Method and assistance system for predicting travel path, and motor vehicle

Abstract

The invention relates to a method and an assistance system (5) for predicting a travel path (11) of a motor vehicle (2). The invention also relates to a correspondingly arranged motor vehicle (2). In the method, a corresponding environment scene in front of the driving direction of the motor vehicle (2) is determined. For the environment scene, plausibility of a plurality of different data sources and / or data originating from the data sources is ascertained, on the basis of which plausibility a travel path (11) can be predicted. The plurality of data originating from the different data sources are fused to one another in a weighted manner as a function of the determined plausibility and a travel path (11) for the motor vehicle (2) is thus predicted.

Description

Technical Field

[0001] The invention relates to a method and an assistance system for predicting a driving path of a motor vehicle and to a correspondingly designed motor vehicle. Background Art

[0002] For various assistance functions of a motor vehicle as well as for automatic vehicle guidance, an estimate, i.e. a prediction, of the expected movement or the expected movement path or movement range of the motor vehicle can be useful. Such a prediction or estimation can be carried out, for example, based on driving dynamics data of the motor vehicle. However, this can lead to errors in different situations, which in turn can lead to further errors in the function or system. Therefore, improvements compared to conventional solutions are desirable for improved functionality, performance and safety corresponding to assistance functions and automation.

[0003] For example, EP1844373B1 describes a method for route prediction in a driver assistance system of a motor vehicle, in which a dynamic route hypothesis is created based on the driving dynamics data of the vehicle. First, an infrastructure route hypothesis is created based on data describing the traffic infrastructure from an information source. A weighting factor describing the reliability of the infrastructure route hypothesis is calculated based on the characteristics of the information source. In addition, a fusion of the infrastructure route hypothesis and the dynamic route hypothesis weighted according to the calculated weighting factor is provided there to form a final route hypothesis.

[0004] The possible use of the predicted driving route is described in DE 10 2018 127 270 A1. A control unit for a vehicle is provided there, which includes a control device that enables the driver of the vehicle to make control inputs for the longitudinal and / or lateral guidance of the vehicle. The control unit is provided there to determine the permissible driving path for the vehicle based on the current state of the vehicle. The driving path has multiple dimensions and displays the permissible range that can be technically and / or accident-free by the vehicle for each dimension. The multiple dimensions include spatial dimensions and driving dynamics dimensions. In addition, a controller is provided to implement the driver's control input in such a way that the vehicle remains within the driving path while driving. Therefore, automatic longitudinal and / or lateral guidance of the vehicle that can be controlled by the driver can be achieved in an effective manner, especially in order to improve the comfort and driving pleasure for the driver. Summary of the invention

[0005] The object of the present invention is to enable a particularly accurate and robust prediction of the travel path of a motor vehicle.

[0006] This object is achieved according to the invention by the content of the independent patent claims. Possible embodiments and developments of the invention are disclosed in the dependent claims, the description and the drawings.

[0007] The method according to the invention can be applied when predicting the driving path of a motor vehicle. This driving path or driving corridor extends away from the motor vehicle in the driving direction of the motor vehicle and specifies the area that the motor vehicle is expected to pass or travel through at each current time point. In the method steps of the method according to the invention, the corresponding environmental scene currently in front of the driving direction of the motor vehicle is determined. In this case, generally speaking, one or a class of environmental scenes or a specific or unique environmental scene that is only obtained at each position can be determined. In order to determine the corresponding environmental scene, for example, the current position of the motor vehicle can be determined and taken into account, for example, with respect to a predetermined, especially world-fixed coordinate system and / or with respect to certain features in each environment, which define or jointly determine each environmental scene, such as a roundabout, a connection point, etc.

[0008] Environmental scenes in the current sense can be, for example, a crossroad, an upcoming turn, a roundabout, a playground, a junction, etc. Thus, an environmental scene here can be a spatially limited area defined by its type and / or characteristics, which can be driven or driven through by a motor vehicle. Likewise, corresponding maneuvers of the motor vehicle, such as driving towards a certain feature, driving through a certain area, leaving a certain area or feature, etc., can further define the respective environmental scene. For example, driving towards a roundabout, entering a roundabout, driving in a roundabout and / or leaving a roundabout can be different environmental scenes.

