METHOD FOR CONTROLLING A PERSONAL PROTECTION DEVICE

Description

State of the art

[0001] The invention relates to a method and a computing unit for controlling a personal protection device.

[0002] DE 10 2011 118 149 A1 discloses a method for operating a safety system of a motor vehicle. In this method, a current traffic situation is classified. Furthermore, the probability of an accident type occurring from a plurality of predetermined accident types in the classified current traffic situation is determined. If the determined probability exceeds a predetermined threshold, a trigger threshold for activating at least one occupant protection device is adjusted.

[0003] The present invention is intended to enable an adaptation of the trigger threshold depending on the road surface being traveled on. Disclosure of the invention

[0004] The invention describes a method for controlling a personal protection device for a vehicle. The personal protection device is controlled when a crash is detected by a crash detection function, in particular as a function of at least one signal from a crash sensor. The crash detection function is sensitized as a function of a signal from an environmental sensor system. This sensitization also occurs as a function of a state of a surface traveled by the vehicle, in particular a road (original page 6). To determine a state of a surface traveled by a vehicle, in particular a road, a state value representing this state is determined. The vehicle has at least one inertial sensor.The method is characterized in that the state value is increased depending on at least one first signal detected by the inertial sensor if the first signal reaches or exceeds a signal threshold and / or the state value is decreased if the first signal falls below the signal threshold. The crash detection function is not sensitized if a condition of a surface traveled by the vehicle, in particular a road, is assessed as disruptive to the crash detection function. The condition of a surface traveled by the vehicle is assessed as disruptive to the crash detection function if the condition value exceeds a condition threshold.

[0005] This embodiment of the method offers the advantage that the state value reflects a realistic state of the surface traveled by the vehicle, in particular a road. To prevent every small vehicle movement or vibration from being incorporated into the state value, the first signal must first reach a signal threshold before the state value is increased.

[0006] The signal threshold and / or multiple signal thresholds can be different, or defined / specified, depending on the vehicle and sensor. This also makes it possible to ensure optimal adaptation of the process for different vehicles.

[0007] This method can be carried out in a vehicle equipped with a corresponding sensor. Alternatively, the method can also be carried out on an outsourced computing unit, which, for example, receives and evaluates sensor values recorded in one or more vehicles. The transmission can take place wirelessly or via the Internet, for example. The information obtained about the road being traveled can then be shared with multiple vehicles, for example via a corresponding communication interface such as a cloud. In this way, even vehicles without a corresponding integrated function or without the necessary sensors receive information about the condition of a specific area, a specific road, or a specific road section.

[0008] Once determined, the status value can be made available to one or more additional functions. The additional function can, for example, be implemented in the vehicle and designed to trigger safety devices and / or a personal protection system and / or restraint devices. It is also conceivable that the function is intended to control a chassis, such as a damper or air suspension.

[0009] The condition of a surface, especially a road, can be understood, for example, as whether the surface has a good or poor surface. For example, whether the surface has a lot of potholes, a lot of rubble or stones lying around, or whether the surface has a very smooth surface. Of particular importance here is the effect of the condition of the surface and / or the road on inertial sensors and / or crash sensors in a vehicle.

[0010] An inertial sensor can be understood, for example, as a rotation rate sensor or acceleration sensor. Pressure sensors can also be used as crash sensors in addition to these sensor types. The first signal detected by at least one inertial sensor can therefore be understood as a signal representing a rotation rate and / or an acceleration value.

[0011] This method offers the advantage of being able to determine the condition of the surface a vehicle is traveling on very easily and cost-effectively. Furthermore, the determination is based on sensor measurements, which can also be used for crash detection. Consequently, the provision of the obtained information creates synergistic effects, which in turn contribute to increasing the safety of road users.

[0012] In a further embodiment of the method, the state value is increased by a predefined penalty value if the first signal exceeds a signal threshold.

