Teaching of trajectories in motor vehicles

Abstract

The invention relates to a method for operating a motor vehicle (10), having: - operating the motor vehicle (10) in a teaching mode, in which the driver can drive through and thereby teach a trajectory (T); - operating the motor vehicle (10) in a tracking mode, in which the motor vehicle (10) at least partially autonomously tracks the trajectory (T) or an adapted variant thereof (TA); wherein in the teaching mode a maximum adjustable steering angle can be limited, and this steering angle limit can be variably set depending on the driving situation. Furthermore, the invention relates to a controller (12) for implementing such a method.

Description

TECHNICAL FIELD

[0001] The invention relates to a controller and a method for operating a motor vehicle. The motor vehicle can be a passenger car or a truck, in particular. In general, the invention is directed to the field of at least partially autonomously supporting a driver of a motor vehicle when carrying out a desired driving maneuver. BACKGROUND

[0002] From DE 10 2014 220 114 A1 a solution is known, in which a driver can learn (einlernen) a desired trajectory, i.e. a theoretical movement path along which the vehicle shall move. To this end, the driver can drive through in a learning mode and thereby predefine the respective trajectory. Subsequently, the vehicle shall autonomously drive through the trajectory. The background is that the driver can thus let the motor vehicle autonomously carry out a parking process to be so frequently performed.

[0003] DE 10 2014 220 144 A1 has taught here that for autonomously driving through (Abfahren) or following (Nachfahren) a trajectory it is important that then a relatively large steering angle is available for the vehicle to use. This is required, for example, to compensate for adjustment differences. But if the driver has adjusted a correspondingly large and in particular maximum steering angle when learning, the scope of play for compensating adjustment differences or for adjusting a sufficiently large additional steering angle when autonomously following the trajectory is correspondingly reduced.

[0004] DE 10 2014 220 144 A1 therefore proposes to limit the adjustable steering angle during the learning operation, for example via a counter torque generated in time in the steering system.

[0005] It has proven that this solution can lead to a loss of comfort from the driver's point of view and more precisely to a steering behavior that is perceived as unnatural during the learning operation. SUMMARY

[0006] It is therefore the task of the invention to improve the learning of a trajectory from the driver's point of view.

[0007] This task is solved by the subject matter of the attached independent claims. Advantageous refinements are specified in the dependent claims.

[0008] It has been recognized that the driver can perceive the steering behavior of the motor vehicle as unnatural due to the previously fixed limitation of the adjustable steering angle in the learning mode. In other words, the driver can perceive the steering behavior in the learning mode as an unnatural deviation from the steering behavior of the motor vehicle in the general driving operation, in which the corresponding limitation typically does not exist, due to the artificially limited steering angle.

[0009] Thus, in contrast to the prior art, the application proposes that the actual requirements for the steering angle limitation are examined in a teaching run (hereinafter also referred to as teaching mode) and / or the steering angle limitation is dynamically adapted to the current driving situation. In other words, the steering angle limitation is preferably only set when this is actually required. Similar to the prior art, a corresponding requirement can thus arise that, in comparison to the teaching mode, an additional steering angle reserve is still available for use when tracking the taught trajectory, for example, in order to be able to compensate for adjustment differences when tracking the trajectory. However, an embodiment of the application provides that it is investigated whether the taught trajectory can be adapted and / or changed afterwards, with knowledge of the current driving situation, such that the trajectory can be tracked at least partially autonomously even without a corresponding steering angle reserve or with at least a reduced steering angle reserve. It is thus not mandatory to directly limit the steering angle in the teaching mode or such a limitation can be reduced.

[0010] In particular, a method for operating a motor vehicle is proposed, which has

[0011] - operating the motor vehicle in a teaching mode (or teaching run), in which the driver can drive through and thus teach a trajectory;

[0012] - operating the motor vehicle in a tracking mode, in which the motor vehicle tracks the trajectory or an adapted variant thereof (for example, adapted afterwards by smoothing) at least partially autonomously;

[0013] wherein the maximum adjustable steering angle can be limited in the teaching mode and this steering angle limitation can be variably set depending on the driving situation.

