DE102006056836B4 - Method and device for controlling personal protective equipment - Google Patents

Abstract

Method for controlling personal protection means (PS), wherein the personal protection means (PS) in response to an acceleration signal of an acceleration sensor (BS 1 to 3), which is arranged on the bumper, are driven, wherein for the control of a time course of a signal derived from the acceleration signal is evaluated such that depending on at least one signal value, which identifies an ejection of an impact object from the bumper, at least one feature is generated and that depending on the at least one feature the control is characterized in that the ejection based on the time course of the acceleration signal is detected, wherein the at least one signal value follows an absolute minimum.

Description

State of the art

The invention relates to a method and a device for controlling personal protection means according to the preamble of the independent claims.

Out DE 10 2005 000 657 A1 a method for offset detection for a pedestrian protection device is known. In this case, depending on an acceleration signal from acceleration sensors, which are arranged on the bumper, an offset of a point of impact is determined in comparison to the vehicle centerline. Absolute values, difference values, summation values or maximum or minimum values of the sensor data can be used for this.

The DE 10 2005 042 512 A1 discloses a method for activating a pedestrian protection device of a motor vehicle.

The DE 10 2004 042 467 A1 discloses a method and apparatus for generating a tripping signal for a pedestrian protection device.

The DE 10 2004 018 356 A1 discloses a collision object discriminating device that can be installed in a vehicle.

Disclosure of the invention

The method according to the invention or the device according to the invention for controlling personal protection devices with the features of the independent patent claims have the advantage that a more precise characterization of the impact object is achieved by the generation of the features as a function of at least one signal value which indicates the ejection of an impact object from the bumper is possible because it has been found that the relaxation in the Auswerfvorgang the impact object of the bumper is a more pronounced signal than the signal that characterizes the penetration. This makes a better assessment of the impact possible. This ejection signal is the more pronounced the lower the mass of the impact object is. However, the determination of the mass is an essential component of a pedestrian discrimination algorithm, so that according to the invention this discrimination is considerably improved.

The method according to the invention or the device according to the invention use a signal derived from the acceleration signal, wherein the acceleration signal here can also represent signals from a plurality of acceleration sensors. This derived signal may be the acceleration signal itself, for example, after low-pass filtering, path-based, or velocity-based, or combinations thereof. The features may then be generated from the derived signal, and then the detected signal value identifying the ejection of the impact object, and these features are then evaluated to determine the drive.

As an interface, an integrated circuit, a discrete circuit, combinations thereof, multiple integrated circuits or a software interface may be provided. The acceleration sensor, which is arranged on the bumper, is installed behind the bumper fascia. The acceleration sensor system may be one, two, three, four or more acceleration sensors. These acceleration sensors are usually provided with a micromechanically produced sensor element and have signal conditioning and transmitter modules.

The evaluation circuit is usually a microcontroller, but other processor types or ASICs are possible. According to the invention, the evaluation circuit has an evaluation module, which may be hardware and / or software-based. Accordingly, the detection module, the element of the Auswertemodüls is formed in hardware and / or software. This also applies to the feature module and for the control module, which is not the element of the evaluation module, but the evaluation circuit.

The personal protective equipment is airbags, belt tensioners, crash-active headrests, roll bars and pedestrian protection in particular, such as a liftable bonnet, exterior airbags and other well-known pedestrian protection.

The measures and refinements recited in the dependent claims, advantageous improvements of the method specified in the independent claims or the device specified in the independent claims for the control of personal protection means are possible.

It is particularly advantageous that, based on the time profile of the acceleration signal, the ejection of the impact object is detected. This signal value is detected as an absolute maximum following an absolute minimum. These designations can be seen in particular within a given time after the impact. Ie. The absolute minimum here denotes the largest minimum in terms of magnitude and then the absolute maximum of the acceleration signal follows. These may be individual acceleration signals or sums of acceleration signals.