[0013] This embodiment offers the advantage that severe unevenness is given greater weight, thus reducing the likelihood of misinterpreting crash signals. Furthermore, by increasing the condition value by a penalty when the signal threshold is exceeded, the number of acceleration peaks and consequently the frequency of vibrations are incorporated into the condition value. This is also an important metric for determining the condition of a surface traveled by the vehicle.

[0014] Exceeding the signal threshold means that the signal was previously below the signal threshold and after exceeding it is above the signal threshold.

[0015] In a further embodiment of the method, the penalty value is increased depending on a determined and / or detected amplitude of the first signal.

[0016] This embodiment of the method offers the advantage that, when determining the condition value of a surface, unevenness of varying severity is also weighted differently. This makes it possible to obtain an exact measure of the actual condition of the surface using the condition value. In this embodiment, the penalty value can be added to the condition value slightly offset from the time of the violation, so that the peak amplitude can be determined first. Alternatively or additionally, the peak amplitude can also influence one or more subsequent penalty values.

[0017] In a further embodiment of the method, the state value is continuously increased as long as the first signal exceeds a signal threshold, wherein the state value is increased in particular linearly.

[0018] The state value is continuously increased as long as the first signal is above a signal threshold, unlike the previously mentioned penalty value, by which the state value is only increased once when a signal threshold is exceeded.

[0019] This embodiment offers the advantage that the time period during which the first signal exceeds the signal threshold is taken into account when determining the condition value, or rather, is incorporated into it. This time period is also a measure of the condition of a surface being driven on. In particular, longer journeys on very bumpy surfaces are given greater weight when determining the condition value.

[0020] In a further embodiment of the method, the state value is continuously decreased as long as the first signal falls below a signal threshold. In particular, the state value is decreased linearly.

[0021] This design offers the advantage that the condition value can decrease again when the surface, especially the road, is in good condition. This, in turn, improves the quality of the condition value. One-off or rare events, such as driving over a sidewalk, are thus not given too much weight, thus avoiding misinterpretations of the condition of the surface being driven on.

[0022] In a further embodiment of the method, the state value is increased over time. The magnitude of the increase, i.e., how quickly the state value increases per unit of time, depends on a determined and / or recorded amplitude of the first signal.

[0023] This design allows for more severe unevenness to be incorporated into the condition value over a longer period of time, thus giving them a higher weighting. This offers another option for adjusting parameters so that the ultimately determined condition value represents the most accurate measure of the surface's actual condition.

[0024] In a further embodiment of the method, the state value increases more when the first signal exceeds the signal threshold per time interval than it decreases in the same time interval when the first signal falls below the time threshold.

[0025] This embodiment does not refer to one-off increases or decreases in the condition value, but rather to continuous increases or decreases in the condition value over time. This embodiment in turn makes it possible to use the condition value to provide an optimal representation of the actual surface or road condition. This embodiment can be used in addition to or as an alternative to increasing the condition value by a penalty value when a signal threshold is exceeded. If no penalty value is added, the faster increase in the condition value when the first signal exceeds the signal threshold can ensure that times in which significant vibration is measured are given greater weight than times without vibration, or times during which the first signal falls below the signal threshold.

[0026] A personal protection device can be understood as any device designed to protect a person involved in an accident. This includes, for example, airbags for vehicle occupants, seat belt pretensioners, airbags for people outside the vehicle, and other safety devices, such as mechanisms for raising a hood to protect pedestrians.

[0027] A crash detection function is understood to be a function that determines whether a crash has occurred, particularly based on at least one signal from a crash sensor. Corresponding crash detection functions are known from the prior art.

[0028] Sensitization of the crash detection function means that a crash is detected more quickly or earlier, or that a corresponding signal is output by the crash detection function. This can be understood, for example, as lowering a crash threshold. Sensitization can also be understood as modified processing of a signal detected by a crash sensor. By processing at least one signal from a crash sensor, a personal protection device can be triggered earlier under the same boundary conditions (e.g. accelerations, forces acting on the vehicle, etc.) than without the processing. For example, a multiplication factor or a value added to or subtracted from the signal can change. A comparable result can also be achieved by lowering a crash threshold.

