Method and system for controlling the light emission of a vehicle lamp

Description

[0001] The present invention relates to a method for controlling the light emission of a vehicle lamp. Furthermore, the invention relates to a system for controlling the light emission of a vehicle lamp.

[0002] Vehicle lights serve, among other things, to alert traffic to changes in driving behavior and to warn of dangerous situations. For example, the illumination of a brake light can alert traffic behind to a vehicle's deceleration. With adaptive vehicle lights, the light emission can be adjusted according to the traffic situation. This can increase the visibility of a vehicle in a potentially dangerous situation and thus improve road safety. For example, a second brake light can be activated in addition to the brake light during an emergency stop. Furthermore, ambient information can be incorporated into the adaptive adjustment. However, an adjustment of the light emission of a vehicle light according to the traffic situation is not always easy to interpret.

[0003] WO 2020 / 087352 A1 discloses a method for controlling a vehicle's lighting system. Environmental information, such as objects in the vehicle's surroundings and the distance between an object in the surroundings and the vehicle, is collected, and based on this environmental information, the vehicle's lighting system is automatically adjusted. Artificial intelligence can be used to analyze the environmental information.

[0004] DE 10 2004 041 429 A1 describes a method for controlling a vehicle's rear fog light. External data automatically determines whether the rear fog light is switched on, off, or left in its switched state. Using an approximate target switching state determined by artificial intelligence, the maximum speed is limited by automatically intervening in the engine control system when visibility falls below a specified range.

[0005] WO 2020 / 013754 A1 discloses a method for controlling a vehicle lighting device. The characteristics of the various headlights can be adapted to the weather. An external control center can receive tasks from an operator via a user interface and execute them using artificial intelligence.

[0006] Further methods for controlling the light emission of a vehicle lamp are known from US2021 / 114514 A1 and DE102016216680 A1.

[0007] The disadvantage of these devices and methods is that, despite adaptive control of the light emission, the lighting function of the light emission is inadequately perceived and interpreted by other road users.

[0008] The invention is therefore based on the object of providing a method and a device in which an adaptive light emission is better perceived and interpreted by other road users.

[0009] According to the invention, this object is achieved by a method having the features of claim 1 and a system having the features of claim 9. Advantageous embodiments and further developments are specified in the dependent claims.

[0010] In the method according to the invention, a road user positioned in the light emission direction of the vehicle light is detected. The light emission of the vehicle light is changed, whereupon reaction data regarding the road user's reaction to the change in light emission is recorded. Forecast data indicating the expected reaction to the changed light emission is determined. The recorded reaction data are compared with the determined forecast data, and if the forecast data deviate from the reaction data, it is determined which adjustment of the light emission of the vehicle light is predicted to reduce the deviation. The light emission of the vehicle light is controlled according to the determined adjustment.

[0011] A change in light emission can affect a tail, brake, indicator, and reversing lamp (SBBR) or high-mounted brake lights, a separate indicator light, a front light, or a side light. The detected road user can be, for example, a driver of another motor vehicle, a cyclist, or a pedestrian. The pedestrian can be in front of, behind, or to the side of the vehicle within a distance defined by the vehicle's detection radius.

[0012] The reaction data provides information on the reaction of the detected road user to a change in light emission. For example, it can contain data from a distance measurement to another road user.

[0013] Predictive data describes data generated, particularly during the process, that contains information about the desired behavior of road users in response to a change in light emission. For example, an increase in the intensity of the brake light would be expected to increase the distance between the vehicle and the detected road user.

[0014] A deviation between the forecast data and the reaction data can therefore occur, for example, if the distance has not increased, even though an increase in the distance would be expected due to the changed light emission. Adjusting the light emission to a change in light emission can be performed several times in succession until the deviation between the forecast data and the reaction data is within a tolerance range.

[0015] The method advantageously adjusts the setting of the vehicle light independently based on the comparison of the measured reaction data of the road users and the forecast data.

[0016] By recording reaction data, the control system can advantageously be better adapted to the individual way in which the detected road user reacts. For example, a first road user detects braking by the vehicle in front, in which the brake light is illuminated with low light intensity, very quickly, and no further adjustment is necessary. A second road user might react more slowly in such a case, for example. Since the reaction data then do not match the forecast data, a control action is taken, for example by increasing the intensity of the brake light's light emission. This advantageously means that the driver does not have to pay additional attention to how other road users react to changes in the light emission of their own vehicle, whether in traffic flow or in dangerous situations.

