Method and device in a motor vehicle for pedestrian protection

Abstract

Method for setting at least one triggering criterion (A1) of a protection system (2) in a motor vehicle (1) having at least one environment sensor (3), comprising the following steps: a) Specification of the trigger criterion (A1) for a standard driving situation, b) detecting at least one detection area (S) with the at least one environment sensor (3) of the motor vehicle (1), characterized in that the method additionally comprises the following steps: c) determining whether at least one obstacle (22) restricting the detection by the environment sensor (3) is present in the detection area (S), d) Calculating at least one reduced detection range (SR) of the environment sensor (3) for detecting potential collision objects (8) as a function of the at least one detected obstacle (22). e) calculating whether collision objects (8) can be hit by the motor vehicle (1) at a predeterminable maximum speed (V) without having previously been detected by the environment sensor (3) in the reduced detection range (SR), f) Changing the triggering criterion (A1) to a changed triggering criterion (A2) if step e) shows that collision objects (8) can be hit by the motor vehicle (1) without having previously been detected by the environment sensor (3) in the reduced detection range (SR).

Description

Prior ArtThe present invention relates to a method for setting at least one triggering criterion of a protection system, in particular for the protection of external road users who are important for protection, in particular of a pedestrian protection system, for use in a motor vehicle, and to a corresponding device for carrying out the method.In the event of an accident of a motor vehicle, reversible and non-reversible restraint systems are intended to protect the driver from severe consequences. There are various systems for alleviating accident consequences for the driver, which can be divided into sensors and actuators.Passive safety actuators in the vehicle interior include, for example, active seats, belt tensioners or airbags. There are various embodiments of airbags, for example, a driver's airbag from the steering wheel, knee airbags for protecting knees during forward displacement and from slipping under the belt, window airbags for protecting the head during lateral impact and for preventing objects from entering the passenger compartment from the outside. Active seats can change shape in the event of an accident and thus, for example, prevent slipping under the belt or bring the driver into a more favorable position (e.g. drive to the rear, so that the driver gets more space to the steering wheel and thus the maximum accelerations for the driver can be reduced). Seatbelt pretensioners reduce the so-called belt slack and couple the driver to the vehicle. As a result, the forward displacement of the driver is reduced or the driver can be decelerated more uniformly with the vehicle.Collisions of a motor vehicle with a pedestrian can be mitigated by, for example, raising the engine hood and / or igniting additional external airbags for pedestrians in order to mitigate the impact severity of the pedestrian on structural parts of the vehicle, in particular the engine block or the A-pillars.Various sensors are used to determine an accident and the accident type. The main sensor is usually an acceleration sensor which is installed in the center of the vehicle in a manner as protected as possible. Such a main sensor may be sufficient for a simple detection of an accident, but is not as powerful and insensitive to errors as a multi-sensor system.Pedestrian accidents are frequently detected in such a way that additional contact sensors, e.g. acceleration sensors, are mounted in the front area of the engine hood of a vehicle in order to measure the weak acceleration values that a pedestrian causes in the vehicle as early as possible and to detect the accident in good time.Additionally or alternatively, a pressure tube sensor can be used, which consists of a silicone tube with usually two pressure sensors at the ends. The hose is installed behind the bumper. When the bumper is pushed in by the leg of a pedestrian, the pressure in the hose increases or a pressure wave is generated in the hose. The sensors detect the pressure increase and can determine the impact position of the pedestrian on the vehicle from the travel time difference of the pressure wave. Acceleration sensors at various locations in the vehicle may plausibilize the pressure sensor signal. Pressure sensors have the advantage over acceleration sensors that they can react very quickly, and the pressure hose at the vehicle front has the advantage that it can cover a large area with relatively few sensors (usually two pressure sensors). It is possible to determine the impact point to a precision of, for example, 5 cm.A further important group of sensors which are used to detect possible accident situations in a motor vehicle are so-called environment sensors which determine environment data from the environment of a motor vehicle and which thus monitor the environment of a motor vehicle and which serve for the detection and classification of detection and classification of possible collision partners.From the document DE 103 36 638 A1, a device is known which uses a surroundings sensor system to determine and classify the shape and dimensions of an object in a vehicle environment. In this case, the