[0029] The condition of an area traveled over by the vehicle, in particular a road, can be received and / or determined in a variety of ways. Either a determination can take place in the vehicle or data can be received which contains information about the condition of the area traveled over by the vehicle. This can be determined, for example, using the method described above. Conditions of areas, in particular roads, can also be stored on a digital map, which can be retrieved using this method. Alternatively, information regarding the road condition can also be received from other vehicles or a central location. For example, several vehicles could send information about the road condition to a cloud, which then makes it available to other vehicles. The data can also be evaluated and processed in the cloud.For this purpose, it is sufficient, for example, to transmit only the vehicle sensor data to the cloud or an external server. A meaningful procedure for determining a status value representing the condition of the area being driven on could be implemented accordingly on the cloud or the external server.

[0030] The method presented here offers the advantage of reducing misinterpretations of crash signals, such as the false activation of personal safety devices, especially when the vehicles are on surfaces or roads whose condition negatively impacts crash sensing. This is especially true for bumpy roads with potholes and / or rocks.

[0031] In a further embodiment of the method, a crash is determined when a crash signal determined on the basis of at least one signal detected by a crash sensor exceeds a crash threshold value.

[0032] In general, the signs can be changed when detecting signals, so that an overshoot becomes a undershoot. Such modifications to the method naturally represent possible alternatives, which are also intended to fall within the scope of this application.

[0033] In an advantageous embodiment of the invention, the crash detection function is sensitized by adjusting the crash threshold value, in particular by reducing it.

[0034] This embodiment of the invention offers the advantage that a decision as to whether a personal protection device should be triggered can be made earlier. Consequently, it is possible to react more quickly to an impending collision and thus ensure optimized activation of the personal protection device. This increases the safety of all road users.

[0035] In a further embodiment of the method, the crash detection is sensitized depending on at least one environmental signal if an impending collision is detected based on the at least one environmental signal.

[0036] For example, an impending collision can be detected by determining a collision probability based on at least one environmental signal. For this purpose, variables that can be obtained by evaluating the environmental signal can be incorporated, if necessary. For example, the direction of travel of a crash object, for example another vehicle, the vehicle's own direction of travel, the speed of various objects, dimensions of objects, etc. Objects can be understood to mean all objects and / or potential collision objects that can be detected by the environmental sensors. Further information about the ego vehicle can also be incorporated, such as the speed, the steering angle or further evaluations from other control units. For example, information could be used regarding the current braking or steering pressure and how quickly a driving maneuver can be carried out.By combining all the information, it is possible to determine whether a collision is imminent or the probability of one. For example, a probability threshold can be defined, above which a collision is assumed to be imminent. If this threshold is exceeded, a collision is detected.

[0037] In an advantageous embodiment of the invention, the state of an area traveled over by the vehicle, in particular a road, is determined using the method described above for determining a state value representing a state of an area traveled over by a vehicle, in particular a road.

[0038] Instead of assessing the condition of an area traveled by a vehicle based on a condition value, other methods can also be considered. For example, information about the condition of a road can be obtained from a digital map. Alternatively, this information can be retrieved from a cloud, for example. With these alternative methods of recording road conditions, a condition value does not necessarily have to be recorded and / or evaluated. The information can be transmitted in any way, for example, using a zero for a "good" and a one for a "poor" road condition.

[0039] This embodiment offers the advantage that misinterpretations of crash signals can be reduced, especially when the vehicle is on surfaces or roads that are in poor condition.

[0040] In cases where the condition threshold is not exceeded, sensitization can occur. For sensitization, for example, a crash threshold can be lowered. Multiple condition thresholds can also be defined, and if these thresholds are exceeded or undershot, defined sensitizations are applied, for example, gradual adjustments of the crash threshold or multiple crash thresholds.

[0041] In a further embodiment of the method, the crash threshold is increased if a condition of a surface traveled by the vehicle, in particular a road, is assessed as disruptive to the crash detection function. The condition of a surface traveled by the vehicle is assessed as disruptive in particular if the condition value exceeds a condition threshold.

