Method and apparatus for adjustment of environment-based driver assistance functionality

Abstract

Adjustment of environment-based driver assistance system functions. A method for operating a motor vehicle (1) using a driver assistance system (2) is disclosed, the method comprising the following steps: for at least one driver assistance system function, defining at least one criterion (5) regarding an environment of the vehicle for an adjustment of the driver assistance system function, detecting a plurality of features (6) of the environment of the vehicle, estimating a probability (7) of an occurrence of the at least one criterion based on a combination of the detected features, and adjusting the driver assistance system function (8) based on the estimated probability.

Description

TECHNICAL FIELD

[0001] The present invention relates to a method for operating a motor vehicle using a driver assistance system. Furthermore, the present invention relates to a device for adjusting a driver assistance system function, and to a motor vehicle. BACKGROUND

[0002] Driver assistance systems support the driver in order to improve driving comfort and / or safety while driving. The respective environmental conditions and characteristics of the respective vehicle environment are important input variables for various available functions. For example, hands-off driving should only be available if there are no pedestrians in the vicinity. Furthermore, the vehicle should only switch from a standstill to the use of ACC (adaptive cruise control) parking and starting if the traffic lane in front of the vehicle is free of pedestrians or obstacles.

[0003] The response of a driver assistance function to environmental situations can in principle be influenced by undesired behavior or risks. For example, the vehicle can thus turn to a pedestrian without the driver giving a reason to do so. In a different example, the ACC parking and starting can follow an initial target vehicle, but a pedestrian appears on the traffic lane in front of the vehicle and is put in danger.

[0004] One possibility for handling these situations is the use of electronic horizon (EH) information or information about the electronic horizon of the environment. Assumptions can be made based on this information to reduce the actual occurrence of the above-mentioned risks. For example, the probability of a pedestrian on a highway is very low, since pedestrians are not allowed on highways by nature.

[0005] However, the electronic horizon information has conceptual disadvantages. The electronic horizon cannot respond to short-term changes in the environment, for example new road works. Furthermore, the quality and reliability of the electronic horizon information can be different for different regions, in particular, it can be different in the USA compared to, for example, China. In addition, the use of an electronic horizon system leads to additional costs for the driver assistance product.

[0006] Therefore, there is a need for a solution that provides an alternative to the electronic horizon system and that excludes or at least reduces the above-mentioned risks.

[0007] Driver assistance systems known to date are disclosed, for example, in US 2016 / 0272201 Al, US 9,238,467, US 2016 / 0176397 Al, US 9,463,806 and US 2015 / 0344027 Al. In US 2016 / 0272201 Al, environmental information is acquired in real time and the set values of the control logic of the driver assistance system are changed accordingly. In document US 9,238,467, data associated with one or more events are received via an input interface, the associated risks are evaluated and a corresponding state change of the driver assistance system is implemented. In document US 2016 / 0176397 Al, alternative driving routes for autonomous driving are determined for avoiding risks arising due to oncoming objects, wherein autonomous driving continues to be safe. In document US 9,463,806, a decision signal is generated by evaluating acquired sensor signals and is further processed in an autonomous driver assistance system. In document US 2015 / 0344027 Al, an environmental observation system for detecting the environment of a driving area is implemented. The environmental detection unit comprises a radar sensor and an imaging device. SUMMARY

[0008] Against this background, it is an object of the present application to provide an advantageous method for operating a motor vehicle using a driver assistance system. It is a further object to provide a corresponding device and a corresponding advantageous motor vehicle.

[0009] These objects are achieved by the method as claimed in claim 1, the device as claimed in claim 13 and the motor vehicle as claimed in claim 15. The dependent claims contain further advantageous designs of the application.

[0010] The method for operating a motor vehicle using a driver assistance system according to the present application comprises the following steps: For at least one driver assistance system function, at least one criterion regarding the environment of the vehicle is defined for the adjustment (e.g. availability, behavior, calibration, parameterization) of the driver assistance system function. A number of features of the environment of the vehicle are detected. The probability of the occurrence of the at least one criterion is estimated based on the combination of the detected features. The driver assistance system function is then adjusted, e.g. defined as available, based on the estimated probability.

[0011] If the estimated probability is below or exceeds a threshold value (e.g. a defined threshold value), the driver assistance system function is preferably adjusted (e.g. defined as available). In particular, the adjustment can require that the driver assistance system function is defined as available and / or calibrated and / or configured (e.g. configured to be more gradual or more sensitive) and / or completely or partially deactivated.

[0012] The driver assistance system advantageously comprises at least one sensor. At least one feature of the environment of the motor vehicle can be detected by the at least one sensor of the driver assistance system. In an advantageous design of the application, ultrasonic sensors, for example parking sensors, and / or radar sensors and / or video cameras and / or lidar sensors and / or sonar sensors and / or devices for acquiring electronic horizon information can be used to detect features of the environment of the vehicle.

[0013] In other words, the application comprises an algorithm for estimating properties or features in the environment of a vehicle, in which data acquired by sensors already present in the vehicle, in particular video cameras and radar, are advantageously used. The evaluation or estimation of the data can then be used to estimate features of the vehicle environment and / or to decide on the setting of a driver assistance system function, or to set said function.

[0014] The application provides the further advantage that different features are used and appropriate feature combinations are implemented in order to estimate the probability of the occurrence of at least one criterion, for example an exclusion criterion. Thereby a more reliable and realistic estimation than would be possible by considering only a single feature or criterion can be achieved. Thus, the driver assistance system function, in particular its availability and its properties, can therefore be adapted to the respective vehicle environment more reliably than with driver assistance system functions based on the prior art.

[0015] With the method according to the application, the at least one criterion regarding the environment of the vehicle can comprise the occurrence of oncoming traffic and / or the permissibility of pedestrian traffic on the carriageway and / or the presence of pedestrians on the carriageway and / or the permissibility of two-wheeled traffic on the carriageway and / or the presence of two-wheeled vehicles on the carriageway. The above-mentioned criteria are given by way of example and further criteria are therefore likewise possible.