speed of the object can also be taken into account in the classification. Depending on this classification, protective measures can be activated.DE 10 2014 215 057 A1 discloses a method for determining a probability of a collision of a means of transportation with a pedestrian. In this case, the position, the head orientation and the direction of movement and / or speed of the pedestrian are determined.From the document DE 10 2011 118 658 A1, a method for determining the extent or size of a collision object and the collision speed as well as the collision coverage is known. Based on the information about the collision thus acquired, a targeted control of protection systems is carried out.DE 103 34 699 A1 discloses a device for actuating an actuator system for protecting pedestrians for a motor vehicle, in which a first signal of a contact sensor system or a threshold for comparison with a first signal of a contact sensor system is changed as a function of a second signal of a surroundings sensor system, the actuator system being actuated as a function of the comparison.Disclosure of the InventionDescribed here is to disclose a particularly advantageous method and a particularly advantageous device for setting at least one triggering criterion of a protection system for a road user.This is described by a method according to claim 1 and by a device according to claim 10. The dependent claims provide particularly advantageous developments of the method and the device on active protection systems in motor vehicles (e.g. stability control ESP or brake control) and comfort systems (e.g. lane keeping assist) are also networked to an ever greater extent with passive safety systems.Protection systems in motor vehicles often have different functional characteristics for different types of safety-relevant situations to which the motor vehicle may be exposed. Such feature embodiments are a collection of various security features that are triggered in a particular situation. There are frequently, for example, the functional characteristics "collision case" and "front collision", "side collision", "pedestrian collision", etc. Depending on the functional characteristic, protection systems are used in a motor vehicle in an adapted manner. The task of sensors and protection systems is to assign and select a present situation to the respective correct functional characteristic. In present situations, a plurality of functional characteristics can also be selected parallel to one another.In the functional embodiment "collision case", such a system (e.g. airbag deployment algorithm) is set, for example, on the basis of environment sensors. Environment sensors (mono / stereo camera, radar, lidar, ultrasound) capture the environment and ascertain a possible imminent collision and its collision type.An airbag control device can be set more sensitive in the event of an imminent collision, so that the restraint systems can be triggered more quickly.For example, a radar sensor can be used to predict a front collision with a vehicle and thus to select a corresponding function characteristic. During the predicted time at which the accident should take place, the activation threshold for restraint systems is reduced (i.e. more sensitive and earlier reaction) within the scope of this functional embodiment. If a possible accident is then registered by the classic passive safety sensors, it is possible to react to this more quickly or in a more time-directed manner, since the plausibility check duration (the time required by the system to check whether a collision is actually present with great probability) can be limited. Depending on the design level, a front collision or a side collision can be predicted, for example, using radar sensors, or else a reaction can be made to a rear collision. In the case of the collisions, it is possible to distinguish between different accident encounters, for example vehicle, truck, pedestrian or a fixedly anchored object and in each case a corresponding characteristic function can be selected.The functional expression "collision case" always has the prerequisite that an accident has already taken place. By means of systems operating in an anticipatory manner, only the reaction time is shortened, whereby the vehicle occupants can be better adjusted to the accident (create more space in order to reduce kinetic energy and thus avoid acceleration peaks). The basic functionality of collision sensing with acceleration sensors, etc., remains. The goal in the case of functional characteristics "Frot Collision" and "Side Collision" is generally a more rapid deployment of the restraint systems in the interior of the vehicle. In contrast, the functional expression "pedestrian collision" usually also aims to make the triggering of the pedestrian protection systems more robust (but certainly with a low rate of false triggerings), since the acceleration signal in the event of a collision with a pedestrian is very small and can be distinguished only poorly from other situations. In simulations, a so-called "leg impactor" (a special dummy leg) is used, which weighs only about 6 kg and must lead to triggering. A small animal, on the other hand, should not necessarily lead to triggering (bird strike, impact of small animals such as rabbits and the like). Mis-triggering is undesirable because pedestrian protection systems are often designed irreversibly and a mis-triggered protection system causes costs.Environment sensors