[0014] For teaching the trajectory, but also for tracking the trajectory, all methods known from the prior art can be used. For example, for the purpose of teaching, the start position and / or the target position can be stored. Furthermore, a speed profile, a steering angle and / or location information depending on the time can be stored to describe the trajectory, in particular a route section between the start position and the target position.

[0015] For tracking or in other words driving through the trajectory, the motor vehicle can adjust the corresponding steering angle and / or speed depending on the location at least partially autonomously. For this purpose, the control device can access or actuate the corresponding actuators in the vehicle in the evaluation of the information taught in the teaching mode and in particular stored.

[0016] The steering angle can be understood as any angular quantity which describes the steering behavior of the motor vehicle. For example, it can be the steering wheel angle or a wheel steering angle measured at the vehicle wheels. The rack position or a general operating quantity of the steering system from which the corresponding steering behavior of the motor vehicle (hereinafter also only referred to as vehicle) can be inferred can also be used as the steering angle.

[0017] To limit the steering angle, methods known in the art can be used. For example, an electric counter torque can be applied, which counteracts the steering movement (e.g. acting on a rack in an electromechanical steering device) preset by the driver. In the case of a steer-by-wire steering system, a counteracting actuator can generate a corresponding counter torque and / or an actuator coupled with the rack can not convert the steering wheel angle into a corresponding steering-effective movement at the vehicle wheels.

[0018] Therefore, to limit the steering angle, at least one control device of the motor vehicle can be manipulated and / or an output control signal can be output in order to cause the limitation.

[0019] The control device can furthermore detect the driving situation according to any one of the methods described below. Depending on the detected driving situation, the steering angle limitation can be changed and thus dynamically adapted to the current driving situation.

[0020] The steering angle limitation can have at least two states, namely a non-limitation state and a limitation state. In the non-limitation state, the maximum adjustable steering angle can correspond to the technically and / or mechanically and / or system-determined maximum possible steering angle, for example. In the limitation state, the steering angle can have a value below the mentioned maximum possible steering angle. In other words, the limitation of the maximum possible steering angle can be selectively activated or deactivated. This thus corresponds to a setting of a variable steering angle limitation. In particular, a change between these states can be made depending on which driving situation is currently present.

[0021] Generally, if a substantially unrestricted maneuverability of the motor vehicle is present as a driving situation, for example because there are no obstacles in the vehicle environment, the steering angle limitation can be dispensed with and / or the steering angle limitation can generally be low. In this case, it can be assumed that the trajectory for the tracking can be adapted in such a way that the trajectory can be driven through with a steering angle below the maximum possible steering angle. The remaining difference between the steering angle required to drive through the adapted trajectory and the maximum possible steering angle is then reserved as a still available gap range for changing the steering angle to compensate for possible adjustment differences.

[0022] If, on the other hand, the maneuverability of the vehicle is affected, for example due to the presence of obstacles and / or so-called narrow sections in the vehicle environment, the steering angle limitation can be activated. The maximum adjustable steering angle is then preferably below the maximum possible steering angle. This takes into account the case that the trajectory can only be adapted within a reduced range or not at all as a result of the smaller remaining maneuvering gap range. It is therefore also not ensured that the steering angle required to track the trajectory can be reduced by adapting the trajectory, and the described steering angle gap range is reserved to compensate for possible adjustment differences.

[0023] The solution presented here is generally advantageous in that the steering angle limitation is more in line with the requirements. In particular, the steering angle limitation is only activated when it is actually required to ensure sufficiently precise tracking of the taught trajectory. The steering behavior of the vehicle in the teaching mode is thereby improved from the driver's perspective, since the steering behavior is perceived as less limited.

[0024] One refinement provides that the driving situation is detected (and / or evaluated) in the event of the use of at least one of the following:

[0025] - a surrounding sensor system of the motor vehicle, which for example comprises at least one radar sensor, lidar sensor, distance sensor, ultrasonic sensor or video camera;

[0026] - a communication system of the motor vehicle, in particular a so-called Car2X communication system, which is set up for communication with at least one vehicle-external unit, for example with other vehicles, mobile radio participants, server devices or internet service providers;

[0027] - map information, for example stored in a navigation system of the vehicle or callable from an external server, which describes the vehicle environment;

[0028] - a computer-implemented model, in particular a machine learning model, which is set up to determine the setting of the steering angle limitation;

[0029] - a computer-implemented method for evaluating and / or classifying the driving situation.

