MO8, MO9) are arranged along the central longitudinal axis of the vehicle (LA). Method (100) according to one of the Claims 3 until 5 , characterized in that the lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) each by a filtering or an integration can be prepared. Method (100) according to one of the Claims 1 until 6th , characterized in that the comparison is made via a difference formation or via one of the two position-dependent lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8 , y; aMO9, y) or lateral movement information (aA, y, aB, y, aC, y) standardized difference formation or a quotient formation of the at least two position-dependent lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or lateral movement information (aA, y, aB, y, aC, y) is performed. Method (100) according to Claim 7 , characterized in that when determining the type of collision, the comparison result is compared in a first further comparison with at least one threshold value in order to identify the type of collision. Method (100) according to Claim 8 , characterized in that when determining the type of collision at least one of the position-dependent lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or lateral movement information (aA, y, aB, y, aC, y) is compared in at least one second further comparison with at least one threshold value, the comparison result of the first further comparison and the comparison result of the at least one second further comparison with one another can be combined to identify the type of collision. Method (100) according to Claim 8 or 9 , characterized in that, when determining the type of collision, time courses of the position-dependent lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y) or lateral movement information (aA, y, aB, y, aC, y) are evaluated, the first further comparison and / or the second further comparison being carried out several times at different times, the comparison results of the first further comparisons and the second further comparisons are combined with one another in order to identify the type of collision. Method (100) according to one of the Claims 1 until 10 , characterized in that when determining the type of collision it is recognized whether the collision of the vehicle (1) with an obstacle (H) is a frontal collision with full overlap or an offset collision without angles and predominantly rotational properties or a Offset collision with an angle and strong lateral components or an offset collision with an angle and strong lateral components and a hooking with an obstacle (H). Sensor arrangement (10) for a vehicle (1), with at least two sensors (14), which are each designed, at a corresponding measuring location (MO1, MO2, MO3, MO4, MO5, MO6, MO7, MO8, MO9) in the vehicle ( 1) to record a position-dependent, superimposed lateral acceleration (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y), and one Evaluation and control unit (12), which is designed to record the position-dependent, superimposed lateral accelerations (aMO1, y; aMO2, y; aMO3, y; aMO4, y; aMO5, y; aMO6, y; aMO7, y; aMO8, y; aMO9, y), characterized in that the evaluation and control unit (12) is designed to carry out the steps of the method (100) for determining a type of collision of the vehicle (1) according to one of the Claims 1 until 11 perform. Sensor arrangement (10) according to Claim 12 , characterized in that the at least two measuring locations (MO1, MO2, MO3, MO4, MO5, MO6, MO7, MO8, MO9) have different distances (rA, rB, rC, rD, rE, rF) to a center of gravity (SP) of the Vehicle (1) have. Sensor arrangement (10) according to Claim 12 or 13th , characterized in that the individual sensors (14) along the central vehicle longitudinal axis (LA) through the center of gravity (SP) of the vehicle (1) and / or along a peripheral left side of the vehicle (1) and / or along a peripheral right side of the vehicle (1) are arranged. Computer program product with program code, which is stored on a machine-readable carrier, for performing the method according to one of the Claims 1 until 11 when the program is executed on an evaluation and control unit (12).

Images