[0009] In this case, the environment scene can also be influenced or defined to be distinguished from other environment scenes by the current environment conditions, such as weather conditions, light conditions, local traffic density, roadway conditions, such as dryness, moisture or snow and / or the like. Thus, for example, a certain roundabout in the dark and with a snow-covered roadway and / or a relatively high traffic density may constitute a different environment scene than the same roundabout in the daytime, with a dry roadway and / or with minimal traffic density (especially without vehicles outside the motor vehicle).

[0010] In another method step of the method according to the present invention, the credibility of multiple, especially different data sources and / or data derived from, i.e., from various data sources are obtained for the corresponding environmental scene obtained or depending on or corresponding to the corresponding environmental scene obtained. This involves a data source or data based on this, i.e., the driving path can be predicted thereby. Therefore, multiple data sources or data can have different or transformed absolute and / or relative credibility in different environmental scenes. The credibility can give or describe the corresponding trustworthiness, reliability, actual availability, etc., as given or expected in the corresponding environmental scene. Therefore, for example, when it snows or rains, the laser radar data may have a lower credibility than the laser radar data in the environmental scene without precipitation or the camera or map data in the same environmental scene. Similarly, the camera data may have a lower credibility than the map data or the trajectory data describing the trajectory of the front traffic participant, etc., when the relatively large curvature or relatively strong curvature of the road ahead is relatively large. Similarly, the credibility of the radar data of the lane boundary may depend on the type of lane boundary, for example. Thus, radar data characterizing the position or course of a lane boundary may have greater plausibility in a first environment scenario in which the lane boundary is formed by a curb or a guardrail, for example, than in a second environment scenario in which the lane boundary is formed by grass or only by road markings.

[0011] In a further method step of the method according to the invention, a plurality of, i.e. selected or all, of the data originating from different data sources are fused with one another in a weighted manner according to the determined plausibility. Thus, for example, the driving route is predicted or updated by fusing the data in a corresponding predefined fusion and prediction model or by using the fused data as input for a predefined prediction model. The data used here can be point data or individual data or data streams.

[0012] Data fusion already enables an improved, for example more accurate or more reliable prediction, i.e. an estimate, of the driving path, for example compared to a prediction of the driving path based on a single data or data source. However, a scenario-driven approach is also proposed here. The identification of the corresponding environmental scene can usually be simpler, more accurate and more reliable than the prediction of the driving path itself. The allocation or selection of credible data and data sources, i.e. the determination of the credibility of different data and / or data sources for different environmental scenes, can allow for a broad understanding to be obtained independently of the situation, which therefore cannot be recognized in real time, i.e. based on current sensor or measurement data, for example, for the corresponding motor vehicle itself. Therefore, for example, compared to conventional approaches, for example, based only on the corresponding available data without further information, the weighted sum of the data related to the corresponding environmental scene, thereby taking into account the different characteristics and situations of the different environmental scenes, can enable a more robust, more accurate and more reliable prediction of the driving path.

[0013] In a possible configuration of the present invention, the data and / or data source include different data and / or data sources, for example, depending on the corresponding or local availability, i.e., for example, in relation to the equipment of the motor vehicle, foreign vehicles or traffic participants in the corresponding surroundings of the motor vehicle or in the corresponding environmental scene, the local traffic infrastructure, the existence or integrity of the corresponding data, etc. This can be pre-given map data. This can be a corresponding SD or HD map. Such map data can, for example, contain, i.e., give the curvature of the path point ahead in the driving direction for the motor vehicle. Based on such path points and / or curvatures, the possible forward trend of the corresponding road can be determined. This trend can in turn be used as the basis for predicting the driving path, i.e., can be included in the prediction of the driving path.