[0030] Any of the above-mentioned variants can be used to evaluate the free maneuverability of the motor vehicle. In other words, the driving situation can thus define and / or evaluate to what extent the vehicle can be maneuvered freely, i.e. to what extent obstacles in the vicinity of the vehicle limit the (free) maneuverability.

[0031] By means of the surrounding sensor system, obstacles in the vehicle environment and / or free spaces can generally be detected. In general, the surrounding sensor system can map the environment of the vehicle in a known manner.

[0032] With the communication system, the motor vehicle can obtain information which preferably defines the current maneuverability or in general the freedom of movement of the motor vehicle. For example, the motor vehicle can obtain information from a control device of a parking lot about the occupancy of parking spaces, the presence of narrow sections within the parking lot or in general the lane boundaries within the parking lot. It is likewise possible with regard to individual parking spaces or parking spaces which are not part of a multi-storey car park.

[0033] The map information can relate in a similar manner to information which defines the freedom of movement of the motor vehicle, for example, as a result of the traffic infrastructure and / or the general traffic routing at the motor vehicle's resting place. Thus, the vehicle environment can be described in terms of the map information, inter alia, as to whether there is a possibility of turning, there is a lane branching, there is a lane narrowing or in general which spatial conditions exist along the lane, for example, as a result of structural restrictions.

[0034] The model and in particular the machine learning model can be implemented in a control device of the motor vehicle, wherein the control device is a computer. But the model can also be implemented in a computer device outside the vehicle, which then but preferably transmits the result of the determination of the steering angle limitation to the control device of the motor vehicle (for example, via mobile radio). The model and in particular the machine learning model can obtain information about the driving situation of the vehicle. In particular, the model and in particular the machine learning model can obtain any of the information described herein, by means of which the vehicle environment and / or the driving situation can be described. In particular, the model and in particular the machine learning model can obtain information of the mentioned surroundings sensors, communication systems and / or map information. The model and in particular the machine learning model can use this information as an input variable. The model can output the setting of the steering angle limitation as an output variable. In particular, the model can output whether the steering angle limitation is activated (for example, whether the above-described limited or unlimited state should be activated). Alternatively or additionally, the model can output the degree of the steering angle limitation, for example, which value the maximum adjustable steering angle should take in particular.

[0035] Instead of a calculator-based model, a general computer-implemented method and / or a method for detecting and in particular evaluating and / or classifying the driving situation can also be used. The steering angle limitation can generally be determined via a characteristic line, a characteristic field or a table value.

[0036] The optional machine learning model can be trained and / or created on the basis of training data sets and in the framework of a machine learning process. These training data sets can describe driving situations and the steering angle limitations present there. For example, they can be objective and preferably validated driving situations, in which the steering angle limitation is evaluated as appropriate or inappropriate. On the basis of such training data sets, the machine learning model can learn a relationship between the driving situation or the variables describing the driving situation and the steering angle limitation felt to be objectively appropriate there.

[0037] Generally, models and in particular machine learning models can mathematically model a relationship between input quantities and output quantities. In the framework of a learning process or training process, the connection between these quantities and / or the weights of the quantities and / or the so-called intermediate layers can be defined, set or in other words learned. In a manner known per se, the machine learning model can here link the input quantities and the output quantities via different layers, wherein the layers each have nodes which are linked via a connection to the nodes of an adjacent layer. In particular, the machine learning model can be an artificial neural network or be based on such a network.

[0038] Generally, at least one driving situation parameter can be obtained which describes the current driving situation. Within the scope of the method, it can be provided to check whether the driving situation parameter meets a predetermined steering angle limitation criterion. If this is the case, the steering angle limitation can be activated and / or limited to a predetermined value, in particular a value below the maximum possible steering angle. If this is not the case, the respective limitation can be stopped. The driving situation parameter can be, for example, the minimum distance of the motor vehicle to the environment and / or to obstacles in the environment.

