The invention relates to a driver assistance system for a vehicle, also called ego vehicle in the following, for predicting a lane area ahead of the ego vehicle to ascertain whether an object, like a vehicle preceding the ego vehicle, is located in the predicted lane area. Therein, the driver assistance system comprises at least one environmental sensor for capturing lane markings of a roadway traveled with the ego vehicle, at least one capturing means for capturing at least two ego motion parameters of the ego vehicle, and a control device, which is configured to determine a lane area ahead of the ego vehicle depending on the at least two ego motion parameters of the ego vehicle according to a determination method. Furthermore, the invention also relates to a vehicle with such a driver assistance system, as well as to a method for predicting a lane area ahead of an ego vehicle to ascertain whether an object is in the predicted lane area.

Diverse driver assistance systems are known from the prior art. Therein, such driver assistance systems can have very different automation levels, starting from level 1 , such as for example the ACC (Adaptive Cruise Control), up to level 5, fully autonomous driving. For implementing these driving functions assisting the driver or also fully autonomous driving functions, it is required that such systems are able to detect objects, obstacles or other traffic participants ahead. In addition, it is also important in most of the systems to estimate whether such objects or obstacles are located in the lane or the driving area presumably to be traveled by the ego vehicle, for example to adapt the speed and/or to maintain a certain safety distance to a preceding vehicle. For example, if a preceding vehicle is detected, but it is not in the lane of the ego vehicle, thus, it either is not required to warn the driver or to automatically decelerate the vehicle in such a case. Therefore, it is required to not only ascertain whether objects ahead such as for example preceding vehicles are basically present, but also to ascertain as reliably as possible whether or not a vehicle preceding the ego vehicle is in the lane area to be traveled by the ego vehicle. This operation is also referred to as "target selection". Therein, the target can represent any object ahead of the ego vehicle. Thus, if such a "target" is selected, this means that this target is located in the lane area to be traveled by the ego vehicle. If no target is selected, thus, preceding traffic participants are not located on the lane to be traveled by the ego vehicle. If there are several vehicles located on the lane to be traveled by the ego vehicle, then, ideally, only that one of the preceding vehicles is selected as target, that has the shortest time to collision, which can be determined based on the position, speed and acceleration levels of the preceding vehicles as well as of the ego vehicle.

In order to be able to perform such a classification, the lane area to be traveled or probably to be traveled has again to be determined or predicted. For such a prediction, it is further known to use ego motion parameters of the ego vehicle. They for example represent the current speed, steering angle turn or similar parameters, which describe the current ego motion of the ego vehicle. Based on the current ego motion of the ego vehicle, the lane area presumably traveled by the ego vehicle can thereby be predicted at least for a certain distance.

A generic driver assistance system is for example disclosed in US 8,577,552 B1 . Herein, a lane area to be traveled in future can be predicted based on ego motion parameters of an ego vehicle and it can be ascertained based thereon whether or not a vehicle preceding the ego vehicle is in this lane area. Therein, for capturing and positioning the preceding vehicle, environmental sensors of the ego vehicle can be used. Furthermore, it is disclosed that camera-based systems can be used for lane marking recognition to provide warnings upon departure from the lane.

US 201 1 /0270466 A1 also describes determining a predicted lane area based on ego motion parameters of the ego vehicle.

However, the prediction of the lane area ahead of the ego vehicle based on ego motion parameters has the great disadvantage that such a prediction is sometimes only reliable for very short distances. Especially in a curvy route, reliable prediction is restricted to an extremely short area in front of the ego vehicle. For example, if the ego vehicle travels straight ahead towards a right-hand bend on a right lane, thus, the predicted lane area would extend straight based on the ego motion parameters in this situation such that for example traffic participants located on the left lane or already located in the right-hand bend would erroneously be recognized as directly preceding vehicles. In contrast, actually preceding vehicles located on the same lane, which are already in the right-hand bend, either would not be captured as target objects.

Therefore, it is the object of the present invention to specify a driver assistance system for a vehicle for predicting a lane area ahead of the vehicle, a vehicle and a method, by means of which a more reliable prediction of the lane area ahead is allowed even for larger distances to the vehicle.

This object is solved by a driver assistance system, a vehicle and a method having the features according to the respective independent claims. Advantageous configurations of the invention are the subject matter of the dependent claims, of the description as well as of the figures.

A driver assistance system according to the invention for an ego vehicle for predicting a lane area ahead of the ego vehicle to ascertain whether an object, like a vehicle preceding the ego vehicle, is in the predicted lane area, comprises at least one environmental sensor for capturing lane markings of a roadway traveled with the ego vehicle, at least one capturing means for capturing at least two ego motion parameters of the ego vehicle and a control device, which is configured to determine a first lane area ahead of the ego vehicle depending on the at least two ego motion parameters of the ego vehicle according to a first determination method. Furthermore, the control device is configured to determine a second lane area ahead of the ego vehicle depending on sensor data of the at least one environmental sensor, which relates to the capture of the lane markings according to a second determination method and to predict the lane area ahead depending on the first lane area and/or the second lane area.

In addition to the first method, according to which the first lane area is determined based on at least two ego motion parameters of the vehicle, a second method for determining the second lane area can also be advantageously used according to the invention, according to which this second lane area ahead is determined based on the captured lane markings. By the possibility of combination of these two determination methods, the accuracy and reliability of the prediction of the driving area ahead can be considerably increased. Hereby, the driving area to be traveled can in particular be considerably more accurately predicted also for route sections further ahead. Also in curve situations, for which the lane area to be traveled can only be very inaccurately predicted or only for a very short distance with respect to the ego vehicle alone according to the first method based on the ego motion of the vehicle, now, the lane area ahead can be considerably more accurately and more long-range described by capturing the lane markings. In the prediction of driving areas ahead by environmental capture, there is basically the problem that an environmental capture per se can sometimes be very inaccurate and can therefore result in a very unreliable prediction of the driving area ahead. For example, if cameras are employed for lane marking recognition, thus, it can for example be the case that lane markings can be very poorly recognized in bad weather conditions such as for example rain. It can also occur that lane markings are basically poorly recognizable or are also partially not present at all on roads. Lane markings can also be occulted by other road users near in front of the ego vehicle or on sides on the ego vehicle. In addition, environmental sensors can be soiled, which impairs their capturing accuracy. Glaring effects can also occur, for example by direct solar radiation or diverse reflections such that the environmental capture and accordingly also results derived therefrom like a predicted driving area ahead are more or less accurate according to situation. However, the invention exploits the realization that especially by a combination of these two methods, that is the prediction of the first lane area based on at least two ego motion parameters as well as the prediction of a second lane area based on the environmental capture, in particular the capture of the lane markings, the possibility of utilizing especially the strengths of these respective methods in particularly advantageous manner and therein compensating the weaknesses thereof at the same time is provided. Hereby, enormous advantages both with respect to accuracy and with respect to range can be

advantageously achieved. By the possibility of combination of the first determination method and the second determination method provided by the invention, a considerably better adaptation to situation can be achieved. For example, the lane area to be predicted can be considerably more accurately described by the second determination method even in large distances to the ego vehicle and the advantages of the first method with regard to very short distances to the ego vehicle can further be utilized. Thus, the lane area to be traveled can be considerably more accurately predicted even on very curvy route courses.

