DE10020084A1 - Method and device for controlling the release of a vehicle's crash protection device uses acceleration signals to form an acceleration input value invoked to determine a release criterion - Google Patents

Abstract

One or more acceleration signals (23) form an acceleration input value (28) that is invoked to determine a release criterion. To increase accuracy of recognition, a crash contact signal (25) from one or more crash contact sensors (20) forms a crash contact input value (30) that is linked to the acceleration input value so that operating threshold sensibility is raised if the crash contact input value displays the presence of a crash situation.

Description

The invention relates to a method according to the preamble of Claim 1 and a device according to the preamble of the Proverbs 8

A method and a device of the type mentioned at the beginning are z. B. known from DE 197 53 163 A1. With the well-known System are ge from the signal of an acceleration sensor received input variables are evaluated in a fuzzy classifier and from this the probability of the existence of a Crash situation calculated in which the airbag deploy should. It is understood that the term "acceleration" here and subsequently a translational and / or rotary Ge speed increase as well as a corresponding speed decrease in the meaning of a delay.

The input variables obtained are for this and similar ones known systems compared to the actually occurring Be accelerations strongly dampened, so that B. driving over Potholes or a curb does not become an unwanted one Airbag deployment leads. On the other hand, this means that the system in the event of actually triggering the Air bags justifying crash situation with appropriate carrier reacted.

Especially with side or window airbags, there is a desire to that this with comparatively low acceleration values should be triggered, but must be ensured at the same time  that the trigger does not occur when the user zer z. B. only violently slams the door of the motor vehicle.

The object of the present invention is therefore a method and a device of the type mentioned in that regard to further develop that on the one hand a crash situation in which the impact protection device should trigger, with high safety is recognized, but at the same time an unwanted end solve reliably is prevented.

This object is achieved by the ability specified in claim 1 ren and the device specified in claim 8 solved. At the inventive method and the invention direction will correspond to the determination of the existence the crash situation not just an acceleration signal, son also evaluated a crash contact signal. The crashkon Clock signal is obtained from a crash contact sensor, which is arranged in an area of the motor vehicle in which the relevant for the corresponding impact protection device bulging with high probability takes place. With this Be it can be rich in the case of a side or window airbag e.g. B. is a door assigned to this airbag. By the Crash contact sensor will indicate the presence of a crash situation shown.

Thus, in the present invention, an additional signal relevant for the detection of an impact, which is independent of the actually occurring acceleration is. As a result, such crash situations can also be considered relevant for triggering the impact protection device be known, in which in the conventional method detected acceleration and the corresponding acceleration signal is not sufficiently high for the corresponding triggering are. On the other hand, high accelerations, which in normal operation of a motor vehicle may occur, e.g. B. when the user closes the vehicle door or the vehicle  drives over a curb as when triggering the Auf impact protection device are not relevantly recognized.

In this way, the reliability in the detection ei relevant crash situation increased. The need for that To dampen acceleration signals so that the unwanted off solution of the impact protection device can be excluded can, does not apply or is at least significantly reduced. This in turn leads to a safe and quick release in the In the event of an actual crash situation.

Advantageous developments of the invention are in Unteran sayings.

In the training according to claim 2, the Be acceleration signal obtained acceleration input variable strengthened and thereby "sensitized" the system as a whole. This is an easy to implement variant ante of the method according to the invention.

Generally one is used for triggering the impact protection establishment relevant crash situation thereby determined and he knows that the one obtained from the acceleration signal Acceleration input variable compared with a limit value becomes. The lowering of this limit specified in claim 3 value if the crash contact input variable is present of a crash situation is easy to implement and may alternatively or in addition to that specified in claim 2 NEN further development of the invention can be applied.

In the linear gain specified in claim 4 Acceleration input variable and / or the limit value it is also an easy to implement variant of the method according to the invention.

In claim 5 is specified, the duration of influencing the Limit the acceleration input variable and / or the limit value,  e.g. B. to about 20 ms. This has the advantage of being Situations in which a release should take place from situations in which triggering does not take place should be separated more effectively, since processes that are not too egg triggering, usually run more slowly and thus take place later.

The basic idea for triggering the impact protection device To get over probabilities is to continue tion specified according to claim 6. Especially when opening a page plump multiple sources of information must be logical with each other are linked so that the "fuzzy" fuzzy Offers logic for processing.

An easy to implement and good result Fuzzy logic method is specified in claim 7.

The device according to claim 8 is suitable for implementation of the method according to the invention. Advantageous further training the device according to the invention can be found in the claims Chen 9 to 15.

