Procedure for optimized release / activation of lateral guidance assistance functions from the environment perspective

Abstract

The invention relates to a method for providing, in particular enabling and / or activating, a lateral guidance function (Q) of a vehicle (F), comprising: - Collection of environmental data (UD), - Recording of vehicle data (FD), in particular a vehicle speed (v), - Creating a lane model (SM) to provide the cross function (Q) depending on the environmental data (UD), - Creating a prediction (P) about the safety of the lane model (SM) when providing the lateral guidance function (Q), - Evaluating a history (x_Past, t_Past) when operating the vehicle (F) depending on the prediction (P) and / or the vehicle data (FD), in particular the vehicle speed (v), - Operating, in particular releasing and / or activating, the lateral guidance function (Q) depending on the evaluation.

Description

[0001] The invention relates to a method for providing, in particular enabling and / or activating, a lateral guidance function of a vehicle. Furthermore, the invention relates to a corresponding computer program product, a corresponding control unit, and a corresponding vehicle for implementing a corresponding method.

[0002] Active lateral driving functions (or lateral guidance functions) in vehicle operation should function stably and robustly. This means that if the function can be activated or becomes active, the activity remains active for a certain distance and / or time. The function is not stable if it is activated briefly and then becomes inactive again after a very short distance and / or time due to missing environmental data (e.g. no lane data). Frequent switching between an active and inactive state (so-called active / inactive switching) of the lateral guidance function should be prevented, as this would make the lateral guidance function unstable and / or prevent it from being provided robustly. At the same time, the highest possible availability of the lateral guidance function should be enabled to enable safe and comfortable operation of the vehicle.However, these requirements—enabling high availability and avoiding frequent active / inactive switching—are in conflict with each other. Most solutions avoid frequent active / inactive switching by restricting availability and imposing strict activation conditions.

[0003] The problem arises primarily when environmental data is not recorded consistently or is unstable. This primarily applies to more urban, poorly marked, or complex scenarios. These scenarios are usually driven through at lower speeds (<60 km / h).

[0004] For this reason, known solutions for activating the lateral guidance function often include a speed threshold (usually around 60 km / h or higher).

[0005] Well-known documents are CN 1 09 849 908 A, DE 10 2017 205 245 A1, DE 10 2006 012 997 A1 and DE 10 2021 121 950 A1.

[0006] It is therefore an object of the present invention to at least partially overcome at least one of the disadvantages described above. In particular, it is an object of the invention to provide a method for providing, in particular enabling and / or activating, a lateral guidance function of a vehicle, which enables high availability and avoids frequent active / inactive switching, which increases customer comfort and increases confidence in the lateral guidance function. Furthermore, it is an object of the present invention to provide a corresponding computer program product, a corresponding control unit, and a corresponding vehicle for carrying out a corresponding method.

[0007] The present invention provides a method for providing, in particular enabling and / or activating, a lateral guidance function of a vehicle having the features of the independent method claim. Furthermore, the invention provides a corresponding computer program product, a corresponding control unit, and a corresponding vehicle having the features of the independent claims. Features and details described in connection with the various embodiments and / or aspects of the invention naturally also apply in connection with the other embodiments and / or aspects, and vice versa, so that with regard to the disclosure of the individual embodiments and / or aspects, reference is always made to each other.

[0008] The present invention provides: A method for providing, in particular enabling and / or activating, a lateral guidance function of a vehicle (in particular when the vehicle is in operation or in other words when the vehicle is moving) comprising the following method steps: - Capturing (i.e. sensing, querying and / or receiving) environmental data, (where the environmental data may include, for example, sensed track data and / or received swarm data), - Capturing (i.e. sensing, querying and / or receiving) vehicle data, in particular vehicle speed, - Creating a lane model to provide the cross function depending on the environmental data, (where the lane model may include track sections with associated lane markings, lane boundaries and / or trajectory data that can be used for the lateral guidance of the vehicle along the track sections), - Creating a prediction about the safety of the lane model when providing the lateral guidance function, (where the prediction can evaluate the safety of the lane model, e.g. affirm or deny the safety of the lane model), - Evaluating a history of vehicle operation depending on the prediction and / or vehicle data, in particular the vehicle speed, (where the history may include distances / times driven, which can be evaluated depending on the prediction and / or vehicle speed, thus making it possible to predict whether the vehicle can be safely and stably steered laterally based on its history, current speed and prediction), - Operating, in particular releasing and / or activating, the lateral guidance function depending on the evaluation.