[0016] In the case of a criterion regarding the occurrence of oncoming traffic, the probability of oncoming traffic in an adjacent lane that is not separated from the lane of the vehicle in question is taken into account. In the case of a criterion regarding the permissibility of pedestrian traffic, the probability of pedestrian participation in traffic, i.e. for example the sharing of the carriageway by pedestrians or also the use of the area around the road or carriageway, is taken into account. This can be the case, for example, in the form of a sidewalk or a pedestrian crossing. In the case of a criterion regarding the presence of pedestrians on the carriageway, the probability of the presence of pedestrians on the road or carriageway used by the vehicle is taken into account. In the case of a criterion regarding the permissibility of two-wheeled traffic, the probability of two-wheeled vehicles sharing the road or carriageway used by the vehicle with said vehicle is taken into account. In the case of a criterion regarding the presence of two-wheeled vehicles on the carriageway, the probability of the presence of two-wheeled vehicles on the carriageway on which the vehicle is driving is taken into account.

[0017] According to the application, the number of features of the vehicle's environment can comprise the following features: the speed of the vehicle exceeding a threshold value and / or the target speed of the vehicle exceeding a threshold value and / or the presence of at least one detected pedestrian and / or the presence of at least one detected bicycle and / or the presence of at least one detected motorcycle and / or the presence of at least one detected moped and / or the presence of at least one detected scooter and / or the presence of detected oncoming traffic and / or the presence of a lateral distance to the oncoming traffic exceeding a threshold value and / or the presence of a structural division between traffic lanes and / or the presence of a road edge or road boundary or road limit and / or the presence of an occupied adjacent lane and / or the presence of a double traffic lane limit and / or the presence of a motorway sign and / or the presence of a speed limit sign.

[0018] The above-mentioned features can in principle be selected depending on the driver assistance system function of interest and / or combined with one another, which is adapted to the respective vehicle environment, for example, is intended to be permitted or excluded or changed.

[0019] For example, the at least one driver assistance system function can be a traffic jam assistant and / or a motorway assistant and / or an ACC parking and starting and / or an active high beam regulation system, or the method according to the application can comprise one of the said functions.

[0020] In a further variant, the driver assistance system function can permit hands-off driving. In this method, for example, it can be defined when or in which driving situations the function is usable or unusable. For example, a warning signal can be indicated to the driver, which informs the driver of the end of the availability of the function, for example, of the hands-off driving function. Depending on the estimated probability of the presence of at least one selected exclusion criterion, for example, the time between the indication of individual warning signals can be lengthened or shortened.

[0021] In a further variant, for example, for the adjustment of the decision regarding the availability or configuration of a traffic jam assistant and / or a motorway assistant and / or associated functions, the probability of the occurrence of oncoming traffic and / or the occurrence of a pedestrian can be defined as a criterion, in particular as an evaluation criterion. These features, for example, regarding the presence of at least one detected pedestrian and the presence of no detected oncoming traffic and the presence of a lateral distance from the vehicle to the oncoming traffic exceeding a threshold value and the presence of a structural division between traffic lanes and / or the presence of a road edge or road boundary or road limit and the presence of an occupied adjacent lane, can be combined with one another in order to determine the probability of the occurrence of one of the above-mentioned criteria or a combination of the said criteria. This is an example of a selection of features for determining the probability of the occurrence of oncoming traffic and / or the occurrence of a pedestrian. Different selections and combinations of features are also possible.

[0022] In a further advantageous variant, the probability of the occurrence of a two-wheeled vehicle and / or the occurrence of a pedestrian in front of the motor vehicle is taken into account in the method according to the application for the adjustment of the ACC parking and starting function. These features regarding the presence of at least one detected pedestrian and the presence of at least one detected bicycle and / or the presence of at least one detected motorcycle and / or moped and / or scooter and the presence of at least one detected highway sign can be combined with one another for this purpose. In addition, different selections and combinations are possible.

[0023] In a further advantageous variant, features can be combined for the adjustment of the active high beam regulation system function in order to distinguish between the environment of a motor vehicle located inside or outside of a built-up area. The criteria and features of the vehicle environment to be taken into account in this case are essentially selected with a view to making the available or excluded functions.

[0024] The advantage of the application is that a plurality of features and criteria, in particular exclusion criteria, can be flexibly selected and taken into account in each case. In this way, a plurality of situations in the vehicle environment can be combined into the respective adjustment, in particular with regard to the decision on the availability or configuration of the driver assistance system function. As a result, the safety with regard to the use of the driving assistance system can be significantly increased.

[0025] The device according to the application for adjusting at least one driver assistance function of a motor vehicle on the basis of the properties of the environment of the motor vehicle is designed to implement the previously described method according to the application. The device according to the application can comprise, for example, at least one sensor for detecting features of the environment of the vehicle, a device for estimating the probability of the presence of at least one criterion for adjusting the driver assistance system function on the basis of the detected features in combination, and a device for adjusting the driver assistance system function on the basis of the estimated probability.

[0026] The motor vehicle according to the application comprises at least one driver assistance system function and the aforementioned device according to the application. The device according to the application and the motor vehicle according to the application have essentially the same properties, features and advantages as the previously described method according to the application. In principle, the motor vehicle can be a passenger car, a truck, a motorcycle, a moped or any other vehicle.

[0027] The following describes examples of which conclusions can be drawn from the probabilities indicating specific criteria from the specific features of the environment of the vehicle. For example, it can be detected using data present in the vehicle that there is a high vehicle speed or a vehicle speed that exceeds a threshold value or a target speed of the vehicle that exceeds a threshold value. This feature can indicate that the vehicle is not located within a residential area or built-up area, but rather outside of a town or on a motorway. If the vehicle is not located in a built-up area or residential area, it can be concluded therefrom that the probability of pedestrians or slow two-wheeled vehicles being in traffic is low, in particular below a threshold value.