Furthermore, it is advantageous that the signal is path-based, i. the second integral of the acceleration is generated, for example, by a two-fold integration, whereby here integration is to be understood pragmatically, so that also a summation or other equivalent measures, such as an averaging, which are used for integration, are understood by it. Due to the way-based feature, the relaxation of the bumper is easily recognizable. This feature can then be evaluated either absolutely or in relation to the bumper indentation.

$$\text{Merkmal\_Dv\_2}=\left(\left|\text{Max\_Dv}\right|-\left|\text{Min\_Dv}\right|\right)/\left|\left|\text{Max\_Dv}\right|+\left|\text{Min\_Dv}\right|\right|$$

$$\text{Merkmal\_Dv\_3=}{\displaystyle \int \text{Dv}\left(\text{t}\right)\text{ dt}\left|\left(\text{falls Dv}>0\right)/{\displaystyle \int \text{Dv}\left(\text{t}\right)\text{ dt}}\right|\left(\text{falls Dv}<0\right)}$$

Furthermore, it is advantageous that the signal is generated speed-based, ie by simple integration of the acceleration or other appropriate measures. In this case, a first maximum value and a first minimum value of the signal is used to generate the at least one feature. Since the relaxation of the bumper, ie the return oscillation of the second integral, is caused precisely by positive values of the first integral, the following features can be used for the quantification: $$\text{Merkmal\_DV\_1}=\text{Max\_Dv}/\text{Min\_Dv}$$

$$\text{Merkmal\_Dv\_2}=\left(\left|\text{Max\_Dv}\right|-\left|\text{Min\_Dv}\right|\right)/\left|\left|\text{Max\_Dv}\right|+\left|\text{Min\_Dv}\right|\right|$$

$$\text{Merkmal\_Dv\_3 =}{\displaystyle \int \text{dv}\left(\text{t}\right)\text{dt}\left|\left(\text{if Dv}>0\right)/{\displaystyle \int \text{dv}\left(\text{t}\right)\text{dt}}\right|\left(\text{if Dv}<0\right)}$$

Dv can be either the first integral of the impact nearest sensor or the sum of the first integrals across all sensors.

$$\text{Merkmal\_Acc\_2}=\left(\left|\text{MaxAcc}\right|-\left|\text{MinAcc}\right|\right)/\left|\left|\text{MaxAcc}\right|+\left|\text{MinAcc}\right|\right|$$

$$\text{Merkmal\_Acc\_3=}{\displaystyle \int \text{a}\left(\text{t}\right)\text{ dt}\left|\left(\text{falls a}>0\right)/{\displaystyle \int \text{a}\left(\text{t}\right)\text{ dt}}\right|\left(\text{falls a}<0\right)}$$

Advantageously, the signal can of course also be acceleration-based, in which case the following features can then be used: $$\text{Merkmal\_Acc\_1}=\text{MaxAcc}/\text{MinAcc}$$

$$\text{Merkmal\_Acc\_2}=\left(\left|\text{MaxAcc}\right|-\left|\text{MinAcc}\right|\right)/\left|\left|\text{MaxAcc}\right|+\left|\text{MinAcc}\right|\right|$$

$$\text{Merkmal\_Acc\_3 =}{\displaystyle \int \text{a}\left(\text{t}\right)\text{dt}\left|\left(\text{if a}>0\right)/{\displaystyle \int \text{a}\left(\text{t}\right)\text{dt}}\right|\left(\text{if a}<0\right)}$$

A of D may be either the acceleration signal of the next impact sensor or the sum of all sensor signals.

The features described, with the exception of Dv_2 and Acc_2, assume large values in the case of a pronounced backswing process. Large feature values thus indicate a low mass of the impact object. For the characteristics Dv_2 and Acc_2, the behavior is just inverse.

The choice of the individual characteristics which contribute to the discrimination can be made application-dependent. Each of the selected features is compared with an application-dependent threshold and fed to a trigger logic. The threshold value can be constant or time-dependent. It is advantageous to set this threshold depending on the speed, which may be the airspeed of the car Vehicle or the integrated acceleration signal, and / or to vary from the temperature and / or the determined impact position.