[0039] Alternatively or additionally, the driving situation parameter can be, for example, a classifier which is stored in a map and / or generally depends on the location or comprises such a classifier. This driving situation parameter can describe, for example, the current location, the driving situation there and / or the degree of difficulty of the driving maneuvers required there (for example at a narrow site). Such a parameter and in particular the classifier can be transmitted to the controller which implements the method and / or determined or read out by the controller. For this purpose, a software interface to the map information and / or to a map server external to the vehicle, for example Map / Car2X, can be used.

[0040] For example, the necessity of a steering angle limitation can be identified or learned by the method and / or the controller, and the information or parameters required for determining the degree of the steering angle limitation are preferably obtained via the software interface described above. These information can optionally be linked to the dimensions of the vehicle (for example, a relatively low steering angle limitation can not be required or can be required in a small vehicle).

[0041] In particular, one refinement provides that the maximum adjustable steering angle is set to a reduced value or kept at a predetermined value below the maximum possible steering angle (determined by the system and / or technically possible) if the driving situation corresponds to a maneuver without maintaining a minimum distance to an interference profile in the environment. Interference profiles in the environment can generally be understood to be structures in the environment that are relevant to a collision and / or obstacles or objects that are relevant to a collision. Here, these can be, in particular, architectural structures, such as walls or pillars. Alternatively or additionally, interference profiles can also be classified, that is to say a limitation of the steering angle below the minimum distance to the interference profile can be made depending on the type or class of the interference profile. For example, a fixed obstacle, such as a wall or a pillar, can lead to a stronger limitation of the steering angle than, for example, a vehicle parked on the driving route, which can no longer be stopped in the future when driving through in follow-up mode.

[0042] The possibility of changing the taught trajectory retroactively is reduced if the minimum distance to such a structure is undershot. Such a change can then lead to a risk of a collision with the interference profile when the vehicle is following the adapted trajectory. Therefore, the taught trajectory should be driven through as accurately as possible, and therefore, similar to the prior art and in order to compensate for retroactive adjustment differences, a limitation of the steering angle is preferred.

[0043] Alternatively or additionally, it can be provided that the maximum adjustable steering angle is set to an increased value or kept at a predetermined value if the driving situation corresponds to a maneuver in the minimum distance to an interference profile in the environment.

[0044] The above-mentioned increasing and decreasing values refer to a change in the maximum adjustable steering angle compared to the driving situation that existed before the driving situation currently under consideration. In particular, it is possible to start without a limitation of the steering angle at first and then to reduce the maximum adjustable steering angle. It is also possible to change from a previously limited maximum adjustable steering angle to a corresponding increased value, which has a smaller or no limitation.

[0045] If the minimum distance to an interference profile in the environment is maintained, the maximum adjustable steering angle can be increased, in particular, to the maximum possible steering angle (for example, determined by the system and / or technically). This corresponds to a deactivation of the steering angle limitation, because then the maximum possible or available steering angle is available for use. The maximum adjustable steering angle, that is to say the steering angle that can be adjusted or preset by the driver, then corresponds to the maximum possible steering angle determined by the system or technically.

[0046] The maintenance of the minimum distance to the disturbance profile can in turn be assessed by means of the surrounding environment sensor system, but also by means of map information, if these are compared with the current position of the vehicle. However, likewise the presence and location of disturbance profiles in the environment can be transmitted to the motor vehicle via the communication system of the motor vehicle, if present, and then the minimum distance to these disturbance profiles can be checked.

[0047] As already mentioned, it can be provided that the taught trajectory is adapted at least partially before the tracking. In particular, this can be done if in the teaching mode no steering angle limitation and / or a set increase in steering angle is carried out in the respective section. The trajectory can then be adapted in such a way that it can be tracked with a reduced steering angle, for the reasons mentioned above, in order to ensure the possibility of compensating for the adjustment differences. In other words, the trajectory can be smoothed.