[0028] In a further variant, pedestrians can be detected. The detected pedestrians can indicate that the vehicle is located in a residential area or built-up area. The position of the pedestrians represents the probability of the pedestrians being involved in traffic or being located on the road. It can be concluded therefrom that the probability of pedestrians and / or two-wheeled vehicle drivers being involved in traffic is high, in particular above a threshold value.

[0029] In a further variant, bicycles have been detected. The detected bicycles can indicate that the vehicle is located in a residential area or built-up area. It can be concluded therefrom that the probability of two-wheeled vehicles being involved in traffic is high. If a bicycle has been detected, the probability of a two-wheeled vehicle being located on a traffic lane likewise increases.

[0030] A detected moped or a detected motorcycle indicates a vehicle which can or can not be associated with a specific lane but can also change lanes. If a motorcycle or moped has been detected, the probability of a two-wheeled vehicle being involved in traffic or being located on a lane increases.

[0031] In a further variant, oncoming traffic has not been detected. The absence of detected oncoming traffic in an adjacent lane indicates that the adjacent lane is not an oncoming lane or that oncoming traffic only irregularly occurs on the adjacent lane. It can be concluded therefrom that the probability of oncoming traffic being on the adjacent lane is below or falls below a threshold value. This applies in particular in the case of no oncoming traffic being detected within a specific period of time or a specific route being driven. The detection can be carried out, for example, using radar sensors or video cameras.

[0032] If oncoming traffic has been detected at a long lateral distance from the vehicle or at a lateral distance from the vehicle which exceeds a threshold value without a corresponding detection occurring at a shorter lateral distance, this indicates that the adjacent lane is not provided for oncoming traffic or that oncoming traffic rarely occurs in the adjacent lane. It can be concluded therefrom that the probability of oncoming traffic being in the adjacent lane is low, in particular in the case of oncoming traffic having been detected for a long lateral distance.

[0033] In a further variant, a structure separation or a road edge has been detected. The detection of a structure separation or a division between the lane used by the vehicle and an adjacent lane in the direction of oncoming traffic indicates that the vehicle cannot drive into the oncoming traffic. It can be concluded therefrom, in particular in the case of a detected structure division between the vehicle and the adjacent lane, that the probability of a turn into the oncoming traffic is below or falls below a threshold value.

[0034] In a further variant, the adjacent lane is occupied. The detection of traffic in the adjacent lane driving in the same direction as the vehicle indicates that there is no oncoming traffic in the adjacent lane. It can be concluded therefrom, in particular in the case of a detected traffic in the adjacent lane driving in the same direction as the vehicle, that the probability of oncoming traffic in the adjacent lane is below or falls below a threshold value.

[0035] In a further variant, a double lane marking has been detected. A double lane marking with respect to the adjacent lane indicates that the adjacent lane is provided for oncoming traffic. It can be concluded therefrom, in particular in the case of a detected double lane marking with respect to the adjacent lane, that the probability of oncoming traffic in the adjacent lane is above or exceeds a threshold value.

[0036] In a further variant, a highway sign has been detected. The highway sign indicates that there is no oncoming vehicle in the adjacent lane and no pedestrian in the traffic. It can be concluded therefrom that the probability of oncoming traffic in the adjacent lane is below or falls below a threshold value. It can furthermore be concluded that the probability of pedestrians and bicycles in the traffic is below or falls below a threshold value.

[0037] In a further variant, a speed limit sign has been detected. Depending on the indicated speed limit, the speed limit sign can indicate that the vehicle is not located in an urban area. It can be concluded therefrom, in particular in the case of a vehicle not located in an urban area, that the probability of pedestrians and slow two-wheeler drivers or two-wheeled vehicles, such as scooters or mopeds, in the traffic is below or falls below a threshold value.

[0038] For example, a camera can be used to detect the highway sign or the speed limit sign.

[0039] The above variants can be combined with each other in various ways. A certain number of features should be combined with each other in order to obtain reliable results from the sensor data. The combination of features can be implemented in different ways. It can essentially be implemented so that it adapts to the requirements of the respective function to be adapted (for example the function to be calibrated or allowed or excluded) and to the existing sensors or the desired reliability. For example, if only a radar sensor is available, the detection of an oncoming object can be combined with the detection of a crash barrier in order to achieve an estimate of the probability of oncoming traffic in the adjacent lane. In a different example, if a reliable estimate of the allowed oncoming traffic is to be implemented, the combination of the relevant signals of the detection of the structural division of the lane, the detection of oncoming traffic and the detection of the adjacent occupied lane can be implemented in order to achieve a reliable estimate of the probability of oncoming traffic being allowed in the adjacent lane. BRIEF DESCRIPTION OF DRAWINGS

[0040] Further features, properties and advantages of the present application will be described in detail below on the basis of example embodiments with reference to the accompanying drawings. All features described above and below are not only advantageous individually, but also in any combination of each other. The example embodiments described below represent, however, only examples which do not limit the subject matter of the present application.

[0041] Figure 1 A motor vehicle according to the present application is schematically shown;

[0042] Figure 2 A variant of the method according to the present application is schematically shown in the form of a flow chart;

[0043] Figure 3 A top view of a traffic lane is schematically shown;

[0044] Figures 4-13 A top view of a traffic lane is schematically shown in each case;

[0045] Figure 14 An algorithm in the form of a tree diagram is schematically shown;

[0046] Figure 15 A top view of a traffic lane is schematically shown to display individual points of the tree diagram. DETAILED DESCRIPTION

[0047] Figure 1A motor vehicle according to the application is schematically shown with an arrangement according to the application. The motor vehicle 1 according to the application comprises a driver assistance system 2 or a device for applying at least one driver assistance system function. Furthermore, it comprises at least one sensor 4 for detecting features of the environment of the motor vehicle 1. Furthermore, it comprises a device 3 for estimating a probability of the presence of at least one exclusion criterion on the basis of a combination of the detected features and a device 3 for defining the availability of the driver assistance system function on the basis of the estimated probability. In this design variant, the estimation of the probability of the presence of at least one criterion, for example an exclusion criterion, and the adjustment, for example the definition of the availability, of the driver assistance system function on the basis of the estimated probability can be implemented with one and the same device 3. On the one hand, it is also possible for there to be a device for estimating the probability and a further device for adjusting, for example for defining the availability, of the driver assistance system function.