[0042] This embodiment of the method offers the advantage that misinterpretations of crash signals can be reduced. If, for example, a vehicle is traveling on a very bumpy road, it may happen in unfavorable cases that a situation is misinterpreted by a crash detection function because a crash threshold is exceeded due to the high vibration. By increasing the crash threshold when a surface to be driven over by the vehicle is detected to be in a poor condition or to be disruptive to the crash detection function, such cases can be avoided. This can further increase the safety of all road users. Further advantages arise from the use of a computer program stored on a machine-readable storage medium, which causes a computing unit to carry out a method according to the above explanations when executed on the computing unit.

[0043] A further advantageous embodiment of the invention manifests itself in the computing unit, in particular in a controller for an electrical machine, which is configured by the computer program provided on the computing unit, in particular on a storage medium of the computing unit, and / or by a corresponding integrated circuit to carry out at least one method according to the above statements. This results in synergies, since the computing unit, in particular the controller, not only serves to control the electrical machine, but is also configured to carry out the method according to the invention. The computing unit can, for example, be designed to control an airbag or other personal protection device depending on the adapted crash detection function. Drawings

[0044] Figure 1 shows a schematic process flow. Figure 2shows an example determination of the state value. Figure 3 shows a process diagram. Figure 4 shows a process diagram for triggering a personal protection device. Figure 5 shows another process diagram. Example

[0045] For reliable accident detection, in addition to crash sensors installed in a vehicle, environmental sensors can also be used. For the best possible reliability of crash determination, both pieces of information can be used together. For example, if an evaluation of signals from environmental sensors determines that a collision with an object is imminent, a crash detection function that is based on an evaluation of the crash sensor signals can be sensitized. This means that, in the event of an actual crash, restraint devices or personal protective equipment can be deployed more quickly. However, if the vehicle is traveling on poor terrain, for example, on a very bumpy road, crash sensitization can have negative effects. This problem is solved by the present invention and the following exemplary embodiments.

[0046] In Figure 1 A method for determining a state value representing the condition of a surface traveled by a vehicle is presented. The state value can be determined for all conceivable surfaces on which a vehicle can travel, such as roads, dirt roads, parking lots, deserts, beaches, and / or ice surfaces. For this purpose, the vehicle has at least one inertial sensor, in this example an acceleration sensor. Furthermore, the vehicle includes a computing unit that can receive signals from the acceleration sensor and on which the method described below can be executed. The method starts in step 101.

[0047] In step 102, a first signal is detected by the inertial sensor or the acceleration sensor, which in this embodiment corresponds to an acceleration signal.

[0048] In step 103, the state value representing the state of an area traveled by the vehicle is determined based on the detected acceleration signal. In this example, this is done on the processing unit, which can receive and evaluate the acceleration signals detected by the acceleration sensor. Depending on the first signal detected by the acceleration sensor, the state value is increased or decreased.

[0049] The process ends in step 104. For continuous monitoring or detection of an area traveled by the vehicle, it is advisable to also carry out the process continuously. Therefore, the Figure 1 The process shown is carried out continuously as a loop, so that the method starts again at step 101 after reaching step 104. In this way, the state value changes continuously depending on the state of the area being driven on.

[0050] In Figure 2is shown in detail how a possible calculation of the state value 201 can be carried out. The course of various variables over time t is shown. These variables are the state value 201, the first signal 202, which in this embodiment corresponds to an acceleration signal 202, an event variable 203 and a misuse variable 204. The event variable 203 is always set to one when the acceleration signal 202 exceeds a predefined signal threshold 222. Otherwise, the event variable 203 is zero. The misuse variable 204 remains set to zero as long as the state value 201 does not exceed a state threshold 211. If the state value 201 exceeds the state threshold 211, the misuse variable 204 is set to one. This misuse variable 204 can, for example, be used to communicate a road condition to another function.If the misuse factor 204 is zero, it can be assumed that the road condition has no adverse influence on, for example, measured values from acceleration sensors installed in a vehicle. In contrast, if the misuse factor 204 is one, it can be assumed that the road condition could negatively impact other measured values or evaluation methods, such as crash detection.