The object can be any arbitrary kind of object, like a moving object or also a static object. For example the object can be a vehicle preceding the ego vehicle, like a car, a motorbike or a truck, bicycle, or also a pedestrian, an animal, or any obstacle on the road, like an object fallen from a truck onto the road. Especially, the object preferably is an object in front of the ego vehicle, especially with regard to the vehicle front or the moving direction of the ego vehicle.

Generally, the ego vehicle can also be any arbitrary kind of vehicle, like a car, especially a passenger car, a motorbike, a truck and so on. The term "ego" vehicle is used only for more easily distinguish this vehicle from objects, like one or more other vehicles preceding the ego vehicle.

Further, the driver assistance system can comprise one or also more environmental sensors such as for example a camera, a ToF (Time of Flight) camera, a stereo camera or the like to capture lane markings. Other environmental sensors such as for example radars, ultrasonic sensors, laser scanners or the like can also be provided. By such sensors, roadway boundaries, guardrails, edge developments and so on can for example also be captured, which can be used for determining the lane area, in particular the second lane area. In capturing the lane markings, both lane markings extending to the left of the ego vehicle in direction of travel and lane markings extending to the right of the ego vehicle can be captured and used for determining the second lane course. For example, if only lane markings on one side of the ego vehicle can be captured because they are for example only present on one side, thus, a predetermined standard width can be assumed for calculating the lane course, in particular its width perpendicular to the direction of extension.

For determining the first lane area based on the at least two ego motion parameters, multiple ego motion parameters are preferably used, such as for example the current speed of the ego vehicle, the current steering angle, a yaw angle or a yaw rate, a current acceleration, acceleration variation and so on. Based on one or more of these

parameters, the first lane area can be calculated by extrapolation of the ego motion described by these parameters, wherein a standard value or else the width of the ego vehicle, optionally plus a lateral, fixed value, can be again assumed for the width thereof in direction of extension. Preferably, at least one of the at least two ego motion

parameters is a parameter describing a longitudinal behavior of the ego vehicle, like the ego vehicle's longitudinal speed, and at least one of the at least two ego motion parameters is a parameter describing a lateral behavior of the ego vehicle, like the ego vehicle's yaw rate or the front wheel angle. Especially, the ego motion parameters used for determining the first lane area are linked to each other by a dynamic model of the ego vehicle. Moreover, the more ego motion parameters are used, the more precisely will be the prediction.

In an advantageous configuration of the invention, the control device is configured to calculate the lane area ahead as a predeterminable averaging of the first lane area and the second lane area for predicting the lane area ahead depending on the first lane area and the second lane area. Hereby, both the results of the first determination method and those of the second determination method can be advantageously taken into account and advantageously combined in predicting the lane area. Therein, such an averaging does not necessarily have to consider the results, i.e. the first lane area and the second lane area, of these two determination methods in the same manner, but for example also in weighted manner such that it is allowed hereby to also specifically exploit the strengths of the respective determination methods and to compensate for the weaknesses thereof. The predeterminable averaging can also be computed dynamically, for example as a function of the speed of the ego vehicle, as a function of the sensor dynamic performance, as a function of infrastructure environment, and so on, and therefore is predeterminable in dependency of a parameter, like the speed, performance, and so on.

Therein, it is particularly advantageous if the control device is configured to ascertain at least one first confidence value for the first lane area and to ascertain at least one second confidence value for the second lane area. Furthermore, the control device is preferably configured to calculate the predeterminable averaging as an averaging of the first lane area and the second lane area weighted depending on the at least one first confidence value and the at least one second confidence value.

By the respective confidence values, the reliability of the individual calculated lane areas, i.e. the first and the second lane area, can advantageously be evaluated. Thus, the respective first and second lane area can be incorporated in the averaging

correspondingly weighted by their confidence. Hereby, it is advantageously allowed to maximize the overall confidence of the finally predicted lane area. By such a weighting, thus, it can be particularly advantageously realized that the strengths and advantages of the respective methods are maximally exploited in different situations and the weaknesses thereof are deleted.

Therein, the concerned confidence values indicate an estimated value to the effect, to which extent the respective calculated first lane area and second lane area coincide with a lane area actually to be traveled. With respect to the second determination method, in which the second lane area is calculated based on the sensor data, this confidence value essentially describes the accuracy and reliability of this sensor data. Therein, the confidence value can for example be calculated depending on an image quality of the captured camera image for lane marking recognition, in particular depending on various image areas, which for example correspond to various distances or ranges from the ego vehicle. The image quality per se can again be ascertained based on parameters such as contrast, intensity, contrast courses, intensity courses, resolution and the like. Results of the data interpretation can also be incorporated in the second confidence value, such as for example if lane markings could be recognized at all or could be classified as such, with which likelihood the identified lane markings are actually lane markings, and so on. For a high reliability, the curvature of the lane markings, the lateral position thereof and the orientation thereof also have to be accurately determined such that uncertainties in determining these quantities can also be incorporated in the calculation of the at least one second confidence value. All of these quantities are correspondingly suitable to evaluate the confidence of the second lane area calculated based on the environmental data or sensor data, in particular by the corresponding at least one second confidence value. The at least one first confidence value for the first lane area can be essentially described based on dynamic parameters of the ego vehicle, and is later explained in more detail. Furthermore, the respective confidence value can be the higher, the higher the confidence of the lane area evaluated by it also is. Such a confidence value can for example assume between zero and one inclusive or also correspondingly between 0% and 100%.

In a further advantageous configuration of the invention, the driver assistance system is configured to determine a distance of the object to the ego vehicle and to ascertain the at least one first and the at least one second confidence value respectively depending on the determined distance. As initially already indicated, the confidence of the respective determination methods varies with the distance or the range to the ego vehicle among other things. For example, the first lane area can be very accurately ascertained based on the ego motion of the ego vehicle especially in the near range, while this is only very inaccurately possible for large distances to the ego vehicle. By the above described advantageous consideration of the distance of the object to the ego vehicle, this circumstance can be particularly advantageously accommodated. Hereby, the at least one confidence value can thus be specifically ascertained for the distance both for the first lane area, thus in calculating the second lane area, at which the object is currently located. Thereby, the lane area can be particularly accurately and reliably predicted especially at the location of the object. Thereby, the following ascertainment of whether or not the object is in the predicted lane area can also be provided with particularly high accuracy and low error rate. For example, if multiple vehicles are detected in front of the ego vehicle in direction of travel, thus, the corresponding confidence values can also be calculated for the respective distances.