[0048] The at least one sensor 4 can be, for example, a radar sensor and / or an ultrasonic sensor and / or a video camera and / or a lidar sensor and / or a sonar sensor and / or a device for acquiring electronic horizon information.

[0049] The driver assistance system can comprise, for example, a traffic jam assistance function and / or a motorway assistance function and / or an ACC parking and starting and / or an active high beam adjustment assistance system. The arrangement according to the application and the vehicle 1 according to the application are designed to implement the method described below.

[0050] Figure 2 A variant of the method according to the application is schematically shown in the form of a flowchart. In the method for operating a motor vehicle using a driver assistance system, at least one adjustment criterion with respect to the environment of the vehicle, for example an exclusion criterion for the availability of the function, is defined for at least one driver assistance system function in a first step 5.

[0051] A plurality of features of the environment of the vehicle are then detected in step 6. The plurality of features is preferably detected using a sensor 4. It is possible, for example, to use sensors already present in the motor vehicle.

[0052] In step 7, a probability of the occurrence of the at least one criterion is estimated on the basis of the combination of the detected features.

[0053] In step 8, the driver assistance system function is adjusted on the basis of the estimated probability. For example, the availability of the driver assistance system function can be defined on the basis of the estimated probability. The driver assistance system function is preferably adjusted if the estimated probability of the presence of the at least one criterion is below or exceeds a threshold value. In particular, the driver assistance system function can be defined as unavailable if the estimated probability of the presence of the at least one exclusion criterion exceeds a threshold value.

[0054] In step 8, the method can end after the adjustment of the driver assistance system function. Alternatively, the method can continue with step 6, i.e. it can be implemented in particular in the form of an adjustment method. This is indicated by the interrupted line arrow.

[0055] A design variant of the method according to the application is described below, in which the presence of pedestrians or two-wheeled vehicles, in particular cyclists, on the carriageway is defined as an exclusion criterion. The driver assistance system function of interest can be, for example, a function that enables hands-off driving. In this variant, the following boundary conditions can be taken into account: information about the road type is available (for example in the form of electronic horizon information), and pedestrians are detected by a camera. Further information about the road is available by means of radar and / or camera, for example about the presence of objects, obstacles or available lanes on the road.

[0056] It is then asked whether the information obtained from the electronic horizon, which indicates that the road type is, for example, a multi-lane road, sufficiently reduces the potential risk of driving without a hand on the steering wheel (hereinafter also referred to as hands-off driving). If so, does the detection of obstacles or the presence of adjacent lanes in the same direction by means of camera and / or radar sensors sufficiently reduce the risk? If not, and if pedestrians are present, an algorithm that reduces the time of hands-off driving should be implemented. If pedestrians have been detected in the vehicle's travel path, this can require that no extended hands-off driving time is available for the next 200 meters (m).

[0057] In a first step, extended hands-off driving is available if there is no presence of pedestrians in the vicinity of the lane being used. However, the extended hands-off driving time, for example 120 seconds (s), can be longer than the time that sensors are able to predict. Therefore, information deduced from previous measurements must be used to infer the presence or absence of pedestrians. In traffic situations involving pedestrians, the past has been more related to the distance travelled by the vehicle than to the driving time interval. Since the assumptions about pedestrians are mainly related to urban or built-up areas, the decisive distance should be related to the urban structure (for example, buildings and other structures). For example, the length of a long building block (for example 200 m) can be used as a measure.

[0058] In a second step, extended hands-off driving is available if there is no presence of pedestrians in the vicinity of the lane used by the vehicle in the last 200 m of the travel route. However, the algorithm described previously is not the only indication that allows extended hands-off driving time. It involves an input signal that represents a reduction in hands-off driving in the area used by pedestrians. Therefore, the interface of the algorithm is "pedestrian presence" and the conditions must be redefined.

[0059] In a third step, it is determined whether the vehicle is located in an area used by pedestrians. This is the case if there are pedestrians in the vicinity of the lane within the last 200 m of the driving route. The term "in the vicinity of the lane" can be specified individually. The driver should essentially learn not to let go of the steering in areas where pedestrians and vehicles are or can be on the road or on the sidewalk.

[0060] In a fourth step, it is determined whether pedestrians are present. This is the case, for example, if a pedestrian within the last 200 m of the driving route has been observed laterally or at a distance of less than 15 m in the lateral direction and in the direction of travel or in the longitudinal direction in front of the vehicle at a distance of less than 60 m. The indicated distances are merely examples and can also be defined differently.

[0061] A design variant is described below which involves a feature in which the adjacent lane is occupied. The following definitions apply to the respective algorithm: There is no oncoming traffic in the adjacent lane if the lane is occupied by a vehicle which is stationary or driving in the same direction as the vehicle of interest, i.e. the vehicle using the method according to the invention. The adjacent lane is the lane next to the lane used by the vehicle of interest, for example to the left of the lane used by right-hand traffic or to the right of the lane used by left-hand traffic. Occupied means that the lane is occupied in such a way that oncoming traffic cannot adapt to the gap between the two vehicles or the detected vehicle and the vehicle of interest. The same direction means that the angle between the speed vector of the vehicle of interest and the speed vector of the target vehicle is less than 20°, i.e. preferably 0° ± 10°.

[0062] With regard to the "occupied" feature, it is basically assumed that there is no oncoming traffic as soon as the observed vehicle is driving in the same direction as the vehicle of interest or a stationary target is observed in the adjacent lane. However, this assumption does not apply to motorcycles or scooters, which are usually narrower than vehicles and can change between the different lanes highly dynamically.