[0014] Likewise, the data and / or data source may include a road model for estimating the road ahead of the road that the motor vehicle is currently or expected to travel and / or an output of such a road model, i.e., the road ahead estimated by such a predetermined road model, etc. Such a road model may represent the road ahead, for example, based on corresponding camera data, map data, corresponding current navigation data, for example from a navigation system of the motor vehicle, etc.

[0015] Likewise, the data and / or data source may be or include lane edge recognition or edge building recognition. Such road edge recognition may, for example, specify or characterize a lane edge or a lane building in terms of position, trend, type, etc. Such edge building or lane edge recognition may be provided by different sensors or by a fusion of different sensors or measurement data, such as camera data, lidar data, radar data, etc. Lane edges or edge buildings in the present sense may, for example, be generated by guardrails, curbstones, grass, bridge pillars, and / or the like.

[0016] Likewise, the data and / or data source may be or include group data that give earlier vehicle movements in the corresponding area. Such group data may be given or summarized, for example, in so-called learning maps, i.e. learned map data, in which the trajectories of group vehicles when passing through each area are given or processed at an earlier point in time. Such group data may be called, for example, from a corresponding vehicle-external server device, such as a computing center, a cloud server, a backend, etc.

[0017] Likewise, the data and / or data source may be or include a real-time trajectory at a corresponding point in time, i.e., for example, simultaneously with the motor vehicle or a road user moving directly in front of the motor vehicle in a corresponding environment scene. Such a real-time trajectory may be detected, for example, by means of an environment sensor device of the motor vehicle and / or obtained by vehicle-to-vehicle communication sent by other road users.

[0018] Likewise, the data and / or data source may be or include at least one maneuvering hypothesis of an assistance system of the motor vehicle. Thus, such an assistance system may, for example, create a prediction of the movement or maneuvering of the motor vehicle. This may be done, for example, based on a road model, map data, navigation data, the current driving or operating state of the motor vehicle, etc. For example, such a maneuvering hypothesis or prediction may be given as to whether the motor vehicle will continue to travel in the respective currently traveled lane, or will perform a lane change or a turning maneuver, etc. This may directly give or influence the travel path and may therefore be particularly usefully included in the prediction of the travel path.

[0019] Likewise, the data and / or data source may be or include steering data of the motor vehicle. These steering data may, for example, include or specify the current or just passed steering angle, the rate of change of the current or just passed steering angle, etc. Such steering data may directly specify or influence the driving or movement direction of the motor vehicle and may therefore be particularly usefully included in the prediction of the driving path.

[0020] Likewise, the data and / or data source may be or include the current or just-passed yaw rate of the motor vehicle. In other words, the data and / or data source may therefore be or include operating states or vehicle dynamics data of the motor vehicle. For example, a signal may be processed from the steering data or steering movements or steering wheel manipulations of the motor vehicle driver and the yaw rate of the motor vehicle, which signal may specify, indicate or predict the path or course directly ahead of the motor vehicle in the vicinity directly connected to the respective current position of the motor vehicle. Such a signal may therefore be particularly usefully included in the prediction of the travel path. In this context, the vicinity in the present sense may extend from the motor vehicle in the travel direction to a distance of, for example, 10 m, 20 m, 30 m, etc.

[0021] Likewise, the data and / or data source may be or include a corresponding current driving path prediction of a machine learning device. Thus, such a device may, for example, be or include an artificial neural network trained to predict a driving path. This may particularly involve a device of a motor vehicle. A driving path prediction based on machine learning may, for example, be achieved by correspondingly analyzing, evaluating or processing sensor-based object detection. Thus, in other words, a corresponding device may generate a specific trajectory of a possible path for a motor vehicle. The driving path prediction may be based on data and / or data sources other than the data and / or data sources mentioned here or only on a portion of them, but processed according to other principles. Therefore, such a driving path prediction may constitute a useful factor in the final optimized prediction of the driving path.