[0051] The following explains in more detail how the state value 201 changes depending on the first signal or acceleration signal 202. If the acceleration signal 202 exceeds the signal threshold 222, the state value 201 increases by a predefined penalty value. In this example, the penalty value is fixed and corresponds to the level of the state value 201 immediately after the acceleration value 202 is first exceeded by the signal threshold 222. The state value 201 is always increased by the penalty value when the acceleration signal 202 exceeds the signal threshold 222 from below. This means that the acceleration signal 202 is initially smaller than the signal threshold 222 and subsequently equal to or greater than the signal threshold 222.

[0052] As long as the acceleration signal 202 exceeds the signal threshold 222, the event size 202 is set to one. During this time, the state value 201 increases linearly with a fixed gradient.

[0053] If the acceleration signal 202 falls below the signal threshold 222 again, the event variable 203 is reset to zero. During these periods, the state value 201 decreases. In this exemplary embodiment, the decrease occurs linearly and with the same gradient, but with a reversed sign, as with the increase in the state value 201 when the acceleration signal 202 exceeds the signal threshold 222. If the state value 201 reaches the value zero, it does not decrease further but remains at zero until the acceleration signal 202 again exceeds the signal threshold 222.

[0054] The misuse quantity 204 is set to one precisely when the state value 201 exceeds a state threshold 211. In this exemplary embodiment, the time t 0 205 is reached after a certain time. The misuse quantity 204 remains set to one until the state value 201 falls below the state threshold 211 again.

[0055] In an alternative embodiment, the misuse value 204, after being set to one, i.e., after the state value 201 has exceeded the state threshold 211, remains set to one for a certain, predefined period of time. The misuse value 204 remains set to one during this period even if the state value 201 has fallen below the state threshold 211 again. The period can be set to several minutes, for example. In this exemplary embodiment, it is set to 10 minutes. This embodiment ensures that a reliable determination of the road condition is possible even when driving on roads whose condition changes frequently.

[0056] In an alternative embodiment of the method, the gradients of the state value differ when the acceleration signal 202 exceeds or falls below the signal threshold 222. This means that the state value 201, for example, when the acceleration signal 202 exceeds the signal threshold 222, increases more per time interval than it decreases in the same time interval when the acceleration signal 202 falls below the signal threshold 222. In this embodiment, the state value is not increased by a penalty value when the first signal exceeds the signal threshold.

[0057] In a further embodiment, the state value 201 increases non-linearly when the acceleration signal 202 exceeds the signal threshold 222, but exponentially. In this case, the longer the acceleration signal 202 exceeds the signal threshold 222, the faster the state value 201 increases per unit of time. A non-linear progression can also be provided for a decrease in the state value 201 when the acceleration signal 202 falls below the signal threshold 222. In this embodiment, however, the state value also decreases linearly.

[0058] In a further embodiment, the amplitude of the acceleration signal 202 is also evaluated and is incorporated into the level of the penalty value and the degree of increase in the state value 201. For example, the higher the amplitude of the acceleration signal 202, the greater the gradient can be selected with which the state value 201 is increased per unit of time, as long as the acceleration signal 202 exceeds the signal threshold 222.

[0059] In an alternative embodiment of the method, when the acceleration signal 202 exceeds the signal threshold 222, the state value 201 is not increased by a penalty value. Instead, an increase only occurs during the time periods in which the event size 203 is set to one, and the acceleration signal 202 consequently exceeds the signal threshold 222.

[0060] In Figure 3Another schematic process flow is shown. This starts in step 301. Steps 302 and 303 correspond to steps 102 and 103, respectively.

[0061] In step 302, a signal from an initial sensor is detected. In step 303, a state value 201 is increased or decreased based on this signal. This increase or decrease of the state value 201 takes place in this embodiment according to the first Figure 2 described procedures.