[0063] Figure 3 A top view of a driving road is shown. The vehicle of interest 10 uses the lane 11. The adjacent lane 12 is considered to be occupied and thus free of oncoming traffic, provided that only vehicles driving in the same direction as the vehicle of interest 10 are present on it. In Figure 3 This is the case for the vehicle 13 in Figure 3 The oncoming vehicle (e.g.

[0064] In terms of geometry, the considered area of the object in the adjacent lane is defined as follows: The minimum width (min) (i.e. the relevant distance in lateral direction) can be defined as the sum of half the vehicle width plus the distance from the vehicle of interest to the lane marking plus the distance from the observed object to the lane marking, e.g. min = 1 m + 0.5 m + 0.5 m. The maximum width (max) can be defined as the sum of half the vehicle width plus the distance from the vehicle to the lane marking plus the distance from the observed object to the lane marking plus the minimum lane width, e.g. max = 1 m + 0.5 m + 3.05 m = 4.55 m.

[0065] In terms of length, i.e. the relevant distance in longitudinal direction, a reliable detection distance can be defined. Since the vehicle of interest and the observed vehicle or other observed objects do not necessarily drive straight ahead, the lateral distance dy is preferably compared to the intersection with the trajectory 46 of the vehicle of interest. It is usually not necessary to implement a detection beyond a certain distance in front of the vehicle, since the view is blocked by the vehicle driving in front, e.g. in case of a traffic jam. The length should be configured in relation to false positive detections on a curve.

[0066] The geometric definition is shown in Figure 4 and 5 . Reference sign 15 denotes the distance from the vehicle of interest to the lane marking. Reference sign 16 denotes the distance from the observed vehicle to the lane marking. Reference sign 17 denotes the minimum lane width. Reference sign 18 denotes the distance in driving direction or in longitudinal direction for reliable detection.

[0067] If there is an object with an angle to the lane of less than e.g. 10° and a longitudinal distance of less than 100 m and a lateral distance between 1.5 m and 4.55 m from the trajectory, the adjacent lane can be considered as occupied.

[0068] After the last relevant observed object has disappeared, the occupied lane is preferably observed or evaluated additionally within 1.6 seconds and 30 m. It is known that a vehicle maneuver occurring within a time period of less than 1.6 seconds can be considered as sporty. Assuming that oncoming traffic has almost no probability of sporty or dangerous evasive maneuvers, it can be assumed that oncoming traffic will not appear within at least 1.6 seconds after the last relevant object has been detected which occupied the lane. The resulting two cases are shown in Figure 6 and 7 . Reference sign 19 denotes the route which can be covered within more than 1.6 seconds. Possible evasive operations are denoted by reference sign 20. In the case shown in Figure 6 , the observed vehicle 13 driving in the same direction as the vehicle of interest changes lane. In the case shown in Figure 7In the illustrated variant, an oncoming vehicle changes lanes.

[0069] The concept of an algorithm involving the feature that there is no oncoming traffic in an adjacent lane is described below. It is based on the concept that extended hands-free driving should be reduced when oncoming traffic is allowed in an adjacent lane. The following assumptions can be made here: If there is no oncoming traffic in an adjacent lane within a specific time period or a specific distance or driving route, it is unlikely that there will be regular oncoming vehicles, at least until the next oncoming traffic has been detected. Such an assumption can be made because the absence of oncoming traffic in an adjacent lane implies that oncoming traffic is not allowed in that lane, such as in the case of a highway, or for example when driving at night on a rural road where there is only irregular oncoming traffic.

[0070] The following definition can be made for the algorithm: If a motor vehicle has been detected that is traveling in the opposite direction to the vehicle of interest and at a lateral distance less than the checkable distance, then oncoming traffic has been detected. A motor vehicle is a car, such as a passenger car, motorcycle or truck. The checkable lateral distance is the distance between the vehicle of interest and the observed vehicle, which is less than twice the standard lane width. Traveling means that the vehicle is moving at a speed greater than 10 km / h. The opposite direction means that the angle between the speed vector of the vehicle of interest and the speed factor of the observed vehicle in a plane (2d) is ±10° from 180°.

[0071] It can be concluded that if an oncoming vehicle has been detected at a lateral distance less than two lane widths from the vehicle of interest, then oncoming traffic has been observed. This is shown in Figure 8 in different variants. Here it is assumed that the lane width is considered as the overtaking space for oncoming traffic. This is represented by reference numeral 20. An additional lane is considered as the checkable space between the vehicle of interest and a potential overtaking vehicle and is represented by reference numeral 21. Figure 8 The situation marked by the circle 22 in <0000​​​

[0073] Figure 10 Alternative or other variations are shown. In this case, the intersection is defined between the extended frontal line 26 of the vehicle of interest 10 in the lateral direction and the frontal direction vector 25 of the observed vehicle 13. This intersection should be located at a distance of less than 6 m from the vehicle of interest 10 in the lateral direction.

[0074] Figure 10 The variant shown offers the advantage that early detection is possible and the results indicate that the risk also practically predicts the location where it will occur. However, a drawback is that the potential error is caused by the fact that the observed vehicle's path can only be defined by substantial error. If the observed vehicle is traveling on a curve, this will result in an additional range of dy values. Furthermore, a moderately complex implementation is involved.

[0075] exist Figure 11 Another method is illustrated schematically. Here, the minimum lateral distance 29 between trajectory 28, or the observed path of vehicle 13, and trajectory 27 of vehicle 10 of interest is examined. This distance 29 should be less than 6m (minimum (SPP, target path) < 6m). The advantage of this method is that, compared with... Figure 10 Compared to the variants shown, earlier detection is possible and fewer false detections occur. However, a drawback is the highly complex methodology involved, particularly in defining the paths of the observed vehicles.