[0022] The data and / or data sources presented here as input for predicting the driving path may have different advantages and disadvantages in different environmental scenarios and thus overall enable a particularly robust, accurate and reliable prediction of the driving path in each of a plurality of different environmental scenarios.

[0023] In another possible configuration of the present invention, the credibility is at least partially obtained from a predetermined map, in which a location-specific credibility is given for at least one data source and / or data. Therefore, in such a map, the credibility of at least one data source or data type in the area or at the location is given for at least one determined location or spatial area, that is, when a sensor such as a sensor for detecting an object or a road trend is used there. Therefore, the credibility can be obtained particularly simply and quickly. This can support the real-time application of the method according to the present invention. The map used here can contain particularly robust and reliable given reliability data particularly simply, because the map can be generated separately and is therefore not limited by the respective restrictions of the environmental sensor device of the motor vehicle under various conditions or the environmental conditions given then. In addition, a plurality of different information can be included in such a map, which can also be obtained in different conditions, under different conditions and at different times. The use of the map proposed here can particularly simply take into account the unique characteristics of the location. This can be useful, because the same type of traffic management or infrastructure functions, such as roundabouts, can also have unique characteristics and special features, which cannot be considered or can only be considered less accurately and in detail for the reliability in the area of ​​such characteristics through the generally fixedly given rules and regulations.

[0024] In another possible configuration of the present invention, during the operation of the motor vehicle, environmental data are recorded by means of an environmental sensor device of the motor vehicle, which environmental data characterize the corresponding environmental scene. For example, such environmental data can be recorded when approaching the corresponding environmental scene and / or in the corresponding environmental scene, that is, when passing the environmental scene. Then, based on these environmental data, the credibility is at least partially obtained on-site or in real time, that is, dynamically in the operation of the motor vehicle. In other words, these credibility are then re-asked each time the corresponding environmental scene is approached or each time the environmental scene is passed. Therefore, these credibility can take into account the unique actual situation as well as the given characteristics of the motor vehicle and the corresponding changes. Therefore, for example, the dynamic influence or the influence of the environmental sensor of each motor vehicle that changes over time, such as weather conditions, vegetation changes, pollution, wear or gradually occurring correction errors, etc., can be automatically considered. Therefore, in particular, compared with the fixedly predetermined, that is, static credibility, a prediction of the driving path that better matches the actual situation can be achieved. In order to determine the credibility, for example, predetermined parameters of the environmental data or of various individual sensors used to record the environmental data can be determined and taken into account, such as the achievable resolution, the various variances or distributions of the measured values ​​or data points, signal jitter, error rate, the occurrence or frequency or intensity of artifacts, the signal-to-noise ratio, etc.

[0025] In another possible configuration of the present invention, the distance from the corresponding current position of the motor vehicle to is also obtained according to the obtained corresponding environmental scene, and at least one data source and / or at least a part of the data is used to predict the driving path. The data source or the data is then used, for example, only to predict the driving path to each determined distance. Here, a unique distance can be obtained for different data sources and / or for different data. The trend of the driving path beyond this distance can then be based on other or remaining data sources and / or data. The configuration proposed here of the present invention can achieve a particularly reliable and accurate prediction of the driving path. For example, if different sensors used as data sources or providing data have different effective ranges, this may be the case. Therefore, for example, the first sensor data and the second sensor data provided by the first sensor and the second sensor can be fused to predict the driving path to a determined distance from the motor vehicle. For predicting the driving path in the area after or from this distance, for example, the first sensor data provided by the first sensor can be ignored.

[0026] In another possible configuration of the present invention, only such data sources and / or data are included in the fusion and prediction of the driving path, and the credibility of the data sources and / or data obtained for this corresponds to at least one predetermined minimum credibility, that is, at least as large as a predetermined threshold. In other words, it is possible to discard data sources and / or data whose credibility is less than the predetermined minimum credibility, that is, lower than the predetermined threshold, in a regional manner when necessary. Thus, the unpredictable and possibly unreasonable distortion of the driving path prediction caused by particularly untrustworthy data sources and / or data can be avoided or reduced. If the credibility of at least one predetermined number or amount of data sources and / or data corresponding to the predetermined minimum credibility is retained, at least the configuration proposed here of the present invention can be applied. Otherwise, one, more or all of the data sources and / or data whose credibility is lower than the predetermined minimum credibility can also be included in the fusion or prediction of the driving path. Thus, even in unfavorable conditions or marginal situations, it is also possible to ensure that the driving path is predicted based on data.