The invention is described below with reference to exemplary embodiments len explained in detail with reference to the drawings. Here show:

Fig. 1: a sketched plan view of a motor vehicle, in which a first embodiment of a device for controlling the triggering of an impact protection device is installed;

Fig. 2 is a block diagram of the device of Fig. 1;

. Fig. 3 and 4 are diagrams in which the waveforms of input variables of the apparatus of Figure 1 are shown time-dependent;

. Fig. 5 is a view similar to Figure 1 with a second embodiment of an apparatus for Steue tion of triggering a Aufprallschutzeinrich tung;

Fig. 6 is a block diagram of the device of Fig. 5;

. Fig. 7 to 9: diagrams in which the waveforms of input variables of the apparatus of Figure 5 are shown time-dependent;

Fig. 10 is a block diagram of a third execution of a device for controlling the triggering of a impact protection device;

Are shown time-dependent graphs showing the curves of normalized input variables of the apparatus of Fig 10; Fig 11 to 13.. and

Is shown a diagram in which the total-release willingly option via a maximum ness single trigger willingly of acceleration sensors: Fig. 14.

A device for controlling the triggering of an impact protection device bears the overall reference number 10 in FIG. 1. It is installed in a motor vehicle 12 shown only schematically in the figure, which has wheels 14 shown in dashed lines in the region of its corner points.

The device 10 comprises a central control unit 16 arranged in the front area of the motor vehicle 12, which is connected to a central acceleration sensor 18 and a crash contact sensor 20 . The central acceleration sensor 18 is arranged centrally on the longitudinal axis of the motor vehicle 12 , whereas the crash contact sensor 20 is accommodated in the area of the driver's door.

The central control unit 16 is also connected to a side air bag 22 , which is also arranged in the area of the driver's door.

The crash contact sensor 20 in the present case is a deformation sensor which has two switching states: If there is no or no major deformation of the driver's door, if it has the switching state "open", if there is a greater deformation of the driver's door, it takes on the switching state "closed" included. In a game, not shown, a crash contact sensor is provided which works on the pressure-force principle. If necessary, this can be designed to detect different crash intensities in several stages, e.g. B. in the case of a smaller, a slight impact, the pressure corresponds to a first and, for a larger, a heavier impact to a corresponding pressure, a second closed position. Alternatively, it can also be a light barrier, a strain gauge or a capacitive or an inductive proximity sensor.

10 Fig. 2 shows the device as a block diagram. The central control unit 16 comprises two signal processors 24 and 26 . The signal processor 24 receives acceleration signals 23 from the central acceleration sensor 18 and converts them into an acceleration input variable 28 . The signal processor 26 is fed by deformation signals 25 of the deformation sensor 20 and converts them into a deformation input variable 30 . In the signal processors 24 and 26 , a signal processing can also take place, for. B. filtering and / or digitization.

The input variables 28 and 30 are fed to an evaluation circuit 32 , which in turn comprises an influencing circuit 34 . This essentially consists of a multiplier 36 , on which a timer 38 operates. An output variable 40 is fed from the multiplier 36 to a comparator 42 which is connected to a limit value transmitter 44 which supplies a limit value 43 . A signal 46 can be fed from the comparator 42 to a trigger circuit 48 , which can forward a corresponding trigger signal 50 to the side airbag 22 .

The device 10 functions as follows: from the central acceleration sensor 18 , the acceleration signals 23 are continuously output to the signal processor 24 , converted in this and passed on as an acceleration input variable 28 to the multiplier 36 . If there is no deformation in the area of the driver's door of the motor vehicle 12, the deformation sensor 20 is open, ie the deformation input variable 30 has the logical value zero. In this case, the acceleration input variable 28 is passed unchanged through the multiplier 36 and compared as an output variable 40 in the comparator 42 with the limit value 43 of the limit transmitter 44 . If the output variable 40 is below the limit value 43 , no signal is emitted to the trigger circuit 48 and the side airbag 22 is not triggered. This case is shown in Fig. 3 by a solid line.

However, as soon as there is a greater deformation on the driver's door of the motor vehicle 12 , this is detected by the deformation sensor 20 , ie it assumes the "closed" switching state. The corresponding switching signal 25 is supplied to the multiplier 36 via the signal processor 26 as the deformation input variable 30 , which now has the logical value one. This is recorded by the multiplier 36 and the acceleration input variable 28 is multiplied by a predetermined factor greater than one, which in the present case has the value two as an example. This means nothing other than that the acceleration input variable 28 is amplified by a factor of two. This case is shown in Figs. 3 and 4 by ge dashed lines 40 a, 30 a. In the case of the above-mentioned use of a multi-stage pressure contact switch as a crash contact sensor, the same applies accordingly. So then 40 correspond to a two-stage pressure switch, the dotted lines a, 30 a of the second switch stage, the closed position, that is, for severe impact, while dot-dash Li nien b 40, 30 b, the first switch stage having a closed position for easy impact represent.