[0017] The control can be initiated not only by manually changing the light emission of the vehicle's headlights. It is also activated when an automatic change in the light emission of the vehicle's headlights has occurred; for example, when a hazardous situation has been detected and the light emission of the vehicle's headlights has been automatically changed, even before the driver has reacted.

[0018] According to the method according to the invention, environmental data from the vehicle's surroundings are collected, and a level for the hazard severity in the surroundings is determined based on the environmental data. When the light emission of the vehicle's light changes, the light intensity of the light emission of the vehicle's light is changed depending on the determined hazard severity, so that the light intensity of the light emission of the vehicle's light is adjusted according to the hazard severity level.

[0019] For this purpose, a front camera can be installed in the vehicle to monitor the traffic environment or a sensor system for monitoring rear traffic, such as rear radar, ultrasound, rear camera, or LIDAR. The relative or absolute positions of the road user can also be measured and transmitted using GPS sensors. The environmental data can thus include, for example, image data from a camera, measurement data from a LIDAR sensor, measurement data from a radar sensor, and measurement data from an ultrasonic sensor. In addition, measurement data from a rain sensor, measurement data from a brightness sensor, measurement data from a temperature sensor, measurement data from a humidity sensor, and / or measurement data from an air pressure sensor can be taken into account.

[0020] Using the sensor technology for capturing environmental data, other road users positioned in the direction of the vehicle's light emission can also be detected, regardless of the vehicle's equipment. Furthermore, reaction data can also be captured, i.e., the other road user's reaction to a change in the light emission.

[0021] The calculation of the hazard severity level can include information on slowing traffic, the end of a traffic jam, an accident, impaired visibility due to fog, or difficult driving conditions due to severe weather. Furthermore, information on the speed of the vehicle or other road users or their distance from one another can be used in the calculation. The individual pieces of information can also be categorized according to their severity. For example, the distance to the end of the traffic jam or the thickness of the fog can be taken into account. A hazard severity level is then determined from the information obtained from the environmental data.

[0022] In this context, the luminous intensity of the light emitted by a vehicle's light source refers specifically to the luminous flux, i.e., how much light the light source emits per unit of time that is perceptible to the human eye. The luminous flux corresponds to the physical radiant power but also takes into account the sensitivity of the human eye.

[0023] If the light intensity of the light emission of the vehicle lamp is changed depending on the determined hazard severity, this means in particular that a low light intensity is selected for a low hazard severity, but a high light intensity is selected for a high hazard severity.

[0024] Advantageously, the light intensity is adjusted to the severity of the hazard, allowing road users to identify the current hazard level. For example, if the initial speed is low, the distance is large, and only slight braking is occurring due to slightly decelerating traffic, a low hazard severity is calculated and a low light intensity is selected. This prevents abrupt braking by following road users, which could lead to accidents or disrupt traffic flow. On the other hand, at high speeds, during severe weather, and during heavy braking, for example, a high hazard severity is calculated and a high light intensity is selected. This warns the road user and allows them to initiate a more severe braking maneuver.

[0025] Without any intervention by the driver, the light emission is advantageously regulated in such a way that the visibility of a vehicle is increased in a dangerous situation and that warnings and / or information are optimally recognized by the changing lighting, thus increasing road safety.

[0026] According to a further development of the method according to the invention, when changing the light emission of the vehicle lamp, the light intensity is continuously increased from an initial light intensity to the light emission corresponding to the level of the hazard severity.

[0027] A continuous increase in light intensity is defined as a steadily monotonically increasing change in light intensity. For example, when adjusting light emission in a hazardous situation from a low to a high level of hazard severity, the increase is not abrupt, but rather continuous. Advantageously, other road users are not dazzled by the sharp increase in intensity, preventing them from being caught off guard, thus reducing the risk of an accident.

[0028] Adaptive adjustment of light intensity also allows the dynamic development of the situation to be communicated. For example, if the driver detects a traffic jam and brakes slowly, a warning is provided with only low light intensity, e.g., in the form of a small illuminated area or low brightness. However, if the braking force increases during this braking process because the end of the traffic jam appears faster than expected, the intensity level is continuously increased as the braking force increases.