[0027] In another possible configuration of the present invention, the unreliability or quality of at least a portion of the data is additionally determined. Then, these data are also weighted according to these unreliabilities or qualities, so that greater unreliability, i.e., lower quality, leads to smaller weights. Therefore, the credibility can be based on basic situations, such as geometric limitations or structural situations. In contrast, the unreliability additionally considered here can be the unique characteristics of the corresponding data actually recorded or measured, such as the corresponding clarity, the corresponding signal-to-noise ratio, the corresponding error or artifact rate, etc. Therefore, if necessary, unlike the credibility, the unreliability can also fluctuate randomly, for example, and it is different even when facing the same environmental scene many times. Therefore, the unreliability can be a measurement unreliability that is uniquely generated respectively. Through the configuration proposed here of the present invention, it is finally possible to achieve a particularly robust, accurate and reliable prediction of the driving path as well as situation matching or situation-specific prediction.

[0028] In another possible embodiment of the invention, objects relevant for the guidance of the motor vehicle are determined or selected in the corresponding surroundings of the motor vehicle based on the predicted corresponding driving path. In this case, based on the driving path, for example, objects recognized in the corresponding surroundings of the motor vehicle can be determined, such as other traffic participants, auxiliary or automatic longitudinal or distance control for the motor vehicle, etc., which are more relevant. For this purpose, objects can be classified as relevant, for example, if they are located within the predicted driving path, at least if the objects are at most at a predefined distance from the motor vehicle, are not far away from the motor vehicle, etc. This selection of relevant objects allows for particularly robust, reliable and safe, as well as particularly comfortable vehicle guidance.

[0029] Another aspect of the present invention is an assistance system for a motor vehicle. The assistance system according to the present invention has an interface for detecting various data that can be used to predict the driving path of the motor vehicle, a processing device, such as a microchip, a microprocessor or a microcontroller, and a computer-readable data memory coupled thereto. The assistance system according to the present invention is configured to perform the method according to the present invention, in particular automatically. For this purpose, a run or computer program can be stored in the data memory, which encodes or implements the method steps, measures or processes or corresponding control instructions described in conjunction with the method according to the present invention. Then, the run or computer program can be executed by means of a process device in order to execute the corresponding method or cause its execution. The assistance system according to the present invention can be configured as a controller or a computer module, for example. The assistance system according to the present invention can also include an environmental sensor device, a communication device, a device or module for generating or outputting a control signal, etc. The assistance system can be configured to store the predicted corresponding driving path in a data memory and / or output or provide it via an interface or another interface.

[0030] Another aspect of the invention is a motor vehicle having an environment sensor device for recording or detecting environment data that can characterize the surrounding scene ahead and can be used in particular for predicting the driving path, and an assistance system according to the invention. The assistance system can be coupled to the environment sensor device, for example, via an onboard network of the motor vehicle. The motor vehicle according to the invention can in particular be or correspond to the motor vehicle mentioned in conjunction with the method according to the invention and / or in conjunction with the assistance system according to the invention. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Further features of the invention are apparent from the claims, the drawings and the description of the drawings. The features and feature combinations mentioned above in the description of the drawings and the features and feature combinations mentioned below in the description of the drawings and / or shown individually in the drawings can be used not only in the respectively given combination but also in other combinations or individually without departing from the scope of the invention.