This amplified output signal reaches the comparator 42 as an output variable 40 , where it is compared with the limit value 43 from the limit value sensor 44 . If the output variable 40 is above the limit value 43 , a signal 46 is output to the trigger circuit 48 . This finally emits a trigger signal 50 to the side airbag 22 , whereby an existing propellant charge (not shown) is ignited.

Although the acceleration signal 23 has the same value in each of the two cases just described, which in itself would not lead to a triggering in comparison with the limit value 43 , the fact that the signal 25 is also used by the deformation sensor 20 for evaluation can a situation can be recognized in which the deployment of the side airbag 22 is justified despite the "low" actual acceleration.

The multiplication factor is forcibly reset to one by the timer 38 after a predetermined period of time. The specified time period is a typical crash period, e.g. B. in the order of 20 ms. This prevents a slight amplification of the acceleration input variable 28 that is caused by slight, but remaining, deformation of the door panel, which would lead to false triggering of the side airbag 22 . In addition, this enables situations in which the deployment of the side airbag 22 is desired to be separated more effectively from other situations in which such a deployment is not desired, since the second type of situation generally runs more slowly and therefore takes place later .

A second exemplary embodiment of a device 10 for controlling the deployment of an airbag of a motor vehicle will now be described with reference to FIGS. 5 to 9. Parts which have equivalent functions to parts which have already been described in connection with the first exemplary embodiment have the same reference numerals and are generally not explained again in detail here.

In contrast to the previous exemplary embodiment, the device 10 in the present case comprises not only a central acceleration sensor 18 , but also an additional acceleration sensor 52 , also referred to as an “assistant”, which is arranged in the lateral region of the motor vehicle 12 . As can be seen in the block diagram of FIG. 6, the assistant 52 generates an additional acceleration signal 54 , which is fed to a signal processor 56 . There it is converted into an acceleration input variable 58 , which is fed into a comparator 60 . With this is connected via a multiplier 61 to a limit value 62 which specifies limit value sensor 64 . The multiplier 61 is in turn connected to a timer 65 . A signal 66 can be emitted from the comparator 60 to the trigger circuit 48 .

The acceleration input variable 28 , which is obtained from the acceleration signal 23 of the central acceleration sensor 18 , is fed directly to its own comparator 42 , which is connected via a multiplier 36 to a limit value transmitter 44 supplying a limit value 43 . A signal 46 can also be emitted from the comparator 42 to the trigger circuit 48 .

In the present case, the multipliers 36 and 61 therefore do not work on the input variables 28 or 58 of the acceleration sensors 18 or 52 , but on the limit values 43 and 62 .

In this case, the device 10 functions as follows: the central acceleration sensor 18 and the assistant 52 continuously deliver acceleration signals 23 and 54 , which are converted into acceleration input variables 28 and 58 in the signal processors 24 and 56 . The assistant 52 delivers a step-shaped acceleration signal 54 , which is indicated by the step shape of the acceleration input variable 58 in FIG. 8. The acceleration input variables 28 and 58 are compared in the comparators 42 and 60 with the limit values 43 and 62 , which are provided by the limit transmitters 44 and 64 .

If no deformation of the driver's door is detected by the deformation sensor 20 , this means that it has the switching state "open", the switching signal 25 and the deformation input variable 30 have the logic value zero. As a result, the limit values 43 and 62 do not change in the multipliers 36 and 61 . Such a case is shown in dash-dotted lines in FIGS. 7 and 8. In this case, after both acceleration input variables 28 and 58 are below the limit values 43 and 62 , the side airbag 22 is not deployed.

If deformation of the driver's door is detected by deformation sensor 52 , it assumes its "closed" switching state, as in cf. Fig. 9 illustrates, and this is shown to the multiplicators 36 and 61 via the deformation input 30 , which now has a logic value of one. In the multipliers 36 and 61 , the two limit values 43 and 62 are then reduced by corresponding factors, as shown in dashed lines in FIGS. 7 and 8. In the case of the multiplier 36, the factor is smaller than in the case of the multiplier 61 , the lowering of the limit value 43 is therefore smaller than that of the limit value 62 . As a result, peculiarities of the accelerations detected by the acceleration sensors 18 and 52 can be taken into account depending on the installation location.