The comparison of the features with the threshold values can be carried out in a certain trigger-relevant time window minimum and maximum time from algorithm start. The algorithm start can be determined as a function of exceeding a threshold, for example a noise threshold. Alternatively, as a reference point for the time window and a characteristic feature in the waveform, z. B. the achievement of maximum intrusion are used. Alternatively, instead of a time window, an evaluation window can also be predefined by a further variable, for example an evaluation can take place as long as the absolutely integrated signal is between a lower and an upper limit.

Embodiments of the invention are illustrated in the drawings and are explained in more detail in the following description.

• 1 ein Blockschaltbild mit der erfindungsgemäßen Vorrichtung,
• 2 Module auf der Auswerteschaltung,
• 3 ein erstes Blockdiagramm,
• 4 ein zweites Blockdiagramm,
• 5 ein drittes Blockdiagramm,
• 6a bis c Zeitdiagramme der Signale bei einem mittigen Fußgängeraufprall,
• 7 der Verlauf des Beschleunigungssignals bei einem mittigen Fußgängeraufprall und
• 8a und b der Vergleich von unterschiedlichen schweren Aufprallobjekten in Bezug auf das integrierte und zweifach integrierte Beschleunigungssignal.

Show it

• 1 a block diagram with the device according to the invention,
• 2 Modules on the evaluation circuit,
• 3 a first block diagram,
• 4 a second block diagram,
• 5 a third block diagram,
• 6a to c time diagrams of the signals in a central pedestrian impact,
• 7 the course of the acceleration signal in a central pedestrian impact and
• 8a and b comparing different heavy impact objects with respect to the integrated and dual integrated acceleration signal.

1 shows a block diagram of the device according to the invention SG as a control device for controlling personal protection devices, in particular pedestrian protection. To the control unit SG are three built-in behind the bumper acceleration sensors BS1 . BS2 and BS3 connected via an interface IF. The acceleration sensors BS1 . BS2 and BS3 have a micromechanically produced sensor element, wherein the resulting measurement signal in the acceleration sensors BS1 to 3 is pre-processed to then digitally, for example, to transmit via a powerline data transmission to the interface IF. The interface IF converts the received data into a format for the evaluation circuit μC, ie a microcontroller. To transfer the data from the interface IF to the microcontroller .mu.C, the so-called SPI (Serial Peripheral Interface) bus is used. The microcontroller .mu.C processes the sensor signals in the manner according to the invention by loading the corresponding algorithms and modules from the memory S.

In this case, based on the acceleration signals, the penetration process and the ejection process of the impact object are detected on the basis of the acceleration signal, and a control signal is then generated by signal processing steps involving the generation of features from the signals and the threshold value of the features. This drive signal is also transmitted via an SPI bus to the drive circuit FLIC, which is also present as a single integrated device or a plurality of integrated devices. The drive circuit FLIC has, in particular circuit breaker, which allow energization of the personal protection means in the event of triggering. Other sensors and components which are connected to the control unit SG or are in the control unit SG have been omitted for the sake of simplicity. The control unit SG is usually the airbag control unit or a safety control unit that regulates the entire personal protection in the vehicle.

2 schematically illustrates the software modules that may be present on the microcontroller μC. The microcontroller .mu.C has the evaluation module as main module 20 on, that is a detection module 21 and a feature module 22 having. Furthermore, a drive module 23 provided and optionally an interface 24 to which, for example, sensors can be connected. Additional software modules may be provided. It is possible that these modules are also present in terms of hardware or from a combination of hardware and software, wherein, for example, certain circuits are exclusively assigned to these modules on the evaluation circuit.

The evaluation module 20 searches with the detection module 21 according to the signal characteristics indicating the intrusion and ejection of the impact object. If the ejection was found, then by means of these signal values by means of the feature module 22 then generates the corresponding features and compared with thresholds. These threshold values can be constant or variable as stated above. In particular, a time window or evaluation window control is possible. Was a triggering case by the evaluation module 20 then this becomes the module 23 communicated, which is configured as a drive module. This drive module 23 then generates by means of the evaluation circuit μC the drive signal, which is then transmitted to the drive circuit.