for detecting environment data have only a limited detection range, or a detection range with a predefined shape. For example, a long-range radar (a typical environment sensor) has a detection range of below + / -10°. In contrast, a camera (a further typical environment sensor) has a detection range of + / - 25°, for example. The angles specified here each define a conical region in front of the motor vehicle in which typical surroundings sensors can detect objects.Note that the detection range has a spatial and a temporal component. Along a travel path, a surroundings sensor spatially detects all immovable objects at a certain width (which should be greater than the width of the travel path), which depends on which distance in front of a motor vehicle the surroundings sensor can still scan. For moving objects, i.e. in particular external road users, which can move onto the travel path at a predefinable maximum speed, the situation is different. Depending on the speeds of the motor vehicle and road user, a possibly circular segment-shaped detection range is obtained in the temporal sequence starting from the environment sensor, outside which the road user can move without being detected by the environment sensor. This illustration is used herein in the drawings. As a result of this temporal component of the detection range, it may occur that, for example, a pedestrian is still outside the maximum width detected by the environment sensor when the motor vehicle is far away, but later approaches the travel path even further or even collides with the motor vehicle without reaching the detection range which is increasingly less wide over time at its respective location.Objects that are laterally outside the detection range cannot be detected. Since it is an opening angle, a spatial section is measured, at which the sensor is the point at which the detection range originates. The covered width is close to zero at this point (if the blind area in front of the sensor is neglected). The recognition of objects requires a certain time. In order to develop the detection robust with respect to noise, often only the objects that were visible for a certain time are accepted by the system.Acceleration sensors to measure a pedestrian impact are connected to the vehicle body. Thus, for example, the travel through a striking hole can generate an exactly large signal as a pedestrian who is struck, since acceleration sensors only supply a few pieces of information (measurement channels) in comparison with environment sensors (e.g. video sensor equal to 1,000,000 measurement points or "pixels").In the case of functional characteristics for pedestrian protection or "pedestrian collision", a high degree of robustness should exist in order to ensure protection for pedestrians which is as good as possible. Therefore, for example, the threshold for triggering pedestrian protection is increased (i.e., made less sensitive) if no pedestrian or no potential pedestrian has been detected by the environment sensor. Conversely, the threshold is lowered if a pedestrian has been detected. The problem here is the opening angle of the environment sensor: in the near range in which the accident occurs, the width covered by the sensor is very small (opening angle, origin in the sensor).In the article by S. N. Huang, J. K. Yang and F. Eklund "Analysis of Car-Pedestrian Impact Scenarios for the Evaluation of a Pedestrian Sensor System Based on the Accident Data from Sweden", basic situations for potential collisions of pedestrians and vehicles are described and sensor systems for vehicles for detecting such situations.It should be noted that the operations and systems illustrated here for pedestrians can also be transferred to other road users (who are safe, external, i.e. located outside the motor vehicle), for example playing children, two-wheeled riders, wheelchair riders or the like, with certain restrictions. In the following, therefore, occasionally also collision objects or road users with protection are mentioned. The measures described here in particular in connection with the protection of external road users can optionally also be transferred to other protection functions, such as, for example, the protection of occupants of the motor vehicle.When triggering pedestrian protection systems, in the event of a pedestrian colliding with a motor vehicle, the consequences for the pedestrian can often be mitigated. Examples of suitable protection systems are pedestrian airbags or the folding up of the engine hood. In most systems, the triggering leads to the vehicle not being able to continue to drive and irreversible systems first having to be replaced, which causes costs.It is therefore desirable that pedestrian protection systems be triggered incorrectly only rarely if there is actually no collision with a pedestrian. Depending on the complexity of a pedestrian protection system, mis-triggering may have many causes, e.g., unevenness of the road, collisions with small animals, stone hammer, or objects on a road.Advanced assistance systems for motor vehicles are therefore designed to identify pedestrians in the environment of the motor vehicle as far as possible by suitable sensors even before a collision and to predict potential collisions from the relative speed between pedestrian and vehicle, for example, and, if possible, to prevent them or to mitigate the consequences for the pedestrian by triggering pedestrian protection systems in good time