[0009] The process steps can be carried out in parallel, sequentially or at least partially overlapping.

[0010] First, environmental data can be collected, including swarm data, lane data, object data, environmental information, etc. Based on the environmental data, a look-ahead area or a lane model can be created, which can include route sections with associated lane markings, lane boundaries and / or trajectory data that can be used for the lateral guidance of the vehicle along the route sections.

[0011] Furthermore, the vehicle speed can be checked.

[0012] Furthermore, a prediction can be made, e.g. whether the lane model can generate two lane boundaries that can be used for safe lateral guidance of the vehicle.

[0013] The prediction can include, for example: - Evaluation of live lane detection, - Assessment of plausibility and forecast of swarm data, and if necessary - Evaluation of object data, e.g. data about a vehicle ahead, which can be used for longitudinal guidance.

[0014] The evaluation of the track data can include, for example: - Characteristic points for lane markings or lane boundaries (e.g. look-ahead points at [0s, 0.2s, 0.4s, ...] * vehicle speed) can be observed over time.

[0015] Advantageously, it is possible to observe how these look-ahead points are detected over a certain period of time, whether the detection changes significantly, and / or whether the points are detected at all. If each observed point in a look-ahead region remains within a local boundary (e.g., if y_i < y_proximity) and is measured frequently enough, live lane detection can be considered stable.

[0016] Advantageously, a security value S_Live (e.g. in a value range [0;1]) can be introduced for the live detection, which can be higher the more the live lane detection is considered stable.

[0017] If the look-ahead points remain close (e.g., if y_i < y_proximity for i = 1, 2, ..., n; where n can describe the number of measurements) and sufficient measurement points are acquired within a time t_period depending on the x-position of the points (e.g., if n (x) ≥ n_threshold (x)), then the safety factor S_live can be set high (e.g., to a maximum value, e.g., to 1). For example, n_threshold (x) can decrease linearly with increasing distance (i.e., with increasing x). This allows for the fact that far points cannot / do not need to be measured as often as close points to be taken into account.

[0018] The assessment of the plausibility and foresight of swarm data can include, for example: - Swarm data (e.g. comprehensive: swarm track data and / or swarm trajectories) can be compared with the track data or with the live track detection.

[0019] Optionally, swarm attributes such as roundabouts, construction sites, intersections, road splits, road merging, etc. can be taken into account within a forecast so that the lateral guidance function is not activated in such situations and / or the swarm data for such routes is invalidated.

[0020] Advantageously, a confidence value S_Swarm (e.g., in a value range [0;1]) can be introduced to assess the plausibility and predictability of swarm data. High values can indicate that the plausibility and predictability of swarm data are very good. If the plausibility is valid for a certain period of time, e.g., within a few seconds, and the swarm data exhibits a long predictability (second and meter values), then the swarm data can be assessed as plausible, and the confidence value S_Swarm can be set high (e.g., to a maximum value, e.g., to 1).

[0021] The evaluation of object data can include, for example: - If a control object (e.g. an object, e.g. a vehicle, which can be used for longitudinal guidance) is located between the determined lane boundaries according to the created lane model (which can be based on live lane marking, swarm trajectories, swarm lane data), stable lane generation can be predicted.

[0022] Advantageously, a safety value S_Object (e.g., in a value range [0;1]) can be introduced, which can evaluate the prediction based on the object data. The safety value S_Object can be set high (e.g., to a maximum value, e.g., 1) if the control object is positioned in the lane and longitudinal guidance to the object is also performed.

[0023] Optