Method for evaluating signals from from at least two acceleration sensors

Abstract

The invention relates to a method for evaluating signals from at least two acceleration sensors in a vehicle (10) in the event of a collision, wherein the acceleration sensors are mechanically coupled to one another and the signals from the acceleration sensors describing an initial movement of the acceleration sensors at the start of the collision in a direction against a collision direction (22) are evaluated.

Description

[0001] Description

[0002] title

[0003] The invention relates to a method for evaluating signals from at least two acceleration sensors in a vehicle, in particular in a motor vehicle, during a collision and an arrangement for carrying out the method.

[0004] State of the art

[0005] Restraint systems are passive safety devices in vehicles. These are designed, for example, to secure vehicle occupants in their seats. This is intended to protect them from the effects of excessive acceleration and from collisions with components of the vehicle structure.

[0006] Airbags, for example, are used as restraint systems. These include driver airbags, passenger airbags, side and head airbags, etc. The signals for triggering restraint devices are typically generated based on the evaluation of acceleration signals. The decisions to activate the restraint devices are therefore made based on this evaluation.

[0007] To effectively activate restraint devices, it is necessary to both detect a collision and determine the type or nature of this collision as quickly as possible. For example, it is generally important whether the collision is head-on or offset. The detection of head-on collisions in airbag control units is based, for example, on acceleration sensors that sense in the x-direction. These sensors can usually be located centrally on the tunnel, but also at positions on the vehicle periphery, such as at the front end or symmetrically on both B-pillars, or at other locations. The deployment algorithms used decide whether or not an airbag should be deployed based on the measured acceleration signals and the processed sensor signals.

[0008] As already mentioned, collision or crash type detection is useful to support the deployment decision for individual restraint devices. Different collision types, e.g., collisions with full overlap versus partial overlap, also referred to as offset, require different deployment times and, if necessary, the activation of other and / or additional restraint devices, such as the activation of head airbags in an offset crash. This collision type detection can be achieved, among other things, using sensors in the vehicle's front end using so-called upfront sensors. In particular, a distinction between collision types, namely full overlap versus partial overlap, so-called offset detection, can be made by comparing the signals from two upfront sensors, one on the left and one on the right at the front end.

[0009] Document DE 10 2013 223 781 A1 describes a method for selecting at least one lateral restraint device of a vehicle before and / or during an impact. In the method, sensor signals from an impact sensor system of the vehicle are read in, and an offset impact of the vehicle is detected using the sensor signals. Offset information is provided when the offset impact is detected.

[0010] A method for crash type detection for a vehicle is known from DE 10 2005 042 198 A1. In this method, sensor data from the left and right sides of the vehicle are recorded and evaluated via at least two sensor units. The sensor data recorded after the impact are linked with one another, with a first link of the recorded sensor data being checked for fulfillment of a stability criterion, and at least a second link of the recorded sensor data being compared with a predetermined threshold value after the stability criterion has been met. A first criterion for an asymmetric impact signal is detected if the second link of the recorded sensor data exceeds and / or falls below the predetermined threshold value.

[0011] Depending on the mounting position of the upfront sensors, it is possible that the acceleration signals at the beginning of a collision, especially without an offset, do not point in the direction of the collision, but rather opposite to it. With a directed evaluation of the acceleration signals, only in the direction of the collision, a triggering decision based on the upfront signals at the beginning of the collision is not possible in such a case.

[0012] Since the movements of the upfront sensors become increasingly unstable with increasing duration of the collision and the progressive destruction of the crash structures, a comparison of the left and right upfront signals, as described in the document DE 10 2005 042 198 A1, can lead at later times, even for frontal collisions with full overlap, to an offset being assumed, although this is not the case, which in turn leads to restraint devices not being controlled in accordance with the collision.

[0013] Disclosure of the invention

[0014] Against this background, a method having the features of claim 1 and an arrangement according to claim 12 are presented. Embodiments emerge from the dependent claims and from the description.

[0015] The presented method is used to evaluate signals from at least two acceleration sensors in a vehicle, evaluating the signals generated in the event of a collision. The at least two acceleration sensors are mechanically coupled to each other. The method evaluates the signals that describe an initial or initial movement of the acceleration sensors at the beginning of the collision, opposite to the collision direction.