when detecting an impact. In such cases, false triggering is unlikely.Conversely, this means that delayed deployments or even false deployments are more likely in the event of a detected impact if no pedestrian was previously identified on the collision course. These cases are addressed by the present process. Particular attention is paid here to situations in which obstacles in the lateral region of a planned travel path restrict the detection range of at least one environment sensor.While in the free detection range of the vehicle, the scanning of the environment, in particular in front of the vehicle (as described in the above-mentioned article), is relatively simple and leads to simple physical relationships when at which relative speeds an identified pedestrian can potentially collide with the vehicle and where the collision point is located, this is not the case in the presence of obstacles which can occlude potential collision objects. As will be described in more detail below, it is rather possible, in the presence of obstacles, for a pedestrian to be detected and identified not, or not reliably, by at least one environment sensor of the motor vehicle, even though he and the motor vehicle are located on a potential collision course. Although a collision is then registered by sensor systems of the motor vehicle, it cannot be associated with a pedestrian already identified beforehand, as would be the case without obstacles.A journey without obstacles in the detection range of environment sensors of the vehicle is considered here as a standard driving situation in which specific conditions and threshold values for triggering external protection systems can be predefined. Since a specific protection system can only be triggered or cannot be triggered, it still being possible to determine the exact time of the triggering, a criterion must ultimately be established, in the presence of which a triggering takes place. In general, a protection system processes different information from different sensors, so that the triggering criterion can include the simultaneous presence of a plurality of information, optionally also with different weighting. False triggerings can be largely avoided in this case by checking the data from environment sensors and impact sensors for plausibility, so that triggering only occurs if a sufficient probability for the impact of a pedestrian is speaking. The method described herein involves the protection of pedestrians in the presence of obstacles laterally of the travel path.The method is based on the knowledge that environment sensors of a motor vehicle are typically equipped in such a way that they cover a sufficiently wide range in front of a moving motor vehicle in order to detect road users who (provided that their speed and direction of movement are maintained) can collide with the motor vehicle. It is therefore possible to require the simultaneous presence of two conditions as a triggering criterion for pedestrian protection measures, for example "a pedestrian on a collision course was detected" and "the impact of an object was detected by the contact sensor". In practice, a probability (weighting) can be assigned to the conditions and a certain minimum probability (threshold value) for the presence of a road user's impact can be used as a triggering criterion. This is particularly concerned with the contribution that environment sensors can still perform to observe the environment in front of the motor vehicle in the presence of obstacles (not obstacles which prevent continued travel, but "visibilitys") in the detection range of environment sensors. For this purpose, the method acquires information on the presence of obstacles and their location and size from the scanning data of the environment. From this, a reduced detection range is calculated which remains after subtracting the areas covered by obstacles. Because of the obstacles, external road users (pedestrians, cyclists, children running) can cross the travel path predefined by the motor vehicle with their typical maximum speeds and can be struck by the motor vehicle without ever having entered the reduced detection range of the environment sensor or environment sensors. If the calculation of the reduced detection range therefore reveals that such a situation is present, the triggering criterion is changed in the method described, in particular to the effect that it is no longer necessary for all conditions to be present in a standard driving situation for triggering the protection system. The advantage of this change in the triggering criterion is that protective measures are initiated even if only the impact of an object is detected, but an identification of a potential collision object has not previously taken place. Although it is no longer possible in this way to avoid false triggering, for example due to stone impact or a striking hole, just as in the case of obstacle-free driving, the safety for pedestrians is increased, which is important especially in the presence of numerous obstacles, in particular at intersections and in the urban area.The triggering criterion predefined in step a) for a standard driving situation is, for example, a triggering criterion which is defined for a usual travel without obstacles.The detection range (step b) is preferably detected via a surroundings sensor connected to a device for carrying out the method. Data obtained from a surroundings sensor may be detected via corresponding