[0048] One possibility for this is to increase at least one curve radius (or in other words, the radius of curvature) of the trajectory. Additionally or alternatively, for example, the entry point (Einlenkpunkt) (in particular before the respective curve or curvature) can be displaced forward along the trajectory, while or the exit point (Auslenkpunkt) (for example, from the curve and / or curvature) can be displaced backward along the trajectory. The forward and backward direction designations here refer to the driving direction from the starting position to the target position of the trajectory (where the position is further displaced forward in the direction of the target position). The entry point and the exit point can be distinguished by the fact that the steering angle changes there beyond a pre-set threshold value. Preferably, as a standard, it can also be defined that this change or the steering angle adjusted thereby remains constant and / or is at least not reduced beyond a predetermined limit for a certain driving distance. For example, the entry point can be distinguished by the fact that a larger steering angle is adjusted compared to straight driving or driving with a smaller steering angle. The exit point can be distinguished by the fact that the steering angle is reduced and in particular a driving with a smaller curve radius results therefrom compared to driving with a larger steering angle or in general curve driving.

[0049] The respective displacement of the entry point or the exit point is a simple possibility for smoothing the trajectory. In particular, the risk is thereby reduced that the driver perceives the adaptation of the trajectory as unnatural or an inappropriate deviation from the actually taught trajectory.

[0050] In general, it can be provided in this regard that the adaptation of the trajectory takes place in dependence on environmental information detected in the teaching mode. In particular, the environment can be mapped in the teaching mode, for example by detecting the environment by means of the surroundings sensor system. In this way, the adaptation can take place in consideration of the relevant and / or current environmental conditions. The environmental information can be taken into account, for example, in such a way that it is checked whether a displacement and / or in general a change in the curve radius into and / or out of the turning point leads to a collision risk with the vehicle environment. Alternatively or in addition to the teaching mode, the respective environmental information can also be obtained by any of the other variants described herein. For example, these environmental information can be obtained via a communication system of the vehicle and / or in dependence on map information.

[0051] In summary, a maximum permissible adaptation of the trajectory can be defined in dependence on the environmental information. Additionally or alternatively, the adaptation of the planned trajectory can be checked in view of a collision risk resulting from the adaptation on the basis of the environmental information. If there is no collision risk, the adaptation can be assessed as permissible. Whereas if a collision risk is determined, the adaptation can be assessed as invalid and not implemented.

[0052] The application also relates to a controller for a motor vehicle, which is set up to carry out the method according to any of the preceding claims. The controller is set up to obtain information about the driving situation from all units described herein. For example, the controller can be connected to the surroundings sensor system of the motor vehicle, possibly to a communication system and / or to a unit providing map information. Likewise, the controller can carry out the machine learning model and / or be connected to a device carrying out the machine learning model. Furthermore, the controller can be set up to output parameters and in particular control signals about the steering angle limitation. For example, the controller can actuate an actuator of the steering system in order to set and / or implement the steering angle limitation. In particular, the controller can actuate the actuator to generate the counter torque described herein in order to thereby implement the steering angle limitation.

[0053] In general, the controller can comprise a processor device and / or a memory device. On the memory device, program instructions can be stored, which, when implemented by the processor device, cause the controller to carry out the method according to any of the aspects described herein. BRIEF DESCRIPTION OF DRAWINGS

[0054] Embodiments of the application are explained below with the aid of the attached schematic drawings:

[0055] Figure 1 A vehicle is shown in a schematic top view, which is operated in a teaching mode and has a controller according to one embodiment of the application.

[0056] Figure 2 is according to Figure 1 Schematic diagram of the post-hoc adaptation of the taught trajectory.

[0057] Figure 2A is a schematic illustration of a post-hoc adaptation of an alternative trajectory.

[0058] Figure 3 is a flow chart of a method according to the application, which is implemented by a controller in Figure 1 . DETAILED DESCRIPTION

[0059] In Figure 1 , the vehicle 10 (passenger car) is shown in a highly simplified top view. The vehicle 10 is in an environment 100, which is limited laterally from the vehicle's perspective by building structures 102. These building structures 102 form an interference profile in the vehicle's environment. Exemplarily, it is a wall of a parking lot. In the starting position S shown, the vehicle 10 has a relatively large spacing A to these interference profiles. Exemplarily, only a single spacing A is shown here, however, this spacing can also be defined as a spacing circle around the vehicle 10 with a radius A, or can take on any other profile, which represents the spacing A of the vehicle 10 to the building structures at points relevant to the driving situation.