[0016] The presented method is thus used, for example, to influence a triggering decision and / or to determine a collision type in a vehicle. The method uses at least two acceleration sensors that are mechanically coupled. This mechanical coupling ensures that the accelerations of the two acceleration sensors are interdependent and that these accelerations are recorded accordingly.

[0017] Sensor signals generated by an initial movement of the acceleration sensors at the beginning of the collision are then evaluated to influence the triggering decision or to determine the collision type.

[0018] For example, the period from 10 to 15 ms after the start of the collision can be evaluated. This means that signals caused by movement or acceleration of the acceleration sensors during this period are evaluated.

[0019] The method is based on the finding that, depending on their installation position, the upfront sensors in vehicles may not move in the direction of the collision at the beginning of a crash, but rather against it. Such a movement can occur, for example, for upfront sensors on the vehicle's bending beam. These upfront sensors are designed as acceleration sensors.

[0020] The described arrangement serves to carry out the presented method and is implemented, for example, in hardware and / or software. The arrangement can be integrated into a control unit of a vehicle or designed as such. To carry out the method described herein, the arrangement has an evaluation unit or means for evaluation. Furthermore, the arrangement can be assigned at least one unit that is designed to control at least one restraint device. With the presented method, it is possible to detect a movement opposite to the collision direction at the beginning of a collision and, through this detection, to influence the existing offset detection, which evaluates the subsequent movement of the upfront sensors in the collision direction, as well as the triggering decision.

[0021] In this way, it is possible to influence the triggering decision at a very early stage, when the signals from the acceleration sensors, e.g., the upfront signals, point opposite to the collision direction. This is particularly advantageous in collisions with full overlap, since these collisions, on the one hand, have the earliest required triggering times and, on the other hand, the movement of the acceleration sensors, e.g., the upfront sensors, to be evaluated, occurs opposite to the collision direction.

[0022] This also prevents potentially false offset detection and ensures correct activation of all restraint devices.

[0023] For this purpose, a movement opposite to the collision direction is detected at the beginning of a collision and this information is used to influence the triggering decision as well as the existing offset detection.

[0024] The improved deployment decision, which influences an existing deployment decision by detecting a movement opposite to the collision direction at the beginning of a collision, enables a rapid deployment decision early in the collision phase, thus enabling timely activation of the required restraint devices. Since the existing deployment decision evaluates not only the signals from the acceleration sensors, such as the upfront signals, but also signals from an x-sensor installed behind the crumple zone, e.g., in the airbag control unit, this sensor ensures that the correct collision direction is present in a frontal collision.

[0025] The improved offset detection by influencing an existing

[0026] Offset detection by means of a detected movement opposite to the collision direction at the beginning of a collision enables correct collision type detection and thus the timely activation of the restraint devices required for the determined collision severity and the determined collision type.

[0027] In summary, the detected countermovement of the acceleration sensors, especially the upfront sensors, at the onset of a collision can directly influence the triggering decision. This can also be achieved indirectly via improved collision type determination, which in turn influences the triggering decision.

[0028] Further advantages and embodiments of the invention will become apparent from the description and the accompanying drawings.

[0029] It is understood that the features mentioned above and those to be explained below can be used not only in the combination specified in each case, but also in other combinations or on their own, without departing from the scope of the present invention.

[0030] Short description of the drawings

[0031] Figure 1 shows a schematic top view of a vehicle with two upfront sensors and resulting movements in a frontal collision with full overlap.

[0032] Figure 2 shows a schematic top view of a vehicle with two upfront sensors during an offset collision.

[0033] Figure 3 shows six graphs of examples of sensor signals or features and their combination, including possible threshold values.

[0034] Figure 4 shows a flowchart of a possible sequence of the presented method. Figure 5 shows a schematic representation of a vehicle with an embodiment of the described arrangement for implementing the method.

[0035] Embodiments of the invention

[0036] The invention is illustrated schematically in the drawings using embodiments and is described in detail below with reference to the drawings.