3 shows in a first block diagram the sequence of the method according to the invention. In process step 300 is due to the acceleration sensor BS1 to 3 the acceleration provided as a value. The acceleration can thereby by the process step 301 be filtered, in particular be subjected to a low-pass filtering. As according to 3 the acceleration signal is evaluated as such, three different evaluation branches are presented here, all or only a subset of which can be used according to the invention. In a first branch, the filtered acceleration signal is in process step 308 integrated, and that is then in process step 309 the fraction of the acceleration signal that is positive and has been integrated is divided by the fraction of the acceleration signal that is negative and integrated to produce a feature. In process step 310 For example, this feature is compared to a threshold that may be constant or variable as shown above to determine if that feature indicates a trigger event. This result becomes a trigger logic 306 communicated. This then decides on the basis of this and / or other results, whether the generation of the drive signal is to take place or not.

In another branch is in process step 302 the minimum, namely the absolute minimum, in the acceleration signal, namely the time course, searched. This absolute minimum marks the intrusion of the impact object into the bumper. In process step 303 the following absolute maximum is sought, which identifies the ejection of the impact object from the bumper. With the minimum and maximum then different features can be generated. In process step 304 is generated by a quotient of the maximum and the minimum a feature that in process step 305 subjected to a threshold test. This result also becomes the trigger logic 306 communicated. Another feature can be generated by the fact that in process step 307 an absolute difference of the maximum and the minimum is divided by the summation of the maximum and the minimum. This feature is then in process step 308 also subjected to a threshold. This is the result of the tripping logic 306 communicated. The trigger logic 306 then decides from the results whether the drive signal is generated or not.

4 shows in a second block diagram the generation of the features of the speed-based signal. In process step 400 the acceleration signal is provided, the in process step 401 is filtered. The acceleration signal is in process step 402 integrated in the manner described above. Again, there are three features present that can be generated. In a first branch is then in process step 403 a division of the integrated speed signal, which is positive, performed by the integrated speed signal, which is negative. This quotient is in process step 404 compared with a threshold and the result becomes the trigger logic 405 communicated. In a middle branch is in process step 406 searched for the minimum in the speed signal and in process step 407 the maximum. In process step 408 Then a quotient of maximum and minimum is formed, which is a feature in method step 409 a threshold decision is supplied. The result of the threshold comparison again becomes the trigger logic 405 communicated. In process step 410 becomes analogous process step 307 a difference of the magnitudes of the maximum of the speed signal and the minimum of the speed signal is performed by a sum of the magnitudes of the maximum of the speed signal and the minimum of the speed signal to form a further feature, which in step 411 a threshold decision is supplied. This threshold value is also the tripping logic 405 supplied, which then decides whether the drive signal should be generated or not.

5 shows in a third block diagram the inventive method by means of a path-based signal. In process step 500 the acceleration signal is provided, the in process step 501 is filtered. In process step 502 the double integration takes place in the manner described above. In process step 503 the minimum of the path-based signal is searched. Then in process step 504 formed by forming the difference of the path-based signal with the minimum of the path-based signal, a first feature, the in process step 505 a threshold decision is supplied and this result of Schwellwertentscheiders is the tripping logic 506 fed. In process step 507 This difference is related to the minimum of the path-based signal, so that a second feature is generated, which is the threshold decision 508 is supplied. The result of the threshold decision maker 508 becomes the trigger logic 506 fed. The trigger logic 506 decides depending on the results of thresholds 505 and 508 Whether the drive signal should be generated or not.

It is possible that single or all branches in the 3 . 4 and 5 represented, can be combined with each other.

6 shows the acceleration signal, the first integral and the second integral in a central pedestrian impact detected by a center accelerometer. At time 0.008, the minimum MinAcc, ie the absolute minimum of the acceleration signal is detected over time. At time 0.016, the absolute maximum of the acceleration signal following the absolute minimum is detected. This is marked by MaxAcc. The acceleration signal is marked in the upper diagram. In the middle diagram, the first integral is shown, which shows the corresponding minimum Min_Dv and the maximum Max_Dv, which of course do not coincide with the times because of the integration. In the lower diagram, the second integral is shown, where the minimum indicates MinDs, which is also present at a different time than the characteristic signal values of the acceleration signal.