[0076] The objective is essentially to not only observe the individual occurrences of oncoming traffic, but also to define the overall probability of exclusion criteria for the availability of driver assistance system functions (e.g., hands-free driving). Early detection is challenging given the increasing uncertainty in predicting the paths of the vehicle of interest and the observed vehicles with increasing distance. Early detection of oncoming traffic cannot be guaranteed because the field of vision may be obstructed by the observed vehicle traveling in front of the vehicle of interest. Therefore, early detection only applies to objects not far from the vehicle of interest. After a positive detection result, oncoming traffic can be observed, for example, for 400m or 30 seconds. Other values ​​can be set similarly. If no oncoming traffic is observed within this time or distance, oncoming traffic is not permitted or only occurs very irregularly, thereby reducing the probability of its occurrence. Because the above values ​​are set relatively high, the signal can be reset accordingly if oncoming traffic is detected.

[0077] This is for example in Figure 12The distance 30 represents a distance of 400 m or a distance travelled within 30 seconds. In the diagram shown below, an oncoming traffic has been detected at point 31. For the distance 30, i.e. up to point 32, it is thus assumed that oncoming traffic occurs. If no further detection takes place within this time period, i.e. during the driving of the distance 30, the reset signal is triggered.

[0078] A design variant is described below in connection with the traffic jam assist system. Initially, a distinction can be made between three cases, namely 1. a case in which the adjacent lane is separated from the lane of interest, 2. a case in which no oncoming traffic occurs in the adjacent lane, or 3. a case in which the adjacent lane is occupied. If one of these cases occurs, the traffic jam assist system can be used, for example, or parameters can be adjusted, in particular, hands-off driving can be permitted. The three driving cases described above can be characterised by three separate algorithms. Each of the road cases described above excludes the occurrence of oncoming traffic. Considering only these three driving cases represents a quick and advantageous solution. Furthermore, this is characterised by high reliability, since there are three possibilities of exclusion functions. This approach is robust, in which, for example, by means of filters or statistics, a minimum spacing of 30 m length can be observed in each case, vehicles driving in the same lane in the same direction can be observed frequently or over a long period of time, and close oncoming traffic is not hidden by the separation between the lanes, i.e. is clearly visible. In the last-mentioned case, the detection probability is thus increased by a factor of two to three and thus almost reaches 100%.

[0079] A design variant is explained in detail below, which takes into account features in which a structural division between the traffic lanes exists or is detected. The decision regarding the permission of an extended hands-off driving time requires the detection of an environment or traffic situation which excludes or indicates a high improbability of difficult traffic situations, such as oncoming traffic or pedestrians. If a structural division exists between the lanes of interest and the structural division cannot be adapted to oncoming traffic between the lanes of interest and the vehicle of interest, the occurrence of oncoming traffic in the adjacent lane can be excluded. For example, a radar-camera sensor system can be used to detect the respective structure. For example, visible road edges or road boundaries or crash barriers can be detected with optical devices or using radar.

[0080] A model is defined below to decide whether oncoming traffic is possible or not possible in the adjacent lane. Here, a distinction can be made between the following cases: 1. double lane marking and obstacles, 2. left lane marking and obstacles, 3. right lane marking and obstacles, 4. no lane marking and obstacles, 5. no obstacles, 6. no lane marking and obstacles.

[0081] Further, it can be defined that adjacent lanes are separated if there is a structural division between the adjacent lane and the lane used by the vehicle of interest. An obstacle has to have a certain length in order to be considered a structural division. For example, a pedestrian island or the like does not represent a structural division. A pedestrian island is typically about 10 m long. There is no limited length for a structural division, so the minimum length is defined for example using the length of a pedestrian island. Three times the length of a pedestrian island, i.e. 3x10 m = 30 m, can preferably be assumed as the minimum length of a structural division.

[0082] A small gap, for example a passage in a structural division or an intersection, does not mean that adjacent lanes are no longer structurally separated. The following assumptions are made in order to define this type of bridging passage. A turning lane (U-turn) does not define the end of a structural division. An intersection does not end a structural division, especially if the division of the lanes remains outside the intersection. In order to detect multiple relevant intersections, five lanes in each direction are assumed. The minimum lane width is assumed to be 3.05 m (see https: / / de.wikipedia.org / wiki / Stra%C3%B6nenquerschnitt). This corresponds to a standard lane width. If no separation or division is also provided at the edge of the road, 5 m can be added per division. The width of a division in which pedestrians can be present is at least 2 m. Thus, the intersection length is for example 10x3.05 m + 2x5 m + 2 m, i.e. a total of 42.5 m.

[0083] Reference is made below to Figure 13 The algorithm filter for lane splitting is explained in detail. The x-axis represents the road and the y-axis represents that adjacent lanes are not separated at 0 and that adjacent lanes are separated, i.e. are structurally separated from each other, at 1. It is assumed here that adjacent lanes are structurally separated from each other if the length of a structural division, for example in the direction of travel, is greater than the minimum length of a structural division. For example, the minimum length of a structural division can be defined as a value of 30 m. Furthermore, adjacent lanes are considered not to be structurally separated by a structural division if the length of a lane without a structural division is greater than the minimum length of a structural division that is not present. The minimum length of a non-structural division can be defined separately. It is advantageously defined as a typical intersection length, for example a value of 42 m.

[0084] Figure 13The road shown in Fig. 1 comprises a carriageway extending in the x-direction, which carriageway has two lanes 41 and 42 and an intersection 40. The lanes 41 and 42 are separated from each other by structure divisions 33, 34 and 35 over three sections of the road. The distance in the x-direction between the first structure division 33 and the second structure division 34 is indicated by reference 36. This distance is less than 42 m. A gap in the x-direction is located between the second structure division 34 and the third structure division 35. This is greater in length than 42 m. The length of 42 m is indicated by reference 37. Reference 38 indicates a length of 30 m in the x-direction from the third structure division 35.

[0085] Figure 13 The following figure in Fig. 2 shows the results produced by the algorithmic filter at various positions in the x-direction by means of the curve 43. Starting on the left, the lanes 41 and 42 are initially considered to be structurally separated. The distance between the first structure division 33 and the second structure division 34, indicated by reference 36, is too small to interrupt the structure division. At the end of the second structure division 34, the defined intersection length, indicated by reference 37, is first increased. As a result, the structure division between the adjacent lanes 41 and 42 is first assumed not to be at the x value indicated by reference 44. The minimum length that has been defined for the structure division, which is indicated by reference 38 and is currently 30 m, is first taken into account from the start of the third structure division 35. The structure division of the carriageway is then assumed to start from the point on the x axis indicated by reference 45.