[0037] Figure 1 shows a schematic plan view of a vehicle, designated overall by reference numeral 10. In the vehicle 10, two upfront sensors 14, 16 are attached to a flexural beam 18 in the front end area of ​​the vehicle front 12. The two upfront sensors 14, 16 are thus mechanically coupled to one another via the flexural beam 18 and are embodiments of the acceleration sensors described herein.

[0038] The intrusion of a frontal crash with full overlap, as indicated by an arrow 22, can lead to a deformation of the bending beam 18. This deforms the bending beam 18 in the center towards the vehicle interior 24 and at its ends towards the vehicle front 15. This results in an initial movement of the upfront sensors 14, 16 in a direction 26 opposite to the collision direction 22. As the collision progresses, this transitions into a movement or deformation in the collision direction 22. This further phase is evaluated by the previously existing offset detection systems. In contrast, the presented method evaluates the initial movement of the upfront sensors 14, 16.

[0039] Such an initial countermovement phase occurs primarily in upfront positions on the flexure beam. It is less frequently observed at other positions, such as the headlight mount or the radiator cross member. Figure 2 shows a schematic plan view of a vehicle 50 in which two upfront sensors 54, 56 are attached to a flexure beam 58 in the front end area of ​​the vehicle front 52. Arrows 62 illustrate a frontal collision with partial overlap, in which the upfront sensor 56 facing away from the collision also exhibits an initial movement (arrow 62) opposite the collision direction 62.

[0040] However, the upfront sensor 54 facing the collision will perform no or only a very slight movement against the collision direction at the beginning of the collision, which will quickly change into the collision direction 62 due to the intrusion of the collision on the collision side.

[0041] A frontal collision with full overlap thus leads to a movement of both upfront sensors 14, 16 opposite the collision direction 22 at the beginning of the collision (Figure 1). To determine the initial movement opposite the collision direction 22, a combination of upfront signals from both sensors 14, 16 or features calculated from them can be used.

[0042] Possible features include a low-pass filtered signal, a single or multiple window integral, or a single or multiple integrated signal. The combination can be based either on the signals themselves, on signal features, or based on the results of an evaluation of the signal features of the individual sensors. The evaluation can include a comparison of the signals, the signal features, the results of an evaluation of the signal features, or different combinations of these variables using one or more thresholds.

[0043] Suitable signal combinations include the sum or maximum of upfront signals or features calculated from them, see Figure 3. It should be noted that the signed maximum in the initial counter-movement phase is given by the signal of the collision side with the absolutely smaller signal, ie an implementation as a minimum is also possible.

[0044] Figure 3 shows examples of sensor signals or features and their combinations, including possible threshold values. A first graph 100, with time plotted on the abscissa 102 and an acceleration or a value of a feature plotted on the ordinate 104, shows curves for a collision with full coverage (see the left side of Figure 3), for the left sensor 14 (see Figure 1) by curve 106 and for the right sensor 16 (see Figure 1) by curve 108.

[0045] A second graph 120, on whose abscissa 122 the time is plotted and on whose ordinate 124 an acceleration or a value of a feature is plotted, shows curves in the case of a collision with offset, see right side of Figure 3, for the left sensor 54 (see Figure 2) by curve 126 and for the right sensor 56 (see Figure 2) by curve 128.

[0046] A third graph 140, with time plotted on the abscissa 142 and a combination maximum on the ordinate 144, shows a curve 146 representing the course of the combination maximum in a collision with full overlap. A possible threshold value 150 for detection is also plotted.

[0047] A fourth graph 160, with time plotted on the abscissa 162 and a combination maximum on the ordinate 164, shows a curve 166 representing the course of the combination maximum in the event of a collision with an offset. A possible threshold 170 for detection is also plotted.

[0048] A fifth graph 180, with time plotted on the abscissa 182 and a combination sum plotted on the ordinate 184, shows the course of the combination sum for a collision with full overlap using a curve 186. Furthermore, a possible threshold value 188 for detection is entered.

[0049] A sixth graph 190, on whose abscissa 192 the time is plotted and on whose ordinate 194 a combination sum is plotted, shows the course of the combination sum for a collision with offset with a curve 196.

[0050] Furthermore, a possible threshold value of 198 is entered for detection. The combined feature is compared against a negative threshold. If this value is exceeded, the initial countermovement phase of a collision with full overlap is detected.