Thus, it can be advantageously accomplished that for example in a near range related to the ego vehicle the predicted lane area is usually significantly influenced by the values of the first lane area since the prediction of the lane area is usually considerably better in the near range based on the ego motion of the vehicle of the ego vehicle than a prediction based on environmental data. In contrast, in a far distant area, in which a lane area calculation is usually considerably more inaccurate based on the ego motion of the ego vehicle than a prediction based on the environmental capture, the predicted lane area is significantly influenced by the values of the calculated second lane area. By this calculation of the respective confidence values depending on distance, an optimum adaptation to the respective situation with the best possible result can be achieved. These advantages can also be accomplished by the following embodiment of the invention in the same manner.

According to this further advantageous embodiment of the invention, the first and the second lane area are structured into multiple, determined distance ranges adjoining to each other with respect to the ego vehicle, wherein the control device is configured to ascertain multiple first confidence values for the first lane area and to ascertain multiple second confidence values for the second lane area. Therein, a respective of the multiple first and second confidence values is associated with one, in particular exactly one, of the multiple determined distance ranges. Furthermore, the control device is configured to calculate the predeterminable averaging for a respective of the multiple distance ranges as an averaging of the first lane area and the second lane area weighted depending on the respective first and second confidence value associated with the respective distance range for a respective distance range.

Thereby, the first and the second lane area can therefore first be calculated, and subsequently multiple confidence values can be associated with these calculated lane areas depending on the distances of the respective partial ranges, thus the distance ranges, of the concerned lane areas to the ego vehicle. Thus, the predicted lane area then results from averaging of these two first and second lane areas, which are incorporated in this averaging for respective distance ranges corresponding to their confidence values. Hereby too, a confidence evaluation resolved in location of the first and the second lane area is advantageously allowed such that the predicted lane area is correspondingly dominated in respective distance ranges by the most reliable data of these respective individual first and second lane areas. Subsequently, positions of detected objects can then be compared to the thus predicted lane area to ascertain whether the one or more of the objects are in the predicted lane area.

In a further advantageous configuration of the invention, the control device is configured to ascertain the at least one first confidence value, and in particular also for the case of multiple first confidence values as is described in the last example, depending on at least one of the following parameters: a speed of the ego vehicle, navigation data, which relates to a route course ahead, and capture of an initiation of a predetermined automatic driving maneuver, which involves a lateral offset of the ego vehicle with respect to the currently traveled roadway.

For example, the confidence can be evaluated the higher, the higher the speed of the ego vehicle is since at higher speeds the trajectory cannot be changed as fast because the wheel angle or steering angle and also the speed by itself cannot abruptly change. In contrast thereto, at very low speed, driving maneuvers changing considerably faster, i.e. on shorter travel distance, are possible such that at lower speed the confidence of the lane area calculated based on the ego motion of the vehicle is also correspondingly lower. Navigation data, for example of a navigation device, which can include a GPS receiver or another positioning device, as well as map data, which can be stored in a memory, can also be used to evaluate the confidence. As initially mentioned, the confidence of the lane area calculation based on the ego motion is relatively low in curvy route, while it is for example very high in a straight roadway. If the currently traveled roadway or the route course ahead extends linearly or curvy, can thereby be ascertained based on the corresponding navigation data. Herein, it is not necessarily required to directly obtain the curvature course of the route section ahead from the navigation data, for example it can also be concluded to such one based on the type of the currently traveled roadway. For example, highways typically extend more linearly than rural roads or also urban roads. Such a road type or regional information, such as for example if the ego vehicle is currently in town or outside of town, can also be obtained from the navigation data and it can be used for calculating the at least one first confidence value. Furthermore, it can also be provided that the driver assistance system is configured to allow the driver to adjust a desired lateral offset, which presets, which lateral distance is to be kept by the ego vehicle for example as a safety distance in overtaking procedures or to vehicles parked at the roadside or to the roadway boundary by itself. If such a lateral or side offset is preset to the driver assistance system by the driver, thus, the driver assistance system can control or correct the lateral offset of the ego vehicle during travel to satisfy this specification. For example, if a driver changes this specification during travel, thus, this results in the fact that the driver assistance system effects a lateral offset of the ego vehicle with respect to the currently traveled roadway in implementing this changed specification to adapt to the new specification. Such a lateral offset necessarily also entails a steering angle correction or a curved trajectory resulting therefrom, but which is not to be ascribed to a curvature of the currently traveled lane area or the currently traveled roadway. Thereby, the initiation of such a predetermined automatic driving maneuver also results in reduced confidence of the first lane area calculated during such a driving maneuver. By the above mentioned examples, it is therefore possible to consider numerous different situations in evaluating the confidence of the first lane area. Hereby, the confidence of the first lane area can be particularly accurately estimated. Of course, further parameters can also be considered in ascertaining the first or the multiple first confidence values, such as for example the dynamics of the longitudinal system or the like.

According to another embodiment the control device is configured to ascertain the at least one second confidence value depending on at least one of the following parameters: an environmental condition, like day or night, sun or rain, and so on, a sensor status of the at least one environmental sensor, a quality of the lane markings, especially on the right side and/or on the left side on the lane in traveling direction, and whether or not a vehicle having a size, which is larger than a predefined size, is in front of the ego vehicle, like a vehicle that can mask the road marks. Also many further parameters can be considered.

However, alternatively or also combinable with the previously mentioned examples, the lane area to be predicted cannot only be provided by weighted combination of the first and the second lane area, but numerous further possibilities exist to use both the first and the second determination method in advantageous and simple manner for a particularly accurate ascertainment of the lane area to be predicted, which are described in more detail below.

Therein, according to a further advantageous configuration of the invention, the control device is configured for the purpose of predicting the lane area ahead in dependence on the first lane area and/or the second lane area, to select the first and/or the second lane area as the predicted lane area depending on at least one predetermined parameter. Thus, the predicted lane area can be equated with the calculated first lane area in a first situation specified by the predetermined parameter and for example with the calculated second lane area in another situation specified by the predetermined parameter. Thus, averaging of the values of the first and the second lane area is not required to calculate the predicted lane area, but the first or second lane area can be selected as the predicted lane area according to situation. Hereby, the computing effort is enormously simplified and yet a particularly good adaptation to situation can be achieved by the predetermined parameter, which sets or at least influences the selection. It can also be that the first and the second lane area coincide for the relevant area, i.e. the location, at which the object is located, or at least with respect to the result whether or not the object is in the lane area, such that both lane areas, that is the first and the second, can also be provided as the predicted lane area. Thus, by the at least one predetermined parameter, the strengths of the respective determination methods can be again specifically used in adaptation to a respective situation and the respective weaknesses thereof can be compensated for at the same time.