[0086] The method of the algorithm as a tree diagram will be explained in detail below with reference to Figure 14 Fig. 3. Figure 15 Each point of the tree diagram is shown in

[0087] As a base or root of the tree diagram 50, a decision will be made as to whether oncoming traffic is possible. Right-hand traffic is shown here. Left-hand traffic can be processed in a similar manner, with the corresponding changes relating to the indication of the side. At the next level, it is investigated whether a road edge has been detected. At leaf 51, no road edge has been detected. At leaf 52, a road edge has been detected. If, as at leaf 51, no road edge has been detected, it is investigated whether the adjacent lane is occupied. At leaf 54, it is occupied, at leaf 53 it is not occupied. If it is not occupied at leaf 53, oncoming traffic is possible. If it is occupied at leaf 54, oncoming traffic is not possible. The result that oncoming traffic is not possible is indicated in Fig. 4 by a tick. The result that oncoming traffic is possible is indicated in Fig. 5 by a cross or x. Figure 14 Figure 14

[0088] ​​If a road edge has been detected at leaf 52, it is investigated whether and which lane markings are present. At leaf 55, no lane markings have been observed, at leaf 56 only a right lane marking has been detected, at leaf 57 only a left lane marking has been detected, and at leaf 58 it has been detected that two lane markings are present.

[0089] If no lane markings are present, it is investigated whether the distance from the road edge is greater or less than the vehicle width. At leaf 59, the distance from the road edge is less than the vehicle width, so that oncoming traffic is not possible. At leaf 60, the distance from the road edge is greater than the vehicle width. If this is the case, it is investigated whether a second road edge has been detected. At leaf 60, this is the case. At leaf 62, no second road edge has been detected, so that oncoming traffic is possible. If a second road edge is detected at leaf 61, it is investigated whether the distance between the road edges is greater or less than twice the minimum lane width. At leaf 63, the distance between the road edges is less than twice the lane width, so that oncoming traffic is not possible. At leaf 64, the distance between the road edges is greater than twice the lane width, so that oncoming traffic is possible.

[0090] After leaf 56, it is investigated whether the distance between the detected right lane marking and the road edge is greater or less than twice the minimum lane width. At leaf 66, this distance is less than twice the lane width, so that oncoming traffic is not possible. At leaf 65, this distance is greater than twice the lane width. In this case, it is investigated whether the distance from the road edge is greater or less than the vehicle width. At leaf 67, the distance from the road edge is less than the vehicle width, so that oncoming traffic is not possible. At leaf 68, the distance from the road edge is greater than the vehicle width. Oncoming traffic is thus possible.

[0091] If only a left lane marking has been detected at leaf 57, it is investigated whether a second road edge has been detected. At leaf 69, no second road edge has been detected. At leaf 70, a second road edge has been detected. After leaf 69, it is investigated whether the distance from the left lane marking to the road edge is greater or less than the lane width. At leaf 71, the distance between the left lane marking and the road edge is less than the lane width. At leaf 72, the distance from the left lane marking to the road edge is greater than the lane width, so that oncoming traffic is possible.

[0092] After leaf 71, it is investigated whether the distance from the road edge is greater or less than the vehicle width. At leaf 74, the distance from the road edge is less than the vehicle width, so that oncoming traffic is not possible. At leaf 73, the distance from the road edge is greater than the vehicle width. It is then investigated whether the distance from the road edge is greater or less than the lane width. At leaf 76, the distance from the road edge is greater than the lane width, so that oncoming traffic is possible.

[0093] At leaf 75, the distance from the road edge is less than the lane width. In this case, it is then investigated whether the vehicle of interest is driving up to the lane or at a distance from the lane. This decision can be made by a threshold distance. At leaf 77, the vehicle is driving up to the lane marking, so oncoming traffic is not possible. At leaf 78, the vehicle is driving at a distance from the lane marking, so oncoming traffic is possible.

[0094] After leaf 70, it is investigated whether the distance between the two road edges is greater or less than twice the minimum lane width. At leaf 79, the distance between the road edges is less than twice the minimum lane width, so oncoming traffic is not possible. At leaf 80, the distance between the road edges is greater than twice the minimum lane width. In this case, the decision method continues at leaf 56.

[0095] After leaf 58 (i.e., detection of two lane markings), it is investigated whether the lane width is correct. At leaf 81, the lane width is not correct. At leaf 82, the lane width is correct. After leaf 82, it is checked whether the distance from the left lane marking to the right road edge is greater than the lane width. At leaf 90, the distance between the left lane marking and the road edge is less than the lane width, so oncoming traffic is not possible. At leaf 91, the distance from the left lane marking to the road edge is greater than the lane width, so oncoming traffic is possible.

[0096] After leaf 81 (i.e., in the case of an incorrect lane width), it is investigated whether the lane is too narrow or too wide. At leaf 83, the lane is too narrow, in which case the method continues at leaf 57. At leaf 84, the lane is too wide. In this case, it is then checked whether the vehicle is driving up to one of the lane markings. At leaf 86, the vehicle is driving up to the left lane marking, in which case the method continues at leaf 56. At leaf 87, the vehicle is driving up to the left lane marking. In this case, the method continues at leaf 57.

[0097] At leaf 85, the vehicle is not driving up to a lane marking. In this case, it is checked whether the distance from the road edge is greater or less than the vehicle width. At leaf 88, the distance is less than the vehicle width, in which case oncoming traffic is not possible. At leaf 89, the distance from the road edge is greater than the vehicle width, in which case oncoming traffic is possible.