inputs. The same applies to the determination of obstacles in step c). Data can also be used for this purpose, which are supplied to a device for carrying out the method via connected devices. Data used for ascertaining obstacles also include, in particularThe calculations carried out in step d) can be carried out, for example, in a device for carrying out the described method.Maximum speeds which are taken into account in step e) can be stored as permanently stored parameters in a control device. Depending on the traffic situation, other maximum speeds for possible road users or else collision objects can be stored. For a traffic situation in the city, a maximum speed can be established, for example, on the basis of the maximum possible speed of a running person (for example, between 20 km / h and 25 km / h). For driving situations on the country, other maximum speeds (for example between 25 km / h and 60 km / h) can be stored in order to still safely include even faster road users such as two-wheeled riders.In steps e) and f), calculations or changes of a triggering criterion are carried out depending on whether collision objects have not entered a detection region of the motor vehicle or have entered a detection region of the motor vehicle only with a very short penetration phase. When collision objects have not entered the detection area, it is obvious that they could not be classified. However, even in the case of an only very short penetration phase, which is smaller than a threshold duration, for example, classification may not have been possible.In a preferred embodiment of the method, conventional image processing systems and / or navigation data are taken into account for ascertaining obstacles. In predictive driving assistance systems, image processing systems are common that can identify objects and / or estimate their size. In general, these are stationary or very slow moving objects.In a preferred embodiment of the method, the triggering criterion contains, in a standard driving situation, at least two different conditions for detecting an external traffic participant to be protected, in particular a pedestrian, but the changed triggering criterion weights at least one condition to a lesser or lesser extent. As already explained, in the presence of obstacles, the condition "road user on collision course was detected" can be omitted, so that the triggering criterion is based only on the information of the remaining sensor system. In principle, it is possible not to completely omit a condition, but to take it into account only with less weight. In this way, for example, the overall situation (number of obstacles, detected number of potential collision objects, traffic guidance, etc.) can be taken into account better.In a particular embodiment of the method, the triggering criterion in a standard driving situation contains the conditions that the protection system has identified a road user to be protected, in particular a pedestrian, as a collision object by means of the environment sensor, but this condition is not demanded or is weighted less heavily in the presence of obstacles.In practice, the protection systems in motor vehicles are being developed more and more, so that two or more environment sensors or the separate analysis of two or more detection ranges of an environment sensor are also possible. In such cases, the described method allows, in a particular embodiment, the environment to be divided into two or more detection areas, wherein the conditions for recognizing a collision object as external, protectable road users, in particular pedestrians, can be selected and / or weighted differently for the different detection areas. In this way, the protection system can be adapted even better to different traffic situations. This is helpful in particular because in the case of right-hand and left-hand traffic, no symmetrical conditions prevail with respect to the environment in front of the motor vehicle (for example, obstacles are typically closer to the travel path on the right-hand side of the road in the case of right-hand traffic).In a preferred embodiment, the motor vehicle is equipped with one or more contact sensors, by means of which a location of the contact on the motor vehicle can be determined. That is, not only the fact of a collision is detected, but also the approximate location of the impact on the front of the motor vehicle is detected. In the method described here, a main contact region having a region width and a region position is now determined in which contact with a road user that is important for protection cannot take place without prior detection by the environment sensor in the detection region, so that the triggering criterion for a standard driving situation is retained for this main contact region despite an obstacle. Road users not detected by the environment sensor can typically only arrive in edge regions of the vehicle front because of their low speed relative to the motor vehicle, since otherwise they would, however, first come into the detection region of the environment sensor. For this reason, it can be ruled out with high probability for a main contact region that a road user who was not previously in the detection region is hit there.In a special embodiment of the method, the main contact area is adapted in its area width and / or its area position depending on the number and / or size and / or position of detected obstacles. This means that, for