[0032] The accompanying drawings show:

[0033] Figure 1 : A schematic overview diagram of a portion of a first traffic scene for intuitively illustrating a method for driving path estimation; and

[0034] Figure 2 : A schematic overview of a portion of a second traffic scene, used to intuitively illustrate the method. DETAILED DESCRIPTION

[0035] In the figures, identical or functionally identical elements are provided with the same reference symbols.

[0036] In road traffic, the prediction of driving paths can provide a useful basis for many different assistance functions or automation. In this regard, it is desirable to predict driving paths as robustly, accurately and reliably as possible for a situation-appropriate reaction. One approach is to use and fuse different data. However, it has been shown that not all available data sources or data types contribute to the prediction of driving paths with the same usefulness and accuracy in every situation or every traffic or environmental scenario. In other words, different data sources and data types can have different credibility for the prediction of each driving path in different environmental scenarios.

[0037] For intuitive explanation, Figure 1An exemplary partial overview of a first traffic scene or environmental scene is shown. Here, a partial view of a curve formed by a road 1 is shown, on which a motor vehicle 2 moves. An external vehicle 3 is traveling in front of the motor vehicle 2. The motor vehicle 2 has an environmental sensor device 4 for recording environmental data that characterizes the corresponding environment and thus the corresponding environmental scenario. In addition, the motor vehicle 2 is equipped with an auxiliary system 5 for predicting or estimating the forward driving path of the motor vehicle 2. The auxiliary system 5 comprises, for example, an interface 6, a processor 7 and a data memory 8. Via the interface 6, for example, data recorded by means of the environmental sensor device 4 can be detected. This can be data of a variety of individual sensors of the environmental sensor device 4, which can be of different types in particular. Other data, such as the position and operating state of the motor vehicle 2, movement or trajectory data of the external vehicle 3, which can be received, for example, via a vehicle-to-vehicle data connection, group data and / or map data called from a server device outside the vehicle, etc. can also be detected via the interface 6.

[0038] The prediction of the driving path can in principle be based on individual data from the acquired data. However, for example, the prediction of the driving path based on the current steering angle of the motor vehicle 2 can lead to an incorrect estimation 9 of the driving path, which is schematically shown here.

[0039] Therefore, first, for example, a currently given environmental scene is determined based on at least a portion of the detected data. The detected data are then fused together to obtain an improved prediction or estimate of the driving path. This depends on the correspondingly determined environmental scene. Therefore, it is taken into account here that different data can have different credibility in each environmental scene. Therefore, for example, due to the curvature of the road 1, the camera data can only image the course of the road 1 to a limited extent. In contrast, the road course 10 of the road 1 can be obtained with a higher credibility, for example from map data or estimated based on the trajectory data of the foreign vehicle 3 in front. Therefore, the various detected data are weighted according to their credibility when they are fused. Here, for example, the credibility can be called from a pre-given database stored in the data memory 8, in which the credibility of different data or data sources for various environmental scenes is given. It is also possible, for example, to dynamically determine the credibility at least partially based on the characteristics of specific, correspondingly concrete detected data. Then, a correspondingly improved driving path estimate 11 is obtained as a result.

[0040] However, the absolute and / or relative credibility, as well as the availability of various data or the corresponding data sources, may be different in other environmental scenarios. Figure 2For example, a schematic overview of a part of another environmental situation is shown. Here, another part of the road 1 is shown. In the environmental scenario shown here, the motor vehicle 2 leaves the road 1 at the exit 12. Here, for example, a prediction or estimation of the driving path based only on the steering angle can also lead to an erroneous estimate 9. Similarly, the corresponding prediction or estimation can lead to such an erroneous estimate 9, for example, based on camera data that is limited in its effective range or map data that may be incomplete, not current or inaccurate, etc. However, for example, the roadway edge 13, such as a guardrail, is detected here with a high degree of confidence by the front radar device of the motor vehicle 2, for example, due to the arrangement or geometry of the guardrail given in the detection range of the front radar and the generally reliable recognizability. Correspondingly, here, for example, the steering angle can be less reliable and therefore not used or only used for a part of the driving path estimation 11 in the vicinity of the motor vehicle 2 and, in contrast, is not weighted as strongly as the more reliable radar data of the front radar device, which can be weighted accordingly more highly and can also be used for another part of the driving path estimation 11 that is away from the motor vehicle 2 in the driving direction.