[0098] Figure 15 It is shown in Figure 14Each driving situation appearing in the tree diagram is illustrated. The top row relates to the situation where both lane markings are present. The second row relates to the situation where only the left lane marking is present. The third row relates to the situation where only the right lane marking is present, and the fourth row relates to the situation where no lane markings are present. The minimum lane width is denoted as min LW, and a safety margin of 0.3 m can be defined as 2.6 m - 0.3 m. The maximum lane width is denoted as max LW, and is defined as 4.6 m in the following examples. The actual lane width is denoted as ego LW. It corresponds to the measured lane width minus the safety margin of 0.3 m. For example, a width of 1.8 m can be defined as the vehicle width.

[0099] The relevant or opposite side depends in principle on the traffic direction, i.e. whether right-hand traffic or left-hand traffic is involved. The following applies in the case of right-hand traffic: the relevant side is the right side, and the opposite side is the left side (relevant = right, opposite = left). In the case of left-hand traffic, the relevant side is specified as the left side, and the opposite side is specified as the right side (relevant = left, opposite = right).

[0100] The lane width LW is between 2.6 and 4.4 m (2.6 m

Claims

1. Method for operating a motor vehicle (1) using a driver assistance system (2), characterized in that the method comprises the following steps: - for at least one driver assistance system function (8), defining at least one criterion (5) about the environment of the vehicle for the adjustment of the driver assistance system function (8), wherein the at least one driver assistance system function (8) comprises a traffic jam assistance function, a motorway assistance function, an ACC parking and starting function, an active high beam regulation system assistance function and a hands-off driving permission; - detecting a plurality of different features (6) of the environment of the vehicle, - estimating a probability (7) of the occurrence of the at least one criterion (5) based on a combination of the detected features (6), and - adjusting the driver assistance system function (8) based on the estimated probability (7); wherein the driver assistance system function (8) is further configured to control that the hands-off driving is not available in one of a certain time and a certain distance in response to the probability (7) of the occurrence of the at least one criterion (5); and indicating a warning signal to the driver, which informs the driver of the end of the availability of the hands-off driving, and modifying the time interval between the warning signal and a subsequent warning signal depending on the estimated probability (7).

2. Method according to claim 1, characterized in that the driver assistance system function (8) is adjusted in case the estimated probability (7) is below or exceeds a threshold value.

3. Method according to claim 1, characterized in that the adjustment requires that the driver assistance system function (8) is defined as available and / or calibrated and / or configured and / or fully or partially deactivated.

4. Method according to claim 1, characterized in that the driver assistance system (2) comprises at least one sensor (4) and at least one of the features (6) of the environment of the motor vehicle is detected by the sensor (4) of the driver assistance system.

5. Method according to claim 4, characterized in that an ultrasound sensor and / or a radar sensor and / or a video camera and / or a lidar sensor and / or a sonar sensor and / or a device for acquiring electronic horizon information are used for detecting the features (6) of the environment of the vehicle.

6. Method according to claim 1, characterized in that the at least one criterion (5) comprises: - the occurrence of oncoming traffic, and / or - the presence of pedestrian traffic on a traffic lane, and / or - the presence of a pedestrian on the traffic lane, and / or - the presence of two-wheeled traffic on the traffic lane, and / or - the presence of a two-wheeled vehicle on the traffic lane.

7. Method according to claim 1, characterized in that the plurality of different features (6) of the environment of the vehicle comprises: - a speed of the vehicle that exceeds a threshold value, and / or - a target speed of the vehicle that exceeds a threshold value, and / or - the presence of at least one detected pedestrian, and / or - the presence of at least one detected bicycle, and / or - there is at least one detected motorcycle, and / or - there is detected oncoming traffic, and / or - there is a lateral distance from the vehicle to the oncoming traffic that exceeds a threshold value, and / or - there is a structural division between traffic lanes, and / or - there is a road edge, and / or - there is an occupied adjacent lane, and / or - there is a double traffic lane boundary, and / or - there is a detected highway sign, and / or - there is a detected speed limit sign.

8. The method as claimed in claim 1, characterized in that a warning signal is indicated to the driver, which informs the driver of the end of the hands-off driving function availability, and in the method, the time between the indication of individual warning signals is extended.

9. The method as claimed in claim 1, characterized in that for the adaptation of a traffic jam assistant function and / or a highway assistant function, the probability (7) of the occurrence of oncoming traffic and / or the occurrence of a pedestrian is taken into account, the following features (6) are combined with one another: - there is at least one detected pedestrian, - there is no detected oncoming traffic, - there is a lateral distance from the vehicle to the oncoming traffic that exceeds a threshold value, - there is a structural division between traffic lanes and / or - there is a road edge, and - there is an occupied adjacent lane.

10. The method as claimed in claim 1, characterized in that for the adaptation of an ACC parking and starting function, the probability (7) of the occurrence of a two-wheeled vehicle in front of the motor vehicle and / or the occurrence of a pedestrian is taken into account, the following features (6) are combined with one another: - there is at least one detected pedestrian, - there is at least one detected bicycle, - there is at least one detected motorcycle and / or scooter, and - there is at least one detected highway sign.

11. The method of claim 7 or 10, wherein, The motorcycle comprises a moped or a small motorbike.

12. The method as claimed in claim 1, characterized in that for the adaptation of an active high beam regulation system assistant function, the features (6) are combined in order to distinguish between the environment of the motor vehicle located inside or outside of a built-up area.

13. A device for adapting at least one driver assistance system function (8) of a motor vehicle (1) based on characteristics of the environment of the motor vehicle (1), characterized in that the device is designed to implement the method as claimed in any one of claims 1 to 12.

14. The device as claimed in claim 13, characterized in that the device comprises: - at least one sensor (4) for detecting features (6) of the environment of the motor vehicle, - a device (3) for estimating a probability (7) of the occurrence of at least one criterion (5) for the adaptation of the driver assistance system function (8) based on a combination of the detected features (6), and - a device (3) for adapting the driver assistance system function (8) based on the estimated probability (7).

15. A motor vehicle (1) comprising at least one driver assistance system function (8) and an arrangement according to any one of claims 13 or 14.

Images