example, the region width is reduced by numerous obstacles and / or obstacles located very close to the travel path. Due to the geometric situation of a surroundings sensor when driving between obstacles, a displacement of the main contact region from the center of the front to a side opposite the obstacles is also a reasonable measure in order simultaneously to ensure protection for road users and to reduce the probability of false triggerings. These measures alone for adapting the region width and / or the region position of the main contact region represent a preferred embodiment of the method which enables an adaptation of the triggering criterion of a protection system even in the case of only very rough calculation of the detection region of the environment sensor.In this case, the region width of the main contact region is preferably also determined as a function of the speed of the motor vehicle, in particular also when the speed changes. The range width may tend to be selected to be greater as the speed of the motor vehicle increases, because a road user not detected by the environment sensor may get into the front region of the motor vehicle to an ever lesser extent as the speed of the motor vehicle increases, without being identified beforehand.The device for setting at least one triggering criterion (A1) of a protection system for external road users is correspondingly designed.The device is suitable for carrying out the method described above and serves for the reliable protection of external road users even in the case of obstacles present laterally of the travel path of a motor vehicle, wherein the device increases the protection for external road users in the case of obstacles present without raising the risk of false triggerings too much.The device is in particular a control device which is set up to carry out the described method.Such a device is particularly preferably equipped with further means for detecting a contact of the motor vehicle with a collision object, and with a trigger for pedestrian protection measures when a contact is detected at the contact sensor as a function of the changed triggering criterion. Means for detecting a contact can be, for example, inputs for signals for contacting the motor vehicle with a collision object and / or sensors for detecting such a contact.A computer program which is set up to carry out the described method and a machine-readable storage medium on which this computer program is stored are also to be described here.Details of the method and exemplary embodiments are explained in more detail below with reference to the drawings. The following are shown: FIG. 1 : schematically, a motor vehicle and its environment when driving between obstacles, FIG. 2 : shows the front area of a motor vehicle just before the collision with a pedestrian in a schematic illustration, and FIG. 3 : shows a schematic flow chart for illustrating the sequences in the method described.In FIG. 1, a motor vehicle 1 is shown on a road 19 with a potential travel path 20 between various obstacles 22, which travel forward at a speed L. On both sides of the road 19, a pedestrian path 21 is present. The motor vehicle 1 has a protection system 2 for external road users 7, 8, 9, which move at a predefinable maximum speed V. Motor vehicle 1 scans the environment in a detection range S, which is essentially in the form of a segment of a circle, using a surroundings sensor 3. If obstacles 22, for example parked vehicles, are present, this reduces the detection range S. The covered areas C (shown with hatching) cannot be scanned by the environment sensor 3, resulting in a reduced detection range SR (detection range S without the hatched covered areas C). A embankment 23 or a wall 24, for example, can also cause covered areas, i.e. can be recognized as an obstacle 22. Typical processing systems for scanning data of a surroundings sensor 3 can detect the boundaries of obstacles 22 and determine their extent in the direction of travel. They can also recognize their lateral distance D from the motor vehicle 1 and determine the covered areas C therefrom. In FIG. 1, the potential collision object 8, here a pedestrian protruding as obstacles 22 between two parked vehicles, is thus not recognizable by the surroundings sensor. On the other hand, a first road user 7 on the pedestrian path can be identified without any problems, it also being possible to identify in which direction he is moving and whether a collision is to be feared. However, a second road user 9 cannot be recognized behind the wall 24 in the covered area C, so that he could possibly still be hit on the right outside in the front area without previous identification of the motor vehicle 1. Furthermore, the motor vehicle 1 has a contact sensor 4 (which can also be composed of a plurality of individual sensors), with which the fact of a collision and the approximate location of the impact can be determined. The contact sensor 4 is assigned a main contact region 5, the region width BB and region position BL of which in the front region of the motor vehicle 1 can be designed as a function of the traffic situation. The protection system 2 has a recognition unit 11 which can recognize obstacles 22 by means of image processing or other processing of scanning data of the environment and determine its number, size and position. In a subsequent calculation unit 13, to which the information about obstacles is forwarded, a reduced detection range