[0041] Likewise, there may be a plurality of other basic types of environmental scenarios as well as a plurality of unique, i.e. location-specific, environmental scenarios for which corresponding individual plausibility scores for different data sources and / or data or data types can be determined, i.e., for example, called up from corresponding regulations or dynamically determined and used for corresponding predictions or estimates of the travel path.

[0042] In general, the described examples illustrate how scenario-driven use of multimodal sensor data can be implemented to estimate a vehicle's travel path.

[0043] Reference numerals list

[0044] 1 Road

[0045] 2 Motor Vehicles

[0046] 3 Foreign vehicles

[0047] 4 Environmental Sensors

[0048] 5. Auxiliary systems

[0049] 6 Interface

[0050] 7 Processor

[0051] 8 Data storage

[0052] 9. Miscalculation

[0053] 10Road Trends

[0054] 11Driving Path Estimation

[0055] 12Exit

[0056] 13. Edge of carriageway

Claims

1. A method for predicting a travel path (11) of a motor vehicle (2), wherein during operation of the motor vehicle (2) - obtaining a corresponding environmental scene ahead of the current driving direction of the motor vehicle (2), - determining the credibility of a plurality of different data sources and / or the credibility of data originating from said data sources for the determined corresponding environment scenario, based on which the driving path (11) can be predicted, and - according to the ascertained plausibility, a plurality of the data from the different data sources are weightedly combined with one another and the driving path is predicted thereby (11).

2. The method according to claim 1, It is characterized in that The data and / or data sources include predetermined map data, a road model for estimating the road course ahead (10), an estimated road course ahead (10), lane edge recognition (13), group data of vehicle movements given in advance, real-time trajectories (10) of other traffic participants (3) moving in the corresponding environmental scene at a corresponding time, operating assumptions of an auxiliary system of the motor vehicle (2), steering data of the motor vehicle (2), a yaw rate of the motor vehicle (2) and / or a driving path prediction of a machine learning device.

3. The method according to any one of the preceding claims, It is characterized in that The plausibility is at least partially acquired from a predefined map in which a location-specific plausibility is specified for at least one data source and / or data type.

4. The method according to any one of the preceding claims, It is characterized in that During operation of the motor vehicle (2), environmental data characterizing the corresponding environmental scene are recorded by means of an environmental sensor device (4) of the motor vehicle (2), and the plausibility is at least partially determined dynamically based on the environmental data.

5. The method according to any one of the preceding claims, It is characterized in that A distance from the current position of the motor vehicle (2) is ascertained as a function of the respectively ascertained surrounding scenario, and at least one of the data sources and / or at least a portion of the data is used to predict the driving path (11).

6. The method according to any one of the preceding claims, It is characterized in that Only those data sources and / or data whose plausibility corresponds to at least one predefined minimum plausibility are taken into account in the fusion and prediction of the driving route (11).

7. The method according to any one of the preceding claims, It is characterized in that In addition, an uncertainty of the data is ascertained for at least some of the data, and the data are also weighted as a function of the uncertainty, so that a greater uncertainty results in a lower weighting.

8. The method according to any one of the preceding claims, It is characterized in that Based on the predicted driving path (11), an object (3) relevant to the guidance of the motor vehicle (2) is selected in the corresponding environment of the motor vehicle (2).

9. An assistance system (5) for a motor vehicle (2), comprising an interface (6) for detecting a plurality of data that can be used to predict a driving path (11), a processor device (7) and a computer-readable data memory (7) coupled to the processor device, wherein the assistance system (5) is configured to execute a method according to any one of the preceding claims.

10. A motor vehicle (2) having an environment sensor system (4) for recording environment data characterizing a surrounding scene ahead and having an assistance system (5) according to claim 9.

Images