Therein, it is particularly advantageous if the at least one predetermined parameter represents an ascertained distance of the object to the ego vehicle. Since the qualities of the individual determination methods depend on the distance to the ego vehicle, the selection of the first and/or second lane area depending on the ascertained distance of the object is again particularly advantageous. In order to describe the qualities or reliability of the first and the second lane area herein, it is either not required to ascertain above described confidence values. Therein, the decision criteria can be based on experience values and empirical calculations. Principally, there is still the possibility of providing the above described confidence values as the at least one predetermined parameter such that for example that lane area is selected from the first and second lane area, which has the higher reliability. Again, this would be additionally combinable with a calculation of the confidence values depending on distance.

In a further advantageous configuration of the invention, the at least one predetermined parameter represents a result of a temporally preceding ascertainment of whether the object is in the predicted lane area. Hereby, previous results can for example also be used for example from areas with higher accuracy or reliability. Such previous results can also be used for verification of the current decision or determine the decision between the first and second calculated lane area. For example, both the first and second lane area can be calculated and it can be examined whether or not the object is within the respective lane area, and the temporally previous result can be used for decision in case of contradicting results.

The above described parameters can also be combined with each other in any manner and can be used for selection between the first and/or the second lane area as the predicted lane area to again allow a particularly good adaptation to situation and a particularly high reliability of the finally predicted lane area.

According to a further advantageous configuration of the invention, a near range located in front of the ego vehicle in direction of travel is defined, which extends starting from the ego vehicle in a predetermined direction up to a predetermined first distance to the ego vehicle. Furthermore, a far range located in front of the ego vehicle is defined, which extends starting from the first distance or starting from a second distance larger than the first distance to the ego vehicle in the predetermined direction up to a third distance to the ego vehicle. Furthermore, the control device is configured to examine whether the object is in the near range or in the far range, and if the object is in the near range, to provide the first lane area as the predicted lane area, and if the object is in the far range, to provide the second lane area as the predicted lane area. This advantageous embodiment of the invention is again based on the realization that the calculation of the lane area based on the current ego motion of the ego vehicle is much better and has a considerably higher accuracy especially in the near range than the calculation of this lane area in the near range based on the environmental capture. However, the lane calculated based on the environmental capture in the far range considerably dominates a lane area calculated based on the ego motion of the ego vehicle with respect to its quality. Therefore, it is particularly advantageous to provide the first lane area as the predicted lane area in the near range and to provide the second lane area as the predicted lane area in the far range.

In this approach, the position or the distance of the object to the ego vehicle is preferably first ascertained, and if it is for example in the near range, the first lane area is calculated and equated to the predicted lane area. In this case, the second lane area for example even does not have to be calculated. In reverse manner, this is also the case for the far range. Hereby, computing effort can be enormously saved and yet the predicted lane area can be provided with high accuracy. Therein, the area ahead of the ego vehicle can for example be divided into only two partial ranges, the near range and the far range, which represents a particularly simple embodiment of this method. However, considerably more divisions can also be provided, which in turn provides considerably more flexible possibilities of adaptation.

Correspondingly, it is an advantageous configuration of the invention that an intermediate range between the near range and the far range is defined, wherein the control device is configured to examine whether the object is in the intermediate range, and if the object is in the intermediate range, to ascertain whether the object is in the first lane area, and to provide a result of ascertainment as a first result. Furthermore, the control device is configured to ascertain whether the object is in the second lane area and to provide a result of this ascertainment as a second result, wherein the control device is further configured to compare the first result to the second result, and in case of coincidence to provide the first or the second result, which represent the same result in this case, as an overall result of the ascertainment of whether the object is in the predicted lane area. Thereby, this intermediate range advantageously allows still considerably more differentiating possibility. Especially the matching between the results, i.e. the first and the second results, allows additional verification and thereby a considerably higher safety and reliability of the overall result.

However, if a coincidence between these two results is not present, i.e. thus for example if the object is for example in the first lane area by matching with the respectively calculated first and second lane area, but not in the second lane area or vice versa, thus, there are again multiple particularly advantageous possibilities of making decisions in such a situation, which are optimized both with respect to safety and driving comfort, and which are explained in more detail below.

Therein, according to a variant of the invention, in case that the first result does not coincide with the second result, the control device is further configured to provide as the overall result that the object is in the predicted lane area. This represents a particularly conservative solution with respect to safety. This has the background that the false assumption that the object is in the same lane area as the ego vehicle, can have considerably less severe consequences than the false assumption that the object is not in the lane area of the ego vehicle, although this is the case. In the first case, the ego vehicle is for example erroneously decelerated or the driver is unnecessarily warned, while in the second case a rear-end collision can be the consequence in the worst case. This conservative decision criterion thereby results in a particularly high safety.

In a further advantageous configuration of the invention, the intermediate range is defined such that it has a first partial range and a second partial range adjoining to the first partial range, wherein the first partial range further adjoins to the near range and the second partial range adjoins to the far range. Furthermore, the control device is configured to examine whether the object is in the first partial range or in the second partial range, and in case that the first result does not coincide with the second result, and

in case that the object is in the first partial range, the control device is configured to

• if the first result involves that the object is in the first lane area, to provide the first result as the overall result;

• if the first result involves that the object is not in the first lane area, to provide an overall result of a temporally preceding ascertainment of whether the object is in the predicted lane area, in particular relating to the second partial range, as the current overall result; and

in case that the object is in the second partial range, the control device is configured to • if the second result involves that the object is in the second lane area, to provide the second result as the overall result; and

• if the second result involves that the object is not in the second lane area to provide an overall result of a temporally preceding ascertainment of whether the object is in the predicted lane area, in particular relating to the first partial range, as the overall result.

In the first partial range, thus which is closer to the near range, is it therefore to be assumed that the results of the first determination method based on the ego motion of the ego vehicle have a higher reliability than the results of the second determination method. Thus, if the result of the first determination method, that is thus the first result, turns out conservative, it is therefore determined that the object is in the first lane area, thus, this result is assumed as the overall result. This overall result is then also applicable in all likelihood since it is based on the more reliable data. In case that this nevertheless does not apply, more severe consequences such as rear-end collisions can still be avoided by the conservative result. The same also applies to the second partial range. Usually, the result of the second determination method based on the environmental capture is considerably more applicable to it than the calculation based on the ego motion. Thus, if the result of the second calculation of the determination method turns out conservative in this second partial range, that is it is ascertained that the object is in the second lane area, thus, it is assumed as the overall result. Correspondingly, the likelihood is again very high that this overall result is also applicable, and more severe consequences as accidents can again be avoided otherwise.

However, if it is ascertained in another case for the first partial range, in which the results based on the ego motion are usually more applicable, that the object is not in the first lane area, however, if the second result based on the second determination method turns out different, thus, for additional verification of the first result, the overall result of a temporally preceding ascertainment of whether the object is in the predicted lane area can be advantageously used. If this preceding result for example states that the object has previously either not been located in the predicted lane area, thus, the first result can for example be set as the overall result. Otherwise, it is assumed as the overall result that the object is in the predicted lane area. This also analogously applies to the second partial range, such that here too the overall result of a temporally preceding ascertainment, in particular also from another distance range, can be used in case of conflict. By this approach, considerably more differentiated and thereby more applicable decisions can be provided. The likelihood for false decision is reduced in the same manner and thereby results considerably less often in unnecessary decelerations or unnecessary warning messages.