SR is calculated therefrom. In the event of a collision with a collision object 8 considered as a road user that is important for protection, the motor vehicle 1 has at least one protection component 6, by triggering of which the consequences of a collision for the road user 7 are to be mitigated. It can be seen in FIG. 1 that road users 7 can be identified and tracked by reduced detection range SR of environment sensor 3 even when driving between obstacles. Depending on the number and location of obstacles 22, however, this does not apply to all potential collision objects 8, here likewise a pedestrian who wishes to cross the road 19. This is initially covered by an obstacle 22 and therefore cannot be noted or even identified by environment sensor 3. There is a risk that the collision object 8 will be struck by the motor vehicle 1 during the further course of the trip without it having previously been in the detection range S at all or long enough to be correctly identified. The protection system is to be designed to be safe and robust especially for such processes.FIG. 2 shows a constellation as can result after continuation of the movements of motor vehicle 1 on travel path 20 and collision object 8. The enlarged front region of the motor vehicle 1 with the contact sensor 4 and its main contact region 5 is shown. In the present case, a pedestrian moving at maximum speed V is shown as the collision object 8 shortly before the collision with the motor vehicle 1. Because of the obstacles 22 which are only at a small lateral distance D from the travel path 20, the main contact region 5 is set to a relatively narrow region ready BB and, as is indicated by a dashed line, can also be arranged asymmetrically at the vehicle front, in particular displaced counter to the position of obstacles 22. Such a region position BL takes into account the fact that the reduced detection region SR is also not necessarily symmetrical with respect to the travel path. The collision object 8 will strike the contact sensor 4 outside the main contact region 5 without having been previously identified by the environment sensor because it was either not at all or only too short for identification in the reduced detection region SR. For this reason, it is advantageous for the protection of this collision object 8 if the protection component 6 is triggered as soon as the contact sensor 4 responds and also without the environment sensor 3 having previously identified it as a pedestrian, for example. If the collision object 8 were somewhat faster or the motor vehicle 1 was somewhat slower, its presence in the detection area S would have been sufficiently long for identification and the collision would take place in the main contact area 5. In this case as well, the protective component 6 would be triggered, but only if there is a simultaneous presence of a contact with the main contact region 5 and an identification as a pedestrian by the environment sensor 3. If, however, a stone or a small animal hit the main contact region 5 in this situation or a striking hole were to trigger signals, the protective component 6 would not be triggered because no identification as a pedestrian had previously taken place.FIG. 3 shows a schematic representation of the sequence of the method in the protection system 2. data supplied by the environment sensor 3 are supplied to a query 11 about the presence of obstacles. If obstacles 22 are reported, the data are passed on to the calculation unit 13. If no obstacles 22 are detected, a triggering criterion A 1 remains relevant for a standard driving situation in a storage medium 10. If obstacles are detected, the calculation unit 13 calculates a reduced detection range SR of the environment sensor 3 and triggers means 14 for changing the triggering criterion A 1 into a changed triggering criterion A 2. The changed triggering criterion A 2 can also be stored in the storage medium 10. If a collision triggers a signal of the contact sensor 4, there are two possibilities depending on whether or not the detected contact is located in the main contact region 5. For this purpose, a query 16 is made as to the main contact area 5. in the system chosen here as an example, the area width BB and the area position BL of the main contact area 5 are defined by a means 15 for changing the main contact area 5, which in turn is controlled by the calculation unit 13 depending on the number, type and location of obstacles 22. If no obstacles are present, the main contact region 5 typically extends over the entire contact sensor 4. If the query 16 establishes that the main contact region 5 has been touched, then triggering of the protective component 6 takes place according to the triggering criterion A 1 for a standard driving situation via the triggering 18 both in the presence of obstacles 22 and without obstacles. If, on the other hand, the main contact region 5 is not touched, then triggering of the protective component 6 takes place via the triggering 17 for a situation with obstacles in accordance with the changed triggering criterion A 2, that is to say, for example, even if no pedestrian has been identified by means of the environment sensor 3.By means of the described protection system, road users are protected by suitable triggering criteria in the event of a collision even in the presence of obstacles to the side of the travel path, without the risk of false triggerings of protection components being unduly increased.