[0029] Adjusting the light intensity can also affect the headlights or other vehicle lights, for example if the road user is a pedestrian.

[0030] According to a further embodiment of the method according to the invention, the brightness in the surroundings of the vehicle light is detected and when the light emission of the vehicle light changes, the brightness of the light emission of the vehicle light is changed such that a certain contrast is achieved between the brightness of the light emission of the vehicle light and the detected brightness in the surroundings of the vehicle light.

[0031] The ambient brightness can be measured using sensors at the rear of the vehicle as well as in the entire traffic environment, especially in the area of traffic behind.

[0032] By comparing the ambient brightness with the brightness of the vehicle's light emission, a contrast can be determined. Contrast is defined as the difference in intensity between the ambient brightness and the brightness of the vehicle's light emission.

[0033] The brightness of the light emission of the vehicle lamp is, in particular, the luminance of the light-emitting surface of the vehicle lamp. Luminance takes into account the spatial and directional dependence of the luminous flux emitted by the vehicle lamp. It indicates the surface brightness with which the eye perceives a surface of the vehicle lamp.

[0034] Ideally, road users should be able to clearly see the light emitted by the vehicle's headlights, so the contrast should be correspondingly high. However, too high a contrast is detrimental, as it could dazzle road users.

[0035] Accordingly, in high ambient brightness, for example, due to intense sunlight, the default brightness of the vehicle's light emission will only achieve a low contrast. This makes it more difficult for road users to perceive the light emission. To improve the visibility of the vehicle's light, increasing the brightness is necessary. This ensures that the vehicle's signaling function can be maintained, especially in hazardous situations. Blinding road users is unlikely, as they are already accustomed to the brightness due to the intense sunlight.

[0036] Conversely, in lower ambient light conditions, such as dusk or darkness at night or in a tunnel, the default brightness of the vehicle's light emission will achieve a very high contrast. To avoid dazzling road users, the brightness must be reduced. This constant glare would be particularly annoying when the vehicle is stationary in a traffic jam or when driving in a line of cars. However, the brightness is still set high enough to ensure that the vehicle's light can still be detected.

[0037] According to a further development of the method according to the invention, a defined contrast is set depending on the level of danger severity, wherein a higher contrast is set at a higher level of danger severity.

[0038] Adjusting the contrast based on hazard severity can be coupled with adjusting the contrast based on ambient brightness. To create a defined target contrast that corresponds to the respective hazard severity, a higher brightness is required in high ambient brightness, such as during the day, than at night or when driving through tunnels.

[0039] This is achieved, for example, with a contrast matrix in which the various combinations of low and high brightness, and low and high hazard severity, are assigned corresponding contrasts. For example, with high ambient brightness and a high hazard severity, a maximum contrast would be set, meaning the brightness of the vehicle's lights would be correspondingly high. This would achieve a strong signaling effect to optimally warn traffic. In low ambient brightness, such as in a tunnel, and with a low hazard severity, a minimal signaling effect would be sufficient to inform road users but not dazzle or surprise them.

[0040] In cases of medium ambient brightness and medium hazard severity, the system is adjusted accordingly. The beneficial effect for road users can thus be utilized for all ambient brightness and hazard severity levels.

[0041] According to a further development of the method according to the invention, the vehicle light has a surface with a plurality of light sources, and the light intensity is adjusted via the brightness and / or the luminous area of the vehicle light. The change in light emission can thus be achieved by changing the intensity of the light emission, i.e., by changing the luminance, i.e., by changing the radiation flux per area, or by using a larger area with the same luminance.

[0042] The plurality of light sources can, in particular, be a plurality of light-emitting diodes (LEDs). The light intensity can then be varied, in particular, by adaptive intensity control in the individual LEDs.

[0043] The vehicle light can also include individual LEDs, some of which can be combined to form sub-areas of the respective lighting system. For example, the rear light can consist of several separate subcomponents. Depending on the desired signaling effect, the illuminated area can be made smaller or larger. Enlarging the illuminated area is achieved by individually controllable illuminated area elements of the respective lighting system.

[0044] In addition to or as an alternative to the brightness of a defined area of the vehicle light, the size of the illuminated area of the vehicle light can also be adjusted. This can be adjusted according to the contrast and / or the level of hazard severity.