Furthermore, the invention also relates to a vehicle, especially to the vehicle named ego vehicle before, which is preferably formed as a motor vehicle, in particular as a passenger car. The vehicle according to the invention comprises a driver assistance system according to the invention or one of its configurations. The features, feature combinations mentioned for the driver assistance system according to the invention at its configurations and the advantages thereof similarly apply to the ego vehicle according to the invention and its configurations.

Furthermore, the invention relates to a method for predicting a lane area ahead of an ego vehicle to ascertain whether a object is in the predicted lane area. Therein, a first lane area ahead of the ego vehicle is determined depending on at least two captured ego motion parameters of the ego vehicle according to a first determination method and a second lane area ahead of the ego vehicle is determined depending on captured lane markings according to a second determination method. The lane area ahead is further predicted depending on the first lane area and/or the second lane area.

The advantages mentioned for the driver assistance system according to the invention and its configurations similarly apply to the method according to the invention.

Furthermore, the objective features mentioned in context with the driver assistance system according to the invention and its configurations allow the development of the method according to the invention by further methods steps.

Further features of the invention are apparent from the claims, the figures and the description of figures. The features and feature combinations mentioned above in the description as well as the features and feature combinations mentioned below in the description of figures and/or shown in the figures alone are usable not only in the respectively specified combination, but also in other combinations or alone without departing from the scope of the invention. Thus, implementations are also to be considered as encompassed and disclosed by the invention, which are not explicitly shown in the figures and explained, but arise from and can be generated by separated feature combinations from the explained implementations. Implementations and feature combinations are also to be considered as disclosed, which thus do not have all of the features of an originally formulated independent claim. Moreover, implementations and feature combinations are to be considered as disclosed, in particular by the implementations set out above, which extend beyond or deviate from the feature combinations set out in the relations of the claims.

There show:

Fig. 1 a schematic representation of a vehicle with a driver assistance system for predicting a lane area ahead of the ego vehicle according to an embodiment of the invention;

Fig. 2 a schematic representation of a lane course calculated based on a ego motion of a vehicle in a curve situation;

Fig. 3 a schematic representation of a vehicle according to an embodiment of the invention as well as of the lane areas calculated according to the first and the second determination method;

Fig. 4 a tabular representation of a method for predicting a lane area ahead based on the first and the second calculated lane area according to an embodiment of the invention;

Fig. 5 a flow diagram for illustrating a method for predicting a lane area ahead of the ego vehicle according to an embodiment of the invention;

Fig. 6 a schematic representation of the ego vehicle and the calculated first and second lane areas with an illustration of respective confidence values for various distance ranges according to a first use case for calculating the predicted lane area according to an embodiment of the invention;

Fig. 7 a schematic representation of the egoa vehicle as well as of the calculated first and second lane areas with an illustration of associated confidence values for various distance ranges according to a second use case according to a further embodiment of the invention; and

Fig. 8 a flow diagram for illustrating a method for predicting a lane area according to a further embodiment of the invention. Fig. 1 shows a schematic representation of a vehicle 1 with a driver assistance system 2 according to an embodiment of the invention. The vehicle 1 , which is also referred to as ego vehicle 1 below, further comprises at least one environmental sensor, wherein here four environmental sensors 3 such as for example cameras are exemplarily illustrated. These sensors 3 can also be placed at other positions as well. For example a video sensor can be positioned at the middle to of the front windscreen, one or two laser scanners can be positioned inside the front bumper, a video sensor can be positioned at the middle of the front bumper, a video sensor can be positioned in each side mirror and one video sensor can be positioned at the rear of the ego vehicle 1 . Therein, the environmental sensors 3 are configured to capture at least a part of the vehicle environment and therein especially also lane markings 4 of a currently traveled roadway 5. Also a navigation device, like a GPS sensor and an associated map, can be provided as another environmental sensor 3, for example to determine if the ego vehicle 1 is driving on a straight road of a curvy road. Furthermore, the vehicle 1 comprises at least one capturing means 6, which is configured to capture at least two ego motion parameters of the vehicle 1 . Therein, multiple such capturing means 6 can also be provided, which can for example be formed as a steering angle sensor, speed sensor and so on. Preferably, ego motion parameters such as for example the vehicle speed, the current steering angle turn, a yaw rate or a yaw angle, an acceleration or the like are captured by the capturing means 6. Furthermore, the vehicle 1 , in particular the driver assistance system 2, comprises a control device 7, which is configured to evaluate the captured quantities, i.e. the sensor data captured by the environmental sensors 3 on the one hand as well as the ego motion parameters captured by the capturing means 6.

Therein, the control device 7 is configured to determine a first lane area 12 ahead of the ego vehicle 1 (compare Fig. 3, Fig. 6 and Fig. 7) depending on the at least two ego motion parameters of the ego vehicle 1 according to a first determination method, as well as to determine a second lane area 13 ahead of the ego vehicle 1 (compare Fig. 3, Fig. 6 and Fig. 7) depending on sensor data of the at least one environmental sensor 3 relating to the capture of the lane markings 4 according to a second determination method on the other hand. For determining the first lane area, the control device 7 can for example determine a current trajectory of the vehicle 1 based on the current ego motion parameters and predict the first lane area 12 by extrapolation. For example, the vehicle width, optionally plus a preset value, or also a predetermined standard value can be assumed as the width of this first lane area 12. In order to ascertain the first lane area, the captured lane markings 4 can be correspondingly evaluated. Hereto, a curvature of the lane markings 4 as well as also their lateral position and orientation can for example be determined to ascertain the second lane course 13 therefrom. Though the lane markings 4 are illustrated here on the left hand side of the ego vehicle 1 , lane markings 4 on the right hand side of the ego vehicle 1 can be captured by means of the sensors 3 as well, especially additionally or alternatively.

Furthermore, the control device 7 is configured to predict the lane area ahead of the ego vehicle 1 in direction of travel and to be traveled depending on the first lane area 12 and/or the second lane area 13. Furthermore, the control device 7 is configured to determine based on the driving area finally predicted for a respective time step whether an object, like a preceding vehicle 14 (compare Fig. 3, Fig .6 and Fig. 7), is in this predicted lane area. If this is the case, the control device 7 selects this vehicle 14 as a target, otherwise not. Based thereon, the driver assistance system 2 can perform further driver assistance functions, such as for example controlling the speed of the ego vehicle 1 depending on the selected target to maintain a minimum distance, warning of a collision with the target or the like.