As can be seen from FIG. 8, the acceleration 54 ascertained by the assistant 52 or the acceleration input variable 58 obtained therefrom exceeds the reduced limit value 62 , which is recognized by the comparator 60 , so that this signal 66 is sent to the trigger circuit 48 emits, which ignites the side airbag 22 via a trigger signal 50 .

Also in this embodiment, the timers 38 and 65 after z. B. 20 ms, the multiplication factors are forcibly reset to one, so the lowering of the limit values 43 and 62 is reversed to prevent false tripping.

A third embodiment of a device 10 for controlling an airbag will now be described with reference to FIGS . 10 to 14. Here too, parts that have functions equivalent to the previous exemplary embodiments have the same reference numerals and are generally not explained again in detail.

As in the previous second exemplary embodiment, two acceleration sensors, namely a central acceleration sensor 18 and an assistant 52 , and a deformation sensor 20 are provided in the driver's door of the motor vehicle 12 , see FIG. 5.

In the present case, the evaluation circuit 32 additionally comprises three standardizers 68 , 70 and 72 , in which the acceleration input variable 28 obtained from the acceleration signal 23 of the acceleration sensor 18 , the acceleration input variable 58 obtained from the acceleration signal 54 of the assistant 52 and the one from the Switching signal 25 of the deformation sensor 20 obtained deformation input variable 30 to individual triggers µ _B1 , µ _B2 and µ _C normalized, see Fig. 11 to 13. The normalizers 68 , 70 and 72 are in turn connected to a fuzzy logic 34 which has a maximum value generator 76 and a minimum value generator 78 . The latter is in turn connected to a comparator 42 . In an embodiment not shown, the normalizers also include an integrator, which averages the incoming input variables in the form of window integrals.

. The apparatus 10 illustrated in Figure 10 operates so follow, wherein in the following, only the essential differences will gen eingegan to the previous embodiments:

The maximum value generator 76 continuously determines the maximum value µ _B from the two individual triggering abilities µ _B1 and µ _B2 , which were obtained from the acceleration signals 23 and 54 of the central acceleration sensor 18 and the assistant 52 , according to the formula µ _B = Max [ µ _B1 , µ _B2 ]. This Ma ximalwert μ _B is forwarded to the minimum value generator 78, is fed to the addition μ the rertüre from the strain sensor 20 in the Fah obtained single-Auslösewilligkeit _C and in which the total Auslösewilligkeit μ _total according to the formula μ _total = Min [( 1 + µ _C ). µ _B , 1] is determined.

This formula means that if no deformation of the driver's door is detected by the deformation sensor 20 , that is to say that it has the switching state "open" (the individual triggering sensitivity μ _C thus has the probability value zero), the total triggering sensitivity μ _total that in the Maximum value generator 76 corresponds to the maximum individual triggering response µ _B. This case is represented by a solid line in FIG. 14.

However detected rertüre from the deformation sensor 20, a deformation of the Fah, so it is in the switching state "ge included", is obtained from the distortion input 30 in the ent speaking normalizer 72, a single-Auslösewilligkeit μ forms _C ge which the probability value less than or equal one Has. In this case, the maximum-78 determined in the total Auslösewilligkeit corresponds μ _total up to twice the value of the maximum single-Auslösewilligkeit μ _B. This case is shown in Fig. 14 by a broken line. The overall system is therefore up to twice as sensitive as in the case of an open deformation sensor 20 . In the case of using a two-stage pressure switch as a crash contact sensor, the dashed lines in FIGS . 13 and 14 represent the second switch stage for severe impact.

The respective total willingness to trigger µ _total is then compared in the comparator 42 with a limit value 43 which, for. B. can be set to 0.5. If the limit value 43 is exceeded, a signal 46 is emitted to the trigger circuit 48 , which then triggers the side airbag 22 via a trigger signal 50 .

It goes without saying that instead of the others described Fuzzy logic designs with a different logical connection and / or weighting of the input variables can be used. Further it is understood that the above-described embodiments can also be applied to other impact protection devices can. These include e.g. B. belt tensioners, door releases, Emergency calls, etc. Furthermore, to form the off other input variables are also used which z. B. of speed, temperature or chemical sensors.