[0060] The vehicle 10 comprises a controller 12. Furthermore, the vehicle comprises a surrounding sensor system 14, which is connected to the controller 12 in a data transmission manner. The surrounding sensor system has in the example shown (exemplarily only) two surrounding sensors 16, which are, for example, spacing sensors. Preferably, a plurality of corresponding surrounding sensors 16 are provided in order to detect the spacing to the environment at all vehicle sides and thus to reliably map the vehicle's environment with respect to the interference profile. In a manner known per se, the spacing relationship to the interference profile 102 can be determined, for example, from the information (environment information) detected by the surrounding sensor system 14, as is mapped by the spacing A in Figure 1 .

[0061] Preferably, the controller 12 is also connected to a server 104 outside the vehicle via a data connection D, which is indicated by a dashed line. Thereby, map information can be obtained, from which the environment information can be derived. For example, the position, the extension and / or the classification of the interference profile in the environment can also be inferred in this way, for example, the position of the wall 102. In the knowledge of one's own vehicle position, the spacing information to the environment or to the interference profile there can also be determined by the controller 12 in this way.

[0062] Alternatively or additionally, the server 104 can comprise or implement a machine learning model. To this end, input quantities can be obtained from the controller 12, e.g. the current vehicle position or any environmental information detected by the surrounding environment sensors. Based on this, the necessity and / or degree of a steering angle limitation can be determined by the machine learning model with knowledge of these input quantities and output to the controller 12.

[0063] Not shown separately is that the controller 12 is connected with an actuator for the steering angle limitation. The actuator can be any of the ones mentioned in the general description part, with which a counter torque can preferably be generated to counteract the manual torque applied by the driver at the steering wheel (not shown).

[0064] In the state shown, the driver has activated the teaching mode, e.g. via a corresponding input into the vehicle control system. In a manner known per se, during the teaching mode, e.g. by the controller 12, the Figure 1 The driven trajectory T is indicated in the figure by a dashed line. The trajectory T can be characterized by steering wheel angles T adjusted by the driver depending on the location, or these steering wheel angles can be stored depending on the location. In the example shown, the vehicle 10 first arrives in the region B1. The controller 12 determines here on the basis of the above-mentioned information and in particular any of the environmental information whether there is sufficient maneuverability of the vehicle 10. In particular, it is determined for this that the distance A to the disturbing profile in the environment, i.e. to the wall 102, exceeds a predetermined minimum value. In the region B1, this is the case. The controller 12 therefore determines that no steering angle limitation is required. The maximum possible steering angle adjustable by the driver thus corresponds to the system-determined maximum possible or available steering angle without limitation. In other words, during the teaching of the trajectory T, the driver can steer the vehicle 10 in the region B1 maximally and without limitation.

[0065] In the region B2 along the trajectory T, the distance of the environment or the wall 102 to the vehicle 10 decreases significantly. This is indicated in the course of the driven trajectory T in the region B2 and the significantly reduced distance to the wall 102 compared to the region B1.

[0066] This is determined by the controller 12, with the steering angle limitation being activated. The driver can then no longer adjust the system-determined maximum possible / available steering angle to the (manual) maximum adjustable steering angle. Instead, only a reduced maximum possible steering angle is available to the driver. In the example shown, this continues until the target position Z is reached.

[0067] After the end of the teaching mode, the trajectory T is preferably adapted by the controller 12 or alternatively by the external server 104. This process is described in Figure 2The trajectories T, TA are shown in the diagram in FIG. 1 schematically and simplified. First, the original trajectory T is shown as taught by the driver. This trajectory is drawn with a solid line. Also drawn is the course of the adapted trajectory TA, which is shown with a dashed line. Figure 2 The diagram in FIG. 1 is used here to explain the different courses of the trajectories T, TA. The shown superimposed arrangement of these trajectories T, TA has no particular meaning. Rather, in particular, these trajectories T, TA overlap as much as possible, rather than being spaced apart from one another continuously as drawn only for illustration reasons.