[0062] In step 304, it is determined whether the state value 201 exceeds a state threshold 211. If the state threshold 211 is not exceeded by the state value 201, the method returns to step 302, in which a signal from an initial sensor is again detected and the state value 201 is subsequently updated in step 303.

[0063] However, if the state threshold 211 is exceeded by the state value 201, this information is passed on to a crash detection function. This crash detection function is designed to detect a crash. For this purpose, at least one signal from a crash sensor is read in. In this exemplary embodiment, several acceleration signals are read in. From these acceleration signals, a crash signal is determined by the crash detection function. If this crash signal exceeds a predefined crash threshold, the crash detection function detects a crash. This information can then be used, for example, to trigger a personal protection device. In this exemplary embodiment, a driver's airbag is triggered as soon as the crash signal exceeds the crash threshold.

[0064] If it is now determined in step 304 that the state value 201 exceeds a state threshold value 211, this information is passed on to the crash detection function in step 305.

[0065] The method ends in step 306.

[0066] In an alternative embodiment of the method, the state value 201 can also be continuously transmitted to the crash detection function. If a predefined state threshold 211 is exceeded, this information can be used by the crash detection function, for example, to prevent sensitization of a crash detection. This can be useful, for example, if the crash detection function evaluates data from environmental sensors in addition to the acceleration values or if this data is incorporated into the crash detection. It is conceivable, for example, that a crash detection function is sensitized, for example, a crash threshold is lowered, if the signals from environmental sensors determine that a crash is imminent.If a vehicle is simultaneously on a road that is in poor condition and could therefore influence the crash detection function, for example by measuring high acceleration values that are due to poor road conditions but not to an accident, such sensitization of the crash detection function can lead to misinterpretations of the situation.

[0067] In an alternative embodiment, the crash detection function is not transmitted with the state value 201, but with a misuse value 204, as determined, for example, in the method presented above.

[0068] In Figure 4 A further method is described in which several of the methods described above are combined. The method starts in step 401.

[0069] In step 402, at least one signal from an inertial sensor is detected. This signal is evaluated in step 403, and a state value 201 is determined based on the signal. This can, for example, correspond to the first Figure 2 described procedure.

[0070] In step 404, the state value 201 is compared with a state threshold 211, and a misuse value 204 is generated. This is provided to the function executed in step 444 for determining a crash threshold.

[0071] In step 455, signals from at least one environment sensor are detected, in this embodiment a camera. These signals are evaluated in step 405. Objects in the vehicle's surroundings are detected, and it is determined whether a collision with one of these objects is imminent. In this embodiment, a probability of such a collision is provided to the function executed in step 444 for determining a crash threshold.

[0072] In step 444, a crash threshold is determined. This is initially fixed, but can be sensitized if an impending collision is detected. Whether a collision is imminent is determined based on the collision probability determined in step 405. In this exemplary embodiment, sensitization corresponds to a reduction of the crash threshold, so that a crash is assumed even at a lower crash signal.

[0073] In this exemplary embodiment, whether such a sensitization of the crash threshold is actually performed additionally depends on the misuse variable 204 determined in step 404. If the misuse variable 204 has the value zero, the crash threshold can be sensitized based on a collision predicted in step 405. However, if the misuse variable 204 has the value one, sensitization of the crash threshold is suppressed, and consequently, the crash threshold is not lowered.

[0074] In step 406, a crash detection occurs, with a crash detection function determining whether a crash has occurred and whether a personal protection device should therefore be triggered. For this evaluation, the crash detection function receives signals from crash sensors—in this embodiment, several pressure and acceleration sensors—which were detected in step 466. Furthermore, the crash threshold value determined in step 444 is transmitted to the crash detection function.

[0075] In this exemplary embodiment, the crash detection function determines a crash signal based on the signals detected by the crash sensors. If the determined crash signal exceeds the crash threshold determined in step 444, a passenger protection device is triggered in step 407. In this exemplary embodiment, multiple belt tensioners, airbags, and pedestrian protection devices can be triggered for this purpose.