Claims

Method for setting at least one triggering criterion (A1) of a protection system (2) in a motor vehicle (1) having at least one environment sensor (3) with the following steps: a) presetting the triggering criterion (A1) for a standard driving situation, b) detecting at least one detection region (S) with the at least one environment sensor (3) of the motor vehicle (1), characterized in that the method additionally has the following steps: c) determining whether at least one obstacle (22) limiting the detection by the environment sensor (3) is present in the detection region (S), d) calculating at least one reduced detection region (SR) of the environment sensor (3) for detecting potential collision objects (8) as a function of the at least one detected obstacle (22). e) calculating, whether collision objects (8) can be struck by the motor vehicle (1) at a predeterminable maximum speed (V) without having been detected beforehand by the environment sensor (3) in the reduced detection range (SR), f) changing the triggering criterion (A1) into a changed triggering criterion (A2), if step e) reveals that collision objects (8) can be struck by the motor vehicle (1) without having been detected beforehand by the environment sensor (3) in the reduced detection range (SR).Method according to Claim 1, wherein conventional image processing systems and / or navigation data are used for determining obstacles (22).Method according to one of the preceding claims, wherein the triggering criterion (A1) in a standard driving situation in step a) contains at least two different conditions for detecting a road user (7) to be protected, but the changed triggering criterion (A2) comprises at least one condition less or at least one condition is less heavily weighted.Method according to one of the preceding claims, wherein the triggering criterion (A1) in the case of a standard driving situation in step a) contains the condition that the protection system (2) has identified a road user (7) to be protected as a collision object (8) by means of the environment sensor (3) before a collision, but this condition is not weighted or is weighted to a lesser extent in the case of a reduced detection range (SR) and after a change in the triggering criterion (A1) in step f).Method according to one of the preceding claims, wherein an environment in front of the motor vehicle (1) is divided in step f) into two or more detection areas (E1, E2), wherein the conditions for detecting a collision object (8) as a road user (7) with protection value can be chosen differently and / or weighted for the different detection areas.Method according to one of the preceding claims, wherein the following steps are carried out subsequently to step f): g) detecting a contact of the motor vehicle (1) with a collision object (8) with the aid of at least one contact sensor (4), h) initiating protective measures according to the changed triggering criterion (A2).Method according to Claim 6, wherein the at least one contact sensor (4) can determine a location of the contact on the motor vehicle (1), wherein, as a function of the number and / or size and / or position of detected obstacles (22), at least one main contact region (5) having a region width (BB) and a region position (BL) is determined, in which contact with a road user (7) that is important for protection cannot take place without prior detection by the environment sensor (3) in the reduced detection region (SR), so that for this main contact region (5) the triggering criterion (A1) for a standard driving situation is retained despite the presence of at least one obstacle (22).Method according to Claim 7, wherein the main contact region (5) is adapted in its region width (BB) and / or its region position (BL) as a function of the number and / or size and / or position of detected obstacles (22).Method according to Claim 7 or 8, wherein the region width (BB) of the main contact region (5) is also determined as a function of the speed (L) of the motor vehicle (1).Device for setting at least one triggering criterion (A1) of a protection system (2) for a motor vehicle (1), having the following features: - a storage medium (10) for the triggering criterion (A1) for a standard driving situation, - at least one environment sensor (3) for ascertaining scan data of an environment of the motor vehicle (1) for at least one detection region (S), characterized in that the device additionally has the following features: - a detection unit (11) for ascertaining obstacles (22) which restrict the environment sensor (3) during the detection in the detection region (S), a calculation unit (13) for calculating a reduced detection range (SR) of the environment sensor (3) for detecting potential collision objects (8) as a function of the number and / or size and / or position of obstacles (22) determined and for calculating whether collision objects (8) can be struck by the motor vehicle at a predeterminable maximum speed (V) without having been detected by the environment sensor (3) in the detection range (SR), means (14) for changing the triggering criterion (A1) into a changed triggering criterion (A2) as a function of the result of the calculation in step d),Computer program which is set up to carry out all the steps of the method according to one of Claims 1 to 9.A machine readable storage medium having stored thereon the computer program of claim 11.

Images