[0068] Also shown are the regions B1 and B2 in which there is no steering angle limitation (B1) or a steering angle limitation is enabled (B2). In the region B2 with a steering angle limitation, no adaptation of the trajectory takes place. It can be assumed here that there is still sufficient steering angle reserve for the vehicle to use when autonomously following the trajectory, wherein this reserve corresponds at least to the difference between the limited adjustable steering angle in the region B2 and the system-determined / technically maximum available steering angle. Furthermore, for safety reasons, no adaptation of the trajectory takes place in order not to cause a risk of collision with the environment.

[0069] In the region B1, however, no steering angle limitation takes place. As explained in the general part, this has the advantage that the driver perceives the steering behavior of the vehicle 10 as natural and unrestricted. However, this means that there can be no steering angle reserve in the autonomous driving operation to compensate for possible occurring adjustment differences. This applies in particular in sections of the trajectory in which the driver has actually adjusted the maximum possible steering angle, i.e. has, for example, deflected the steering to the maximum. Currently, there is such a full deflection in the entry steering point E, which is characterized by a change in the steering angle that exceeds a permissible threshold value, for example a change of more than 60%. Subsequently, curve driving with a curve radius R along the trajectory T up to the exit steering point A is carried out, viewed in the driving direction, i.e. from the start point S to the target point Z. From the exit steering point A, a reverse change in the steering angle takes place, which again preferably exceeds a certain threshold value, i.e. for example again more than 60%.

[0070] The trajectory T is now adapted in such a way that the steering angles required for driving through the adapted trajectory TA are limited as much as possible in the region B1. In particular, these steering angles are limited in such a way that the steering angle is preferably below the maximum possible steering angle, i.e. for example a full deflection is not carried out. It can thus be ensured that a certain steering angle reserve remains as it was in order to adapt to adjustment differences when autonomously driving through the adapted trajectory TA.

[0071] In the current example, this is achieved by increasing the curve radius R (to the adapted radius RA) in the curve segment between the entry turning point E and the exit turning point A. Specifically, this is achieved by shifting the entry turning point E further backward (i.e., closer to the starting point S) along the adapted trajectory TA. This results in an adapted entry turning point AE of the adapted trajectory TA. Additionally or alternatively, the exit turning point A can be shifted further forward (i.e., in the direction of the target location Z) along the adapted trajectory TA. This results in the adapted exit turning point AA shown.

[0072] Although there is no steering angle limit in area B1 (which can be advantageous from the driver's perspective), this ensures that the taught trajectory T is sufficiently accurate for drivability in the form of an adapted trajectory TA.

[0073] exist Figure 2A The diagram illustrates another alternative example of trajectory adaptation when traversing an S-shaped track or curve. The trajectory T and its dashed, adapted course TA are plotted superimposed, and the trajectory is adapted along its full length. In this case, the adapted entry turn point AE and exit turn point EE not only move along the original trajectory T but also in a direction not parallel to it. Therefore, it may be necessary to supplement the adapted trajectory TA with a connecting segment VA that connects the starting point S and the target point Z to the adapted entry turn point AE and exit turn point EE.

[0074] exist Figure 3 The text shows the output of a document. Figure 1 A flowchart of an exemplary method implemented by controller 12 in the diagram.

[0075] In step S1, the teaching mode is activated, and vehicle 10 begins to travel along trajectory T. In step S2, which may also be performed simultaneously, surrounding environment information is acquired and / or recorded during the travel of trajectory T (e.g., by means of the surrounding environment sensing mechanism 14). In step S3, which is preferably performed continuously during the travel of trajectory T, it is determined based on the environmental information whether steering angle limitation is required. For this purpose, the distance A between the vehicle and the environment or wall 102 is determined as a standard and / or other standards are considered, such as, for example, the classification of interference profiles. If this distance is below a predetermined threshold, the maneuverability of vehicle 10 is limited, and therefore steering angle limitation is activated. If this is not the case, steering angle limitation is not activated. The activation or deactivation of steering angle limitation can be stored as additional information describing trajectory T.