[0076] If the predefined crash threshold is not exceeded, the method jumps to step 408, where it ends, and no personal protection device is triggered. After completing the method and reaching step 408, the method starts again in step 401.

[0077] In an alternative embodiment of this method, the crash threshold can also be increased in step 444 when determining the crash threshold. If the misuse factor 204 is equal to one, it can be assumed that the road condition influences crash detection. To avoid misinterpretation of a crash situation, the crash threshold can be increased in this situation.

[0078] In an alternative variant, both pieces of information are evaluated together and weighed against each other so that the crash threshold is optimally adjusted.

[0079] In a further embodiment of the method, the misuse quantity 204 is not provided in step 404, but only the state value 201 is transmitted. Consequently, in step 404, a more detailed evaluation of the currently available state value 201 can be performed and, if necessary, a more sensitive adjustment of the crash threshold value can be performed.

[0080] In another embodiment, a vehicle has a crash detection function executed on a processing unit. Furthermore, the vehicle has acceleration sensors for detecting an accident and environmental sensors in the form of a camera, a lidar, and a radar. Furthermore, the vehicle is equipped with a receiving device that can receive statuses from different road sections.

[0081] The crash detection function evaluates the signals from both the surrounding sensors and the crash sensors. If the surrounding sensors detect an impending collision, the crash detection function is sensitized, so that restraint devices are triggered earlier based on the signals from the crash sensors than would be the case under the same conditions (identical accelerations acting on the vehicle) without sensitization of the crash detection function.

[0082] However, the crash detection function is also capable of suppressing sensitization based on the evaluation of the surrounding sensors. This suppression occurs based on the condition of the surface the vehicle is driving over. Since there is no integrated process in the vehicle that can evaluate the condition of the surface currently being driven over, information about the condition of the surface is received. This is done via the receiver device installed in the vehicle.

[0083] The road conditions are stored on an external server, from which they are sent to and received by the vehicle. The information stored on the server can, for example, be collected information from multiple vehicles that perform functions to determine the condition of a road surface and transmit the determined conditions to this server.

[0084] In an alternative embodiment, the conditions of different road sections are already stored in the vehicle, for example, on a digital map. These conditions can be read by the crash detection function and used to suppress sensitization of crash detection.

[0085] In an alternative embodiment, signals about a road condition are sent from other vehicles via a car-to-car communication interface and received by the vehicle.

[0086] In Figure 5 The basic method for controlling a personal protection device for a vehicle is shown, which starts in step 501.

[0087] In step 502, a crash detection function is used to determine whether a crash has occurred. This determination is made based on at least one signal from a crash sensor 521. In this exemplary embodiment, this signal 521 is detected by an acceleration sensor.

[0088] In addition, the determination of whether a crash has occurred is based on a signal from an environment sensor system 522. In this exemplary embodiment, this signal 522 is detected by a camera. The signal detected by the environment sensor system serves to sensitize the crash detection function.

[0089] Furthermore, the determination of whether a crash has occurred is made depending on a condition 523 of a surface traveled by the vehicle. The condition of the road currently traveled by the vehicle can be determined in different ways. On the one hand, it is possible to determine the condition using acceleration sensors in the vehicle, for example, the existing crash sensors. For this purpose, the Figure 2 The method shown in FIG. 1 is used. In this exemplary embodiment, a signal representing this state 523 is received from a cloud on which current data about various route sections is stored.

[0090] If the crash detection function determines in step 502 that a crash has occurred, a personal protection device is triggered in step 503. Which protective device is triggered, after what time, and with what intensity depend on the result of the calculations of the crash detection function. The crash detection function is capable, for example, of detecting different crash severities. It can also determine on which side a crash occurs and whether it is a collision with a vehicle, an object, such as a tree, or a person. The method ends in step 504.

[0091] If the crash detection function determines in step 502 that no crash has occurred, no personal protection device is triggered. Essential to this determination and this invention is the inclusion of the condition 523 of the surface traveled by the vehicle.