Both determination methods for ascertaining the first and the second lane area 12, 13, have weaknesses taken by themselves. The determination of the second lane area based on the environmental data for example is very severely dependent on corresponding environmental conditions and environmental influences such as sight conditions, weather conditions, the quality of the lane markings and the contrast thereof to the underground, i.e. the roadway 5. The quality of the environmental sensor system 3 by itself and the quality of the evaluation and interpretation methods also play a role. However, especially by severe dependency on environmental conditions, which sometimes can be subject to very severe fluctuations, prediction of the lane area based on the environmental capture alone is not particularly reliable. Especially, it either cannot be predicted up to now when the environmental capture, in particular the capture of the lane markings, is better or worse, whereby it would be very risky to only rely on a lane area predicted by the environmental capture. In the same manner, the first determination method taken by itself also has problems as is explained in more detail based on Fig. 2.

Fig. 2 shows a schematic representation of a vehicle 8 not belonging to the invention and a lane course 9 calculated based on the ego motion of the vehicle 8 on the example of a roadway 5 extending very curvy. Furthermore, further traffic participants 10a, 10b, 10c are also on this roadway 5, which travel in the same direction of travel illustrated by the arrow 1 1 as the vehicle 8. In this example, the vehicle is still immediately before the curve and travels straight ahead at this moment. Correspondingly, the calculation of the lane 9 based on the ego motion of the vehicle 8 results in an area extending linearly in front of the vehicle 8. Now, if a target selection is to be effected based on this ascertained lane 9, thus, this results in a false selection in this example, which can have critical

consequences. Therein, it is generally understood by a target selection that another traffic participant is selected, which is immediately ahead on the lane presumably to be traveled by the ego vehicle. In the example illustrated in Fig. 2, thus, the traffic participant 10a would be selected as such a target, which is on the ascertained lane 9. However, this traffic participant 10a is actually on another lane, which will not be traveled by the vehicle 8. In contrast, the other traffic participants 10b and 10c, for which there is a risk of collision, since they are at least partially on the lane to be traveled by the vehicle 8, are erroneously not selected. Such false selections, which correspondingly are to be ascribed to an erroneous prediction of the lane 9, can sometimes have very critical consequences.

Now, the invention in its configurations advantageously provides the possibility of combining the two described methods, i.e. ascertaining the lane area based on the ego motion as well as based on the environmental sensors 3, wherefrom a particularly reliably ascertainable lane area can overall be predicted, which will probably be traveled by the ego vehicle 1 . A selection of these various possibilities of combination is explained in more detail based on the following examples. Though the following examples relate for illustrative purposes only to a preceding vehicle 14 as an example of an object, the invention and its embodiments, especially those described in the following, apply similarly for any arbitrary object, like pedestrians, obstacles, moving or static objects, additionally to or instead of the preceding vehicle 14.

Fig. 3 shows a schematic representation of an ego vehicle 1 , such as for example the one described with regard to Fig. 1 , as well as the first lane area 12 ahead calculated according to the first determination method based on the ego motion parameters of the vehicle 1 , as well as the second lane area 13 calculated according to the second determination method based on the environmental capture, in particular the lane marking capture. Furthermore, a vehicle 14 preceding the vehicle 1 is illustrated here in the temporal progress, that is for multiple different time steps, in different positions.

Usually, it is the case that the first lane area course 12 provides very good and reliable results for low distances or ranges to the ego vehicle 1 . However, with non-straight roadway course, the first lane area 12 only has a very low reliability in larger distance to the ego vehicle 1 . The second lane area 13 calculated based on the environmental data is subject to large reliability variations due to the dependency on the environmental capture such that it can basically be regarded as considerably less reliable especially for short distances to the ego vehicle 1 than the first calculated lane area 12. However, for larger distances to the vehicle 1 , and in particular with curved roadway, it can be assumed that the reliability of the second lane area 13 is considerably higher than that of the first lane area 12. Now, this realization can be advantageously utilized to select the lane area to be predicted from the first lane area 12 and/or the second lane area 13 depending on the distance to the ego vehicle 1 .

Hereto, the area in front of the ego vehicle 1 in direction of travel can be divided into multiple distance ranges. Herein, a near range N can in particular be provided as well as a far range F. In the near range N, the first lane area 12 is then correspondingly provided as the lane area to be predicted, while the second lane area 13 is provided as the lane area to be predicted in the far range F. Thus, if the preceding vehicle 14 is in the near range N, thus, the position of this preceding vehicle 14 is compared to the course of the first lane area 12 and it is for example ascertained, as illustrated here, that the preceding vehicle 14 is in the first lane area 12. If the preceding vehicle 14 is for example in the far range F, thus, the ascertained position of the preceding vehicle 14 is compared to the calculated second lane area 13, and it is ascertained, as illustrated in this example, that the preceding vehicle 14 is in the second lane area. Thus, the preceding vehicle 14 is correctly selected as the target in both cases.

In order to additionally avoid an abrupt transition of this selection concept, it is

advantageous to provide at least one intermediate range. In this example, this

intermediate range is divided into two partial ranges Z1 and Z2. In order to set these individual distance ranges, it can for example be provided to preset a distance d, such as for example 20 m. The boundary between the near range N and the first partial range Z1 can then for example be set by the distance d minus a percentage value related to the distance value d, which is exemplified in Fig. 3 by x%, and the boundary between the second partial range and the far range can be set by a second percentage value related to the distance value d, which is here exemplified by y%. Thereby, the distance d sets the boundary between the first partial range Z1 and the second partial range Z2, while the value d-x% sets the boundary between the near range N and the first partial range Z1 and the value d+y% sets the boundary between the second partial range Z2 and the far range F. The percentage values x and y can be identical or different. In case that x% and y% are each 20 % of the distance d and the distance d is for example 20 m, thus, the boundary between the near range N and the first partial range Z1 is at 16 m and the boundary between the second partial range Z2 and the far range F is at 24 m. The boundaries themselves can be arbitrarily associated with one of the ranges adjoining to them. The distances to the vehicle 1 can also be ascertained by the environmental sensors 3 of the vehicle 1 . The distance can further for example be defined as the smallest distance, for example as the linear distance between a point of the environment and a selected point of the vehicle 1 . Correspondingly, the boundaries d-x%, d and d+y% would then for example extend circularly around the vehicle 1 . However, distances can also arbitrarily differently defined, for example along one of the calculated lane areas 12, 13 or as a distance perpendicular to a transverse axis of the vehicle 1 . However, the same distance definition is preferably taken as a basis for setting these distance ranges N, Z1 , Z2, F and for ascertaining the distance of the preceding vehicle 14. This in particular also applies to the distance ranges A1 , A2, A3, A4 described in more detail below in context with Fig. 6 and Fig. 7.

Providing these intermediate ranges, that is the first partial range Z1 and the second partial range Z2, provides considerably more flexibility and possibilities of adaptation, which overall allow a more reliable and more exact prediction of the driving area ahead. An example, how the decision making can occur for these four defined distance ranges, i.e. the near range N, the first partial range Z1 , the second partial range Z2 and the far range F, is now described in more detail based on Fig. 4.