As expected, a negative acceleration signal can be detected in the first milliseconds, which is caused by the penetration of the leg impactor. It is interesting that after about 10 ms, the acceleration signal shows a pronounced positive signal, ie MaxAcc, whose cause is to be found in the relaxation of the bumper associated with the ejection of the impactor in the direction of travel (conservation of momentum). The sign change takes place at maximum penetration depth, see also the second integral, thereby separating the penetration of the impact object from the ejection. It is striking that the ejection signal is more pronounced than the intrusion signal.

7 shows the acceleration signal at a central impact detected by the central accelerometer. The ejection signal is as shown above and in FIG 7 once again, more pronounced than the intrusion signal. This is the more pronounced the lower the mass of the impact object. Since the determination of the mass of the impact object is an integral part of the pedestrian discrimination algorithm, the above features have been proposed.

8a and b show the mass dependency of the ejection operation using the example of two pedestrian impacts of 13, 5 and 3.5 kg mass, respectively. In 8a is the impact with 13, 5 kg and in 8b the impact with 3.5 kg each represented as speed time and Wegzeitdiagramm. It can be seen that the relaxation of the bumper and the evaluation of the object occurs earlier in the 3.5 kg attempt and appears more pronounced. To quantify the bumper relaxation or ejection process, the above-mentioned features offer.

Claims (10)

Method for controlling personal protection means (PS), wherein the personal protection means (PS) in response to an acceleration signal of an acceleration sensor (BS 1 to 3), which is arranged on the bumper, are driven, wherein for the control of a time course of a signal derived from the acceleration signal is evaluated such that depending on at least one signal value, which identifies an ejection of an impact object from the bumper, at least one feature is generated and that depending on the at least one feature the control is characterized in that the ejection based on the time course of the acceleration signal is detected, wherein the at least one signal value follows an absolute minimum. Method according to Claim 1 , characterized in that the ejection is detected on the basis of the time profile of the acceleration signal, wherein the at least one signal value is detected as an absolute maximum. Method according to Claim 1 or 2 , characterized in that the signal is generated away-based, wherein the at least one feature is generated as an absolute value and / or related to a depression of the shock absorber. Method according to one of the preceding claims, characterized in that the signal is generated speed-based, wherein for generating the at least one feature, a first maximum value and a first minimum value of the signal are used. Method according to one of the preceding claims, characterized . in that the signal is acceleration-based, wherein a second maximum value and a second minimum value of the signal and / or a respective integral of positive and negative acceleration values are used to generate the feature. Method according to one of the preceding claims, characterized in that the signal is generated by a collision-next acceleration sensor or by all acceleration sensors. Method according to one of the preceding claims, characterized in that the activation is carried out as a function of a threshold value comparison of the at least one feature, for which at least one threshold is used, which is performed constant, time-dependent, speed-dependent, temperature-dependent and / or dependent on the point of impact. Method according to Claim 7 , characterized in that the threshold value comparison is performed in a time window. Method according to Claim 7 , characterized in that the threshold value comparison is performed in an evaluation window. Device for activating personal protective equipment (PS) with: - At least one interface (IF), which provides an acceleration signal of an acceleration sensor (BS 1 to 3), which is arranged on the bumper; - An evaluation circuit (.mu.C), which has an evaluation module (20) for evaluating a time course of a signal derived from the acceleration signal, wherein the evaluation module (20) comprises a detection module (21) for detecting at least one signal value, which marks an ejection of an impact object wherein the ejection is detected on the basis of the time profile of the acceleration signal, wherein the at least one signal value follows an absolute minimum, wherein the evaluation module (20) further comprises a feature module (22) for generating at least one feature in response to at least one signal value the evaluation circuit (μC) further comprises a control module (23) for controlling the personal protection means (PS) as a function of the at least one feature.

Images