[0076] In step S4, the vehicle travels along trajectory T and subsequently analyzes that trajectory. Currently, this includes identifying region B1 with steering angle limitations and region B2 without steering angle limitations. In region B1 with steering angle limitations, the trajectory T is then used to adapt to the fitted trajectory TA.Figure 2 Explanatory adaptation. In step S5, the autonomous driving-through mode is activated and the adapted trajectory TA is driven through. There can be a significant time difference of hours or days between steps S1 and S5. In a manner known per se, the driver can thus teach the parking process that frequently occurs from his perspective in step S1 (for example, with regard to a reserved parking space in a parking lot) and in step S5 the parking process is autonomously or independently tracked (i.e., repeated) by the vehicle 10.

[0077] List of reference signs

[0078] 10 vehicle

[0079] 12 controller

[0080] 14 surroundings sensor system

[0081] 16 surroundings sensor

[0082] 100 environment

[0083] 102 wall

[0084] 104 server

[0085] A distance

[0086] B1 region with steering angle limitation

[0087] B2 region without steering angle limitation

[0088] S starting point

[0089] Z target point

[0090] T trajectory

[0091] TA adapted trajectory

[0092] E entry steering point

[0093] A exit steering point

[0094] AE adapted entry steering point

[0095] AA adapted exit steering point

[0096] D data line

[0097] R curve radius

[0098] RA adapted curve radius

[0099] VA connecting section

Claims

1. A method for operating a motor vehicle (10), having - operating the motor vehicle (10) in a teaching mode, in which a driver can drive through and thereby teach a trajectory (T), - operating the motor vehicle (10) in a tracking mode, in which the motor vehicle (10) at least partially autonomously tracks the trajectory (T) or an adapted variant thereof (TA), - in the teaching mode, a maximum adjustable steering angle can be limited, and the steering angle limit can be variably set depending on a driving situation, wherein the steering angle limit has at least two states, namely an un-limited state and a limited state, wherein in the un-limited state the maximum adjustable steering angle corresponds to a system-dictated maximum possible steering angle, wherein in the limited state the maximum adjustable steering angle is smaller than the system-dictated maximum possible steering angle.

2. The method according to claim 1, - the driving situation is detected in the use of at least one of: - a surrounding sensor system (16) of the motor vehicle (10), - a communication system of the motor vehicle (10), which is set up for communication with at least one vehicle-external unit (104), - map information, which describes a vehicle environment, - a computer-implemented model, which is set up to determine a setting of the steering angle limit, - a computer-implemented method, which is used to evaluate and / or classify the driving situation.

3. The method according to claim 2, - if the driving situation corresponds to a maneuver without maintaining a minimum distance to an interference profile (102) in the vehicle environment, the maximum adjustable steering angle is set to a reduced value or kept at a predetermined value below the maximum possible steering angle. wherein 4. The method according to claim 2, - if the driving situation corresponds to a maneuver in a minimum distance to an interference profile (102) in the vehicle environment, the maximum adjustable steering angle is set to an increased value or kept at a predetermined value.

5. The method according to claim 1 or 2, - the taught trajectory (T) is at least partially adapted before tracking, in order to produce an adapted variant (TA).

6. The method according to claim 5, - the adaptation comprises an increase of at least one curve radius (R) of the trajectory (T). characterized in that 7. The method according to claim 5, - the adaptation comprises a forward displacement of at least one entry turning point (E) and / or a backward displacement of at least one exit turning point (A).

8. The method according to claim 5, - the adaptation of the trajectory (T) is carried out taking into account environmental information detected in the teaching mode.

9. The method according to claim 8, - a maximum allowed adaptation of the trajectory (T) is defined in dependence on the environmental information.

10. The method according to claim 2, - the computer-implemented model is a machine learning model.

11. A controller (12) for a motor vehicle (10), which is set up to implement the method according to any one of the preceding claims. ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​ characterized in that ​ ​

Images