[0045] For example, a small illuminated area is activated when the hazard severity is low, while a larger illuminated area is activated when the hazard severity is high. The warning effect is intuitively greater with a larger illuminated area. This allows for a better warning of traffic when the hazard severity is high, for example, when there is an increased risk of collision. In this case, the brightness of the illuminated area is not changed.

[0046] Preferably, both changes in the light emission of the vehicle light—i.e., a change in brightness and a change in the size of the illuminated area of the vehicle light—can be combined. The adjustment can, in turn, be made depending on both the contrast and the level of hazard severity. For example, if the hazard severity is high in a bright environment, a larger illuminated area is activated and the brightness is increased.

[0047] By changing the light emission of the vehicle lamp, the luminous area can be continuously increased from an initial light area to the target light area according to the level of hazard severity.

[0048] Different signal combinations are therefore available for different warning situations and warning levels, which road users can intuitively link to the current situation.

[0049] According to a further embodiment of the method according to the invention, a machine learning method is used to determine the forecast data. The use of a machine learning method advantageously makes it possible to determine particularly reliable forecast data.

[0050] According to a further development of the method according to the invention, reaction data are stored in a database and the reaction data of the database are used for training the parameter set of the machine learning method for determining the forecast data.

[0051] Before the process is used, for example before the vehicle is delivered, an artificial intelligence system is trained. This involves collecting driver reaction data to a wide variety of driving situations, storing this data in a database and using this data for training. During the process, the artificial intelligence, which uses the machine learning process, can determine in real time which reaction is expected in which driving situation in order to create the forecast data. In addition to the actual driver reaction, a wide variety of environmental factors are also taken into account in the machine learning process. The process can therefore create reliable forecast data for a wide variety of driving situations, even for those that were not directly determined in advance and saved in the database.

[0052] Another option for creating the database for determining the predictive data is to continuously store current reaction data during vehicle use or while the method is being used. The data set available for training the machine learning method is thus expanded each time the method is used. This allows the determination of the predictive data to be continuously improved, leading to optimized adjustment of the vehicle's lighting and thus to a safer driving situation.

[0053] The data sets used to determine the forecast data prior to the application of the method can be combined with the current data sets. This allows specific driving situations, especially rare ones, to be included in the database in advance, but the data sets can also be continually expanded to include current driving situations. For example, new developments in traffic situations or unfamiliar environments can also be included.

[0054] The use of artificial intelligence based on machine learning enables automatic recognition of a wide range of environmental factors and overall optimization of the settings. It learns which lighting settings are best for which situations. These situations consist of lighting conditions, surrounding traffic, driver actions, and other environmental factors. The artificial intelligence-based algorithm can learn, in particular, from road users' behavior in relation to the lighting. This includes, for example, a quick reaction or no reaction at all, or driving errors due to glare.

[0055] The large amount of data from various sensors can thus be used to advantage. This ensures that vehicle lighting can be adjusted quickly, and the adjustment is controlled in such a way that an intuitive signal is generated for the driver and the traffic behind.

[0056] The information from the environmental data can be used for the machine learning process to determine the level of hazard severity from the environmental data.

[0057] The artificial intelligence learns from the reaction data whether the assessment of the hazard severity and thus the adjustment of the vehicle lights led to the desired result or whether the level of hazard severity was misjudged.

[0058] According to a further embodiment of the method according to the invention, the reaction data is determined from transmitted data via direct communication between vehicles (car-to-car communication). In this way, for example, the braking activity and / or steering behavior of the detected road user and / or the eye movements of the driver of the detected road user can be transmitted directly.

[0059] As an alternative or in addition to the directly recorded environmental data, data from direct communication between vehicles (car-to-car) can be evaluated. For example, it can be determined whether the traffic behind is also stationary or whether there is movement within the convoy. Furthermore, information about the driver or the vehicle can be transmitted. This advantageously allows the environmental data to be expanded to include data that cannot be directly measured via sensors.

[0060] The system according to the invention for controlling the light emission of a vehicle light comprises a detection unit for detecting road users arranged in the direction of light emission and an adjustment unit for changing the light emission of the vehicle light. Furthermore, the system comprises a calculation unit for detecting reaction data on the road user's reaction to the change in light emission, for determining forecast data indicating the expected reaction to the changed light emission, and for comparing the detected reaction data with the determined forecast data. A control unit is designed to determine, in the event of a deviation between the forecast data and the reaction data, which adjustment of the light emission of the vehicle light predicts a reduction in the deviation and to regulate the light emission of the vehicle light according to the determined adjustment.