[0051] The combination can be based on the signals or on the results of the separate evaluation of the signal characteristics of both sensors.

[0052] For example, an initial counter-movement phase of a collision with full overlap is detected when the feature of the right sensor falls below a negative threshold and, in addition, the feature of the left sensor falls below a negative threshold.

[0053] The determined initial movement against the collision direction can therefore be used to influence the triggering decision or the existing offset detection.

[0054] Possible ways to influence the trigger decision include adjusting, for example, sensitizing, one or more trigger decision thresholds, or logically combining the detected initial movement opposite to the collision direction with other trigger decision features. These also include features based on a central x-sensor and thus ensure the presence of the correct collision direction, e.g., a frontal collision direction.

[0055] Possible influences for the offset detection are an adaptation, e.g. a robustness, of one or more threshold values ​​of the existing offset detection or a logical combination of the detected initial movement against the collision direction with other features of the existing offset detection.

[0056] Figure 4 shows a flowchart of one possible sequence of the described method. In a first step 200, a collision occurs while a motor vehicle is traveling. The effects of the collision are recorded by two mechanically coupled acceleration sensors, and corresponding signals are generated and forwarded to an arrangement of the type described herein. In the arrangement or in an evaluation unit within the arrangement, the received signals are evaluated in a next step 202. Based on this evaluation, a signal is output in a next step 204 to a unit for controlling at least one restraint device, so that an appropriate response is made to the collision.

[0057] Figure 5 shows a vehicle 250 in which two acceleration sensors 252 and 254 are provided. In the event of a collision, these acceleration sensors 252, 254 each provide a signal 256 or 258, which is forwarded to an arrangement 260 for carrying out the method. An evaluation unit 262 is provided in this arrangement 260. This evaluation unit 262 is implemented in hardware and / or software and thus provides means for carrying out the method. Based on the evaluation, the evaluation unit 260 generates a signal which is transmitted to a unit 264 for controlling two restraint devices 266, 268. The restraint devices 266, 268, which are designed, for example, as airbags or belt tensioners, then trigger a reaction to the collision.

Claims

Claims 1 . Method for evaluating signals (256, 258) from at least two acceleration sensors (252, 254) in a vehicle (10, 50, 250) during a collision, wherein the acceleration sensors (252, 254) are mechanically coupled to one another, the signals (256, 258) of the acceleration sensors (252, 254) which describe an initial movement of the acceleration sensors (252, 254) at the beginning of the collision against a collision direction (22) are evaluated.

2. Method according to claim 1, which is used to determine a collision type.

3. Method according to claim 1 or 2, which is used to influence a decision to trigger restraint means (266, 268).

4. Method according to one of claims 1 to 3, wherein at least one restraining means (266, 268) is activated as a function of the evaluation.

5. Method according to one of claims 1 to 4, wherein the signals (256, 258) are evaluated within a period of 10 to 15 ms after the start of the collision.

6. Method according to one of claims 1 to 5, in which the signals (256, 258) from two upfront sensors (14, 16, 54, 56) are evaluated, which are coupled to one another via a bending beam (18, 58).

7. Method according to one of claims 1 to 6, in which features of the signals (256, 258) are first calculated from the signals (256, 258).

8. The method of claim 7, wherein features are calculated by a measure selected from a group consisting of: filtering with a low-pass filter, calculating a simple or multiple window integral, simple or multiple integration.

9. The method according to any one of claims 1 to 8, wherein a combination of the signals (256, 258) of the at least two acceleration sensors (252, 254) or a combination of the features calculated from the signals (256, 258) is evaluated.

10. The method according to claim 9, wherein the combination results from a measure selected from a group consisting of: forming a sum, determining a minimum and determining a maximum.

11. Method according to one of claims 1 to 10, wherein a comparison with at least one threshold value is carried out during the evaluation.

12. Arrangement for evaluating signals from at least two acceleration sensors (252, 254) in a vehicle (10, 50, 250) in the event of a collision, wherein the arrangement (260) has an evaluation unit (262) which is configured to carry out a method according to one of claims 1 to 11.

13. Arrangement according to claim 12, which is associated with a unit (264) for controlling at least one restraining means (266, 268).

Images