Claims (15)

1. A method for controlling the triggering of an impact protection device of a motor vehicle, in which an acceleration input variable is formed from at least one acceleration signal from an acceleration sensor and is used to determine a triggering criterion, characterized in that at least one crash contact sensor from a crash contact signal ( 25 ) ( 20 ) a crash contact input variable ( 30 ) is formed, which is linked to the acceleration input variable ( 28 ) in such a way that the response sensitivity is increased when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 2. The method according to claim 1, characterized in that the acceleration input variable ( 28 ) is at least temporarily amplified when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 3. The method according to any one of claims 1 or 2, characterized in that a limit value ( 43 , 62 ) used to determine the presence of a corresponding crash situation is at least temporarily lowered when the crash contact input variable ( 30 ) indicates the presence of a crash situation . 4. The method according to any one of the preceding claims 2 or 3, characterized in that the acceleration input variable ( 28 ) and / or the limit value are multiplied by a factor at least temporarily when the crash contact input variable ( 30 ) indicates the presence of a crash situation . 5. The method according to any one of claims 2 to 4, characterized in that the duration of the amplification of the acceleration input variable ( 28 ) and / or the lowering of the limit value is limited to a predefinable crash period. 6. The method according to claim 1, characterized in that the input variables ( 28 , 30 , 58 ) are standardized and the standardized values (µ _B1 , µ _B2 , µ _C ) are linked together in accordance with a fuzzylo logic ( 34 ) in such a way that a Total triggering willingness (µ _total ) is formed. 7. The method according to claim 6, characterized in that in the fuzzy logic ( 34 ) determines the maximum µ _{B of} the normalized input variables of the acceleration sensors and the total triggering ability µ _total according to the relationship µ _total = Min [(1 + µ _C ). µ _B , 1] are formed, where µ _{C is} the normalized input variable of the crash contact sensor. 8. Device for performing the method according to one of claims 1 to 7, with
at least one acceleration sensor that provides an acceleration signal,
a signal processor that generates an acceleration input variable from the acceleration signal,
an evaluation circuit which uses the acceleration input variable to determine a triggering criterion,
characterized in that
At least one crash contact sensor ( 20 ) arranged in an impact-relevant area of the motor vehicle ( 12 ) and a corresponding signal processor ( 26 ) are provided, which provides the evaluation circuit ( 32 ) with a crash contact input variable ( 30 ) which the evaluation circuit ( 30 ) does Indicates the presence of a crash situation, and
the evaluation circuit ( 32 ) links the crash contact input variable ( 30 ) to the acceleration input variable ( 28 ) in such a way that the response sensitivity is increased when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 9. The device according to claim 8, characterized in that the evaluation circuit ( 32 ) has an influencing circuit ( 34 ) which at least temporarily amplifies the acceleration input variable ( 28 ) when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 10. Device according to one of claims 8 or 9, characterized in that the evaluation circuit ( 32 ) comprises a comparator ( 42 , 60 ) in which a comparison with a limit value ( 43 , 62 ) is used to determine the presence of a crash situation. is carried out, and the evaluation circuit ( 32 ) has an influencing circuit ( 34 ) which at least temporarily lowers the limit value ( 43 , 62 ) when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 11. Device according to one of claims 9 or 10, characterized in that the influencing circuit ( 34 ) comprises a multiplier ( 36 , 61 ) in which the acceleration input variable ( 28 ) and / or the limit value ( 43 , 62 ) at least temporarily multiplied by a factor when the crash contact input variable ( 30 ) indicates the presence of a crash situation. 12. The device according to one of claims 9 to 11, characterized in that the evaluation circuit ( 32 ) comprises a timer ( 38 , 65 ) by means of which the duration of the increase in the acceleration input variable ( 28 ) and / or the decrease in the limit value ( 43 , 62 ) is limited to a predefinable crash period. 13. The apparatus according to claim 8, characterized in that the evaluation circuit ( 32 ) comprises at least one normalizer ( 68 , 70 , 72 ), with which the input variables ( 28 , 30 , 58 ) are normalized, and a fuzzy logic circuit ( 34 ) , in which the standardized values (µ _B1 , µ _B2 , µ _C ) are linked together in such a way that a total triggering willingness (µ _total ) is formed. 14. The apparatus according to claim 13, characterized in that in the fuzzy logic circuit ( 34 ) the maximum µ _{B of} the standardized input variables of the acceleration sensors ( 18 , 52 ) is determined and the total triggering µ _total according to the relationship µ _total = Min [( 1 + µ _C ). µ _B , 1] are formed, where µ _{C is} the standardized input variable of the crash contact sensor ( 20 ). 15. Device according to one of claims 8 to 14, characterized in that the crash contact sensor works on the pressure-force principle and / or comprises a light barrier, a strain gauge, a capacitive, inductive and / or magnetic proximity switch ( 20 ).