[0061] The system according to the invention serves in particular to carry out the method according to the invention. It has the same advantages as the method according to the invention.

[0062] The calculated and applied vehicle light setting can be displayed to the driver in a combination display in the vehicle. This informs the driver which light display other road users are seeing. The driver can therefore check that the control system and thus the light emission are functioning correctly. Furthermore, the driver is warned if the system has generated a hazard signal due to a hazardous situation that they themselves did not recognize or recognized too late. Both of these factors increase the safety of the driver and other road users.

[0063] The light emission can be visualized, for example, using symbols for the individual lamps, which illuminate as soon as the control is activated. The technology used externally, such as an enlarged illuminated area and / or increased brightness of the light emission, can be implemented analogously in the display. In cases of particularly high hazard severity, an additional acoustic warning signal can be triggered.

[0064] All variants of vehicle light control mentioned in the procedure automatically result in an increase in energy efficiency compared to standard vehicle lights without intensity control.

[0065] The invention will now be explained using embodiments with reference to the drawings. Figure 1 shows schematically the components of the system, Figure 2 is a flowchart of the method steps of a first embodiment of the method according to the invention,

[0066] First, an embodiment of the device according to the invention is described with reference to Figure 1 Explained: The device comprises a plurality of environmental sensors 4 arranged in the vehicle 8. These may include cameras, a LIDAR sensor, a radar sensor, an ultrasonic sensor, a rain sensor, a brightness sensor that measures the ambient brightness in the traffic environment of the vehicle 8, a temperature sensor, a humidity sensor, and / or an air pressure sensor. The data acquired by these environmental sensors 4 is transmitted to a detection unit 1 as environmental data.

[0067] Further external environmental data from other road users 6 (car2car) or traffic infrastructure (car2x) is transmitted to the detection unit 1 via the communication unit 9. The detection unit 1 is also connected to a satellite signal receiver, for example, a GPS (Global Positioning System) receiver 10, so that the current position of the vehicle 8 can be determined. The environmental data received by the detection unit 1 is transmitted to a calculation unit 2, which further processes it.

[0068] From the environmental data, calculation unit 2 calculates reaction data regarding the reaction of road user 6 to a change in light emission. Furthermore, calculation unit 2 creates forecast data indicating the expected reaction to the changed light emission.

[0069] For this purpose, the calculation unit 2 is connected to an internal or external database 11. The database 11 contains previously collected reaction data and an artificial intelligence system trained in advance using a machine learning method. The environmental data transmitted to the calculation unit 2 is used to generate the forecast data using the artificial intelligence system that uses the machine learning method. The reaction data determined using the environmental data are also stored in the database 11 to expand the data set for the machine learning method.

[0070] Calculation unit 2 also determines the hazard severity level from the environmental data. For this purpose, the reaction data stored in database 11 is used to determine the correct hazard severity level from previous driving situations using artificial intelligence.

[0071] Calculation unit 2 also serves to compare the recorded reaction data with the determined forecast data. The result is transmitted to control unit 3, which, if the forecast data deviates from the reaction data, determines which adjustment of the light emission of vehicle lamp 5 predicts a reduction in the deviation.

[0072] The device also comprises an adjustment unit 12 for changing the light emission of a vehicle lamp 5. The results of the determined adjustment of the light emission of the vehicle lamp 5 and the corresponding necessary control are sent to the adjustment unit 12, and the light emission of the vehicle lamp 5 is adjusted by the latter.

[0073] The vehicle light 5 comprises lighting devices for a tail light, brake light, indicator light, and reversing light (SBBR) or a high-mounted brake light, a separate indicator light, and / or a secondary headlight. Furthermore, the vehicle light 5 can also be used as a headlight. The vehicle light 5 comprises a plurality of individual light-emitting diodes (LEDs), some of which are combined into sub-areas for specific lighting functions. In this way, the LEDs can form luminous surface elements. The adjustment unit 12 controls the individual LEDs of the vehicle light 5 depending on the calculated control, thus changing the luminous surface of the vehicle light 5 and the light intensity of the light emission of the vehicle light 5.