Claims

1. Method for actuating a personal protection device for a vehicle, in particular an airbag, the personal protection device being actuated if a crash is detected by means of a crash detection function on the basis of at least one signal from a crash sensor, the crash detection function being sensitized on the basis of a signal from an environment sensor system, the sensitization being effected on the basis of a condition of a surface used by the vehicle, wherein • the condition of the surface used by the vehicle is determined by determining a condition value (201) representing that condition, and • the condition value (201), on the basis of at least one first signal (202) acquired by at least one inertial sensor of the vehicle • is increased if the first signal (202) reaches or exceeds a signal threshold value (222) and / or • is reduced if the first signal (202) falls below the signal threshold value (222), and • the crash detection function is not sensitized if a condition of a surface used by the vehicle, in particular of a road, is assessed as disruptive to the crash detection function, the condition being assessed as disruptive if the condition value (201) exceeds a condition threshold value (211).

2. Method for actuating a personal protection device for a vehicle, in particular an airbag, the personal protection device being actuated if a crash is detected by means of a crash detection function on the basis of at least one signal from a crash sensor, the crash detection function being sensitized on the basis of a signal from an environment sensor system, the sensitization being effected on the basis of a condition of a surface used by the vehicle, wherein • the condition of the surface used by the vehicle is determined by determining a condition value (201) representing that condition, and • the condition value (201), on the basis of at least one first signal (202) acquired by at least one inertial sensor of the vehicle • is increased if the first signal (202) reaches or exceeds a signal threshold value (222) and / or • is reduced if the first signal (202) falls below the signal threshold value (222), and • the crash threshold value is increased if a condition of a surface used by the vehicle, in particular of a road, is assessed as disruptive to the crash detection function, the condition being assessed as disruptive if the condition value (201) exceeds a condition threshold value (211).

3. Method according to Claim 1 or 2, characterized in that when the condition threshold value (211) is exceeded, a misuse variable (204) that (page 14, lines 5 ff.) indicates a road condition that has an adverse effect on crash detection is set to one.

4. Method according to Claim 3, characterized in that the misuse variable (204) remains set to one for a predefined period of time, even if the condition value (201) has fallen below the condition threshold value (211) again.

5. Method according to one of the preceding claims, characterized in that a crash is detected if a crash signal that has been determined - on the basis of at least one signal acquired by a crash sensor - exceeds a crash threshold value.

6. Method according to one of the preceding claims, characterized in that the crash detection function is sensitized by adjusting, in particular lowering, the crash threshold value.

7. Method according to one of the preceding claims, characterized in that the crash detection is sensitized on the basis of at least one environment signal if an impending collision is detected on the basis of the at least one environment signal.

8. Method according to one of the preceding claims, characterized in that the condition value (201) is increased by a predefined penalty value if the first signal (202) exceeds a signal threshold value (222).

9. Method according to Claim 8, characterized in that the penalty value is increased on the basis of a determined and / or recorded amplitude of the first signal (202).

10. Method according to one of the preceding claims, characterized in that the condition value (201) is continuously increased while the first signal (202) exceeds a signal threshold value (222), the condition value (201) being increased in particular linearly.

11. Method according to one of the preceding claims, characterized in that the condition value (201) is continuously reduced while the first signal (202) falls below a signal threshold value (222), the condition value (201) being reduced in particular linearly.

12. Method according to one of the preceding claims, characterized in that the condition value (201) is increased over time, a sharpness of the increase being dependent on a determined and / or recorded amplitude of the first signal (202).

13. Method according to one of the preceding claims, characterized in that the condition value (201) increases more sharply per time interval when the first signal (202) exceeds the signal threshold value (222) than it decreases in the same time interval when the first signal (202) falls below the signal threshold value (222).

14. Method according to Claim 13, characterized in that multiple condition threshold values are defined that, when underrun or exceeded, result in defined sensitizations being effected by means of gradual adjustment of the crash threshold value.

15. Computing unit for actuating a personal protection device for a vehicle, in particular an airbag, which is configured by an applicable integrated circuit and / or by a computer program stored on a storage medium to carry out a method according to one of Claims 1 to 14.