Therein, Fig. 4 shows a table for illustrating a method for predicting a lane area ahead based on the first and/or second lane area 12, 13 according to an embodiment of the invention. In this table, N again denotes the near range, Z1 the first partial range, Z2 the second partial range and F the far range. Herein, E1 denotes the result of an examination of whether the preceding vehicle 14 is in the first lane area 12, while E2 denotes the result of an examination of whether the preceding vehicle 14 is in the second lane area 13. Furthermore, "+" denotes that the preceding vehicle 14 is in the concerned lane area and thus is selected as a target, while "-" denotes that the preceding vehicle 14 is not in the concerned lane area and therefore is not selected as a target. E further denotes the overall result, which is finally provided by the driver assistance system 2 for further utilization. If the preceding vehicle 14 is for example in the near range N, thus, only the first lane area 12 is calculated since it is usually considerably more accurate in the near range N than the second lane area 13. Correspondingly, the position of the preceding vehicle 14 is compared to the first lane area 12. If it is therein ascertained as the first result E1 that the vehicle 14 is in the first lane area 12, thus, this first result E1 is equated with the overall result E. If the preceding vehicle 14 is not in the first lane area 12, thus, the first result E1 is equated with the overall result in the near range N here too. This is illustrated by the third and the fourth line of the illustrated table. The following two lines describe the situation in the far range F. Here, only the second lane area is relevant 13, and the position of the vehicle 14 located in the far range F is correspondingly compared to this second lane area 13. If this examination yields that the vehicle 14 is in the second lane area 13, thus, this also represents the overall result E. If the second result E2 yields that the vehicle 14 is not in the second lane area 13, this also represents the final result E. In the further four following lines of the table, the situation in the first partial range Z1 is described. Thus, if the vehicle 14 is in the first partial range Z1 , thus, both the first lane area 12 and the second lane area 13 are calculated as well as the respective results E1 and E2 of the target selection. If they coincide, as represented in the lines seven and eight, thus, these results E1 , E2 also apply as the final result or overall result E.

With differing first and second results E1 and E2, there are now several possibilities. In the following, one variant of them is explained. For example, if it is ascertained for the first partial range Z1 that the vehicle 14 is in the first lane area 12, but not in the second lane area 13, thus, it applies as the overall result that the vehicle 14 is in the predicted lane area and thus is selected as a target. This has the background that a higher accuracy of the first lane area 12 can still be assumed in the first partial range Z1 and additionally the first result E1 also represents the conservative result, in which safety risks are not to be expected in case of error. Otherwise, if it is ascertained for the first partial range Z1 that the vehicle 14 is not in the first lane area 12, but in the second lane area 13, thus, it can be resorted to an overall result E in a temporally preceding ascertainment for the decision between these two contradicting results E1 and E2. For example, the previous

ascertainment in the zone Z2, especially if the distance between the ego vehicle 1 and the front vehicle 14 is decreasing, can be resorted. For example, if the vehicle 14 was previously in the second partial range Z2, and if it was therein ascertained as the overall result E that the vehicle 14 has been previously in the predicted lane area, which is denoted by "(+)" in the table, thus, it is also assumed as the overall result E in this case that the vehicle 14 is at the current time in the predicted lane area. In contrast, if it was therein ascertained as the overall result E that the vehicle 14 has not been previously in the predicted lane area, which is denoted by "(-)" in the table, it is also assumed at the current time that the vehicle 14 either is not in the predicted lane area as the overall result E.

Alternatively to this approach, the conservative result could also simply always be set as the final result E, that is "+". In other words, an alternative approach is to always consider a conservative„OR" result when the vehicle 14 is in the intermediate zones Z1 or Z2 : If a vehicle 14 is detected either by result E1 or result E2 in the corresponding estimated lane area 12 or 13 respectively, then it is selected in the final result E. This is also similarly the case in the second partial range Z2. With coinciding results E1 and E2, these results are also equated with the final result E. In case that the results E1 and E2 do not correspond in this second partial range Z2, and it has further been ascertained that the vehicle 14 is in the second lane area 13, thus, this is also set as the final result E. In contrast, if it is ascertained that the vehicle 14 is not in the second lane area 13, but in the first lane area 12, thus, preceding overall results E can here also again be considered. For example, if they state that the vehicle 14, which was for example previously in the first partial range Z1 , has not been in the predicted driving area, thus, this is set as the final result E, if so, thus, this is also again set as the current final result E.

This approach, where the temporally preceding overall result E is taken into account, is again illustrated as a flow diagram in Fig. 5. Herein, in step S1 , a preceding vehicle 14 is first captured by environmental sensors 3 of the vehicle 1 and the distance thereof to the ego vehicle 1 is ascertained, for example as the shortest distance. Furthermore, it is examined in step S2 whether this ascertained distance is in the near range N, in the first partial range Z1 , in the second partial range Z2 or in the far range F. If the distance is in the near range N, thus, in step S3, the first lane area 12 is calculated according to the first determination method and the ascertained vehicle position of the vehicle 14 is compared to the first lane area 12 in step S4. If it is ascertained as the result of this comparison in step S5 that the vehicle 14 is in the first lane area 12, thus, the vehicle 14 is selected as the target in step S6, otherwise not in step S7.

In contrast, if it is ascertained in step S2 that the preceding vehicle 14 is in the far range F, thus, the second lane area 13 is calculated according to the second determination method in step S8 and the calculated second lane area 13 is compared to the vehicle position of the vehicle 14 in step S9. If it is then ascertained in step S10 that the vehicle 14 is in the second lane area 13, the vehicle 14 is selected as the target in step S1 1 , otherwise not in step S 12.

If it is ascertained in step S2 that the vehicle 14 is in the first partial range Z1 , thus, both the first lane area 12 is calculated according to the first determination method and the second lane area 13 is calculated according to the second determination method in step S14 and the vehicle position of the vehicle 14 is compared to the respective lane areas 12 and 13 in step S15. Furthermore, it is examined in step S16 whether the results coincide with respect to the fact whether the vehicle 14 is in the corresponding lane area 12 or 13. If this is the case and if it is ascertained in step S17 that the vehicle 14 is in both lane areas 12 and 13, thus, the vehicle 14 is selected as the target in step S18, otherwise not in step S19. In contrast, if it is ascertained in step S16, that the results do not coincide, thus, it is examined in step S20 whether the result according to the first determination method is positive, and if so, the vehicle 14 is selected as the target in step S21 .

Otherwise, a temporally preceding overall result is used in step S22, and if it was positive, the vehicle 14 is selected as the target in step S23, and otherwise not in step S24.