[0074] The adjustment is also transmitted to a display unit 7, which shows the driver the setting of the vehicle light 5.

[0075] In the following, an embodiment of the method according to the invention is described with reference to Figure 2 explained, with further details of the exemplary embodiment of the device according to the invention being described: Before executing the method, the artificial intelligence is trained in a step S0 to determine the level of hazard severity and the forecast data. For this purpose, driver reaction data to different driving situations and associated environmental data are collected, stored in the database 11, and used for training.

[0076] During the journey, a road user 6 located in the light emission direction of the vehicle light 5 is detected in a first step S1. The road user 6 is a driver of a motor vehicle in the light emission direction of the brake light, i.e., behind the vehicle 8. In another exemplary embodiment, the road user 6 can also be a cyclist or pedestrian and can also be located next to or in front of the vehicle 8.

[0077] In a second step S2, environmental data is recorded using the environmental sensors 4.

[0078] This step S2 includes, among other things, measuring the brightness with a brightness sensor, measuring the distance to other road users, and recording the weather conditions. The calculation unit 2 determines a hazard severity level from the environmental data. In the exemplary embodiment, the driver is traveling on the highway at high speeds. It is bright, but it is raining heavily. Due to an accident in the immediate vicinity, the driver must brake sharply. Therefore, a high hazard severity level is determined. The determination is made with the help of artificial intelligence, which has learned from previous reaction data and environmental data to assess the hazard severity and which associated setting of the vehicle light 5 is suitable to achieve a desired result, e.g., to warn following road users.

[0079] In the subsequent third step S3, the light emission of the vehicle light 5 is changed using the adjustment unit 12. The driver brakes, and the brake light illuminates. Half of the available LEDs are activated, and a medium brightness level is selected. In another embodiment, the light emission change can also occur, for example, by flashing or turning on the headlight.

[0080] In the fourth step S4, the detection unit 1 records reaction data on the reaction of road user 7 to the change in light emission. This information originates from the sensors for monitoring rear traffic, for example, rear radar, ultrasound, rear camera, or LIDAR, which are included in the environment sensors 4. However, it can also be transmitted directly by transmitting the braking activity, steering behavior, or eye movement of road user 6 if the road user 6 has a communication interface that can exchange data with the communication interface 9 of the vehicle 8 (car2car). Furthermore, a front camera can be used to observe the traffic environment. In the exemplary embodiment, the calculation unit 2 determines from the images from the rear camera that road user 6 has not yet initiated a braking maneuver.The calculation unit 2 can also take into account data from the radar sensor and / or LIDAR sensor.

[0081] In a fifth step S5, the reaction data are stored in the database 11 of the calculation unit 2 to expand the parameter set for the machine learning method. The database 11 contains previously stored reaction data from other traffic situations, collected either during the use of the method or previously.

[0082] Prediction data indicating the expected reaction to the changed light emission are determined in a sixth step S6. For this purpose, the reaction data from database 11 is accessed, and the pre-trained artificial intelligence is used to determine the prediction data using a machine learning method. The prediction data predict that the road user would have to brake sharply in response to the changed light emission.

[0083] In a seventh step S7, the reaction data are compared with the prediction data. In the exemplary embodiment, a non-braking vehicle 8 was detected as the reaction data, and a sharply braking vehicle 8 was predicted as the prediction data. Accordingly, a deviation between the prediction data and the reaction data is determined.

[0084] Since a deviation has been detected, in an eighth step S8, the calculation unit 2 determines which adjustment of the light emission is predicted to reduce the deviation. This takes into account the hazard severity level. Since this was classified as high, an increase in the illuminated area or an increase in the brightness of the illuminated LEDs is attempted.

[0085] Second, the contrast of the light emission of vehicle light 5 in the surroundings is determined. The brightness detected by the brightness sensor is compared with the currently set brightness of vehicle light 5. The contrast between vehicle light 5, set to a medium brightness level, and the ambient brightness is determined. It is determined that a better contrast can be achieved with a higher brightness level. The machine learning process generates an optimized light emission setting based on these inputs. In this case, the brightness is increased and the light area is continuously enlarged to the largest possible area.