If it is ascertained in step S2 that the vehicle 14 is in the second partial range Z2, thus, both the first lane area 12 and the second lane area 13 are again calculated in step S25 and compared to the position of the vehicle 14 in step S26. If the results coincide in step S27, thus, in case of positive results in step S28, i.e. the vehicle 14 is both in the first and in the second lane area 12, 13, thus, the vehicle 14 is selected as the target in step S29, otherwise not in step S30. In contrast, if the results do not coincide in step S27, thus, it is examined in step S31 whether the result according to the second determination method was positive with respect to the second lane area 13, and if so, the vehicle 14 is selected as the target in step S32. Otherwise, it is again resorted to a temporally preceding overall result in step S33, and if it was positive, the vehicle 14 is selected as the target in step S34 and otherwise not in step S35.

By the above described examples, the lane area ahead can be reliably predicted in particularly simple and computationally particularly little expensive manner. However, there are still numerous further possibilities to provide the lane area to be predicted depending on the first lane area 12 and/or the second lane area 13 with high accuracy. One of these possibilities is now described below based on Fig. 6, Fig. 7 and Fig. 8.

Fig. 6 again shows a schematic representation of a roadway 5 with an ego vehicle 1 and a vehicle 14 preceding the ego vehicle 1 . Furthermore, the first lane area 12 calculated according to the first determination method as well as the lane area 13 calculated according to the second determination method are again illustrated, in particular again for a certain point of time. In this example, respective confidence values W1 , W2 can now further be calculated for a respective lane area 12, 13. Hereto, the area in front of the ego vehicle 1 in direction of travel and correspondingly also the first and the second lane area 12, 13 is divided into multiple determined distance ranges A1 , A2, A3, A4 adjoining to each other, four in this example. However, more or less distance ranges can optionally also be provided. The first distance range A1 can for example extend starting from the ego vehicle 1 between 0 m and 5 m, the second distance range A2 between 5 m and 15 m, the third distance range A3 between 15 m and 30 m, and the fourth distance range A4 between 30 m and 50 m. These distance ranges are only given as order of magnitude and also can be defined differently. Moreover this also applies for the distant ranges presented in Fig. 7.

Although the boundary lines between the respective distance ranges A1 , A2, A3, A4 are linearly illustrated, thus, these boundaries can again extend for example on circular lines around a designated point of the ego vehicle 1 , with respect to which these respective distances are measured. For the first lane area 12, corresponding confidence values W1 are then correspondingly calculated for a respective distance range A1 , A2, A3, A4. For a relatively fast traveling ego vehicle 1 , for example for 70 km/h, the reliability of the first lane area 12 in the first distance range A1 is very good, and in particular here exemplarily indicated with 100 %. With increasing distance, the reliability of the first lane area 12 decreases, here exemplarily over 75 % in the second distance range A1 , 50 % in the third distance range A3 up to 25 % in the fourth distance range A4.

For the second lane area 13 too, respective confidence values W2 are calculated for the respective distance ranges A1 , A2, A3, A4. Herein, a reliability of 80 % exemplarily results for the first distance range A1 , of 75 % in the second distance range A2, of 70 % in the third distance range and of 50 % in the fourth distance range. Here too, the reliability of the second lane area 13 decreases with increasing distance to the ego vehicle 1 , but not as drastically as it is the case for the first lane area 12. The lane area to be predicted can now advantageously be calculated as an averaging of the first and the second lane area 12, 13 for the respective distance ranges A1 , A2, A3 and A4 with weighting corresponding to the respective confidence values W1 , W2 associated with the distance ranges A1 , A2, A3, A4. By such an approach for providing the predicted lane area, the reliability can overall be maximized. Finally, the ascertained position of the vehicle 14 can then be compared to the thus calculated predicted lane area and it can be ascertained based on this comparison whether or not the preceding vehicle 14 is in the predicted lane area.

Fig. 7 again shows a further illustration of the method described to Fig. 6, but in this case for a lower speed of the ego vehicle 1 of for example 30 km/h and for illustrating a different use case as that one shown in Fig. 6. Furthermore, the roadway 5 traveled by the ego vehicle 1 , a vehicle 14 preceding the ego vehicle 1 as well as the first lane area 12 calculated according to the first determination method and the lane area 13 calculated according to the second determination method are again illustrated in this example. Again, the respective confidence values W1 for the first lane area 12 and the confidence values W2 for the second lane area 13 are illustrated for the respective distance ranges A1 , A2, A3, A4.

In this situation too, the reliability of the first determination method or of the first lane area 12 resulting therefrom is very high for the first distance range A1 , in particular 100 %. However, the reliability decreases considerably faster with increasing distance to the ego vehicle 1 due to the lower speed of the ego vehicle 1 and is for example only 5 % in the fourth distance range A4. At a speed of 30 km/h, a distance of 30 m is traveled within about 3 seconds such that the likelihood of a change of the odometric data within these 3 seconds is very high such that the calculation of the first lane area 12 is very unreliable for this fourth distance range A4, while the reliability of the lane recognition is relatively high with 50 % in this range caused by the reliability by the recognition algorithm itself.

For the second lane area 13, there also results another course of the confidence values W2 for the respective distance ranges A1 , A2, A3, A4. Therein, these other values are not necessarily to be ascribed to the lower speed of the ego vehicle 1 , but are to demonstrate the dependency on situation of the environmental capture. The lower confidence values W2 for the first and the second distance range A1 , A2 can for example be ascribed to the fact that for example for short distances to the ego vehicle 1 , information with respect to the lane markings 4 cannot be provided by one of the environmental sensors 3, such as for example the cocoon camera, which provide usually short range detection of road marks. So in the present illustration, if the cocoon camera cannot provide information, for example if the sensor is out of order, only road marks perception by other sensors such as the front widescreen camera is available, but for range from about 10m up to about 50m. Now, the lane area to be predicted can again be calculated as an averaging of the first and the second lane area 12, 13 for the respective distance ranges A1 , A2, A3 and A4 with a weighting corresponding to the respective confidence values W1 , W2 associated with the distance ranges A1 , A2, A3, A4, and a target selection can then be performed based thereon. This approach is for example periodically repeated in respective consecutive time steps.

Fig. 8 shows a flow diagram for illustrating the just described methods according to an embodiment of the invention. Herein, in step S40, the first lane area 12 is calculated according to the first determination method and the second lane area 13 according to the second determination method. Furthermore, in step S41 , a confidence value W1 , W2 is calculated for the respectively calculated lane areas 12, 13 for a respectively defined distance range A1 , A2, A3, A4. Furthermore, in step S42, the first lane area 12 and the second lane area 13 are averaged correspondingly weighted according to their respective confidence values W1 , W2 for a respective of the distance ranges A1 , A2, A3, A4. The result of this averaging is provided in step S43 as the predicted lane area and in step S44 it is examined whether or not the measured position of the preceding vehicle 14 is in the predicted lane area. If this is the case, thus, the preceding vehicle 14 is selected as the target in step S45, otherwise not in step S46. Here too, this method is again continuously repeated in consecutive time steps.

Thus, a driver assistance system, a vehicle and a method are overall provided, which allow providing the lane area ahead with particularly high reliability and particularly adapted to situation by the possibilities of combination of two different methods for determining a lane area ahead.