[0086] The calculated and applied setting of the vehicle light 5 is shown to the driver in a combination display in the vehicle 8. Due to the increased severity of the danger, an acoustic warning signal also sounds.

[0087] The control sequence is now repeated. The new light emission setting is equivalent to the previously described step S3. In the exemplary embodiment, a sharp braking maneuver by road user 6 is now detected due to the change in light emission. The distance between the two vehicles increases. The reaction data recorded after the new light emission setting is stored in database 11 according to step S5 to further improve the machine learning process.

[0088] In this case, the reaction data agree with the forecast data within a predefined tolerance range, so no further adjustment is necessary and the procedure is terminated. List of reference symbols

[0089] 1Detection unit 2Calculation unit 3Control unit 4Environmental sensors 5Vehicle light 6Road user 7Display unit 8Vehicle 9Car 2 Car communication interface 10GPS signal 11Database 12Setting unit

Claims

1. Method for controlling the light emission of a vehicle lamp (5) of a vehicle (8), in which method a road user (6) located in the light emission direction of the vehicle lamp (5) is detected, and the light emission of the vehicle lamp (5) is changed, whereupon reaction data on the reaction of the road user (6) to the change in the light emission are recorded, forecast data are determined which indicate the expected reaction of a road user (6) located in the light emission direction of the vehicle lamp (5) to the changed light emission, the recorded reaction data are compared with the determined forecast data, if the forecast data deviates from the reaction data, it is determined by means of which adjustment of the light emission of the vehicle lamp (5) a prediction of a reduction in the deviation is made, and the light emission of the vehicle lamp (5) is controlled according to the determined adjustment characterized in that environmental data from the surroundings of the vehicle (8) are recorded and a level for the hazard severity in the surroundings is determined using the environmental data, if the light emission of the vehicle lamp (5) changes, the light intensity of the light emission of the vehicle lamp (5) is changed depending on the determined hazard severity, so that the light intensity of the light emission of the vehicle lamp (5) is adjusted according to the level of hazard severity.

2. Method according to claim 1, characterized in that when the light emission of the vehicle lamp (5) is changed, the light intensity is continuously increased from an initial light intensity to the light emission according to the level of hazard severity.

3. Method according to claim 1, characterized in that the brightness in the surroundings of the vehicle lamp (5) is measured, when the light emission of the vehicle lamp (5) is changed, the brightness of the light emission of the vehicle lamp (5) is changed such that a specific contrast is achieved between the brightness of the light emission of the vehicle lamp (5) and the measured brightness in the surroundings of the vehicle lamp (5)4. Method according to claim 1 and 3, characterized in that a defined contrast is set depending on the level of hazard severity, with a higher contrast being set at a higher level of hazard severity.

5. Method according to claim 4, characterized in that the vehicle lamp (5) has a surface comprising a plurality of light sources and the light intensity is adjusted via the brightness and / or the luminous surface of the vehicle lamp (5).

6. Method according to claim 1, characterized in that a machine learning method is used to determine the forecast data.

7. Method according to claim 6, characterized in that reaction data are stored in a database (11) and in that the reaction data of the database are used for training the parameter set of the machine learning method for determining the forecast data.

8. Method according to claim 1, characterized in that the reaction data are determined from transmitted data via Car2Car communication (9).

9. System for controlling the light emission of a vehicle lamp (5), having a detection unit for detecting road users (6) located in the direction of light emission, an adjustment unit (12) for changing the light emission of the vehicle lamp (5) and a calculation unit (2) - for recording reaction data on the reaction of the road user (6) to the change in the light emission, - for determining forecast data indicating the expected reaction to the changed light emission, - for comparing the recorded reaction data with the determined forecast data, a control unit (3) which is designed to determine, in the event of a deviation of the forecast data from the reaction data, which adjustment of the light emission of the vehicle lamp (5) predicts a reduction of the deviation and to control the light emission of the vehicle lamp (5) according to the determined adjustment characterized in that environmental data from the surroundings of the vehicle (8) can be recorded and a level for the hazard severity in the surroundings can be determined using the environmental data, if the light emission of the vehicle lamp (5) is changed, the light intensity of the light emission of the vehicle lamp (5) can be changed depending on the determined hazard severity, so that the light intensity of the light emission of the vehicle lamp (5) can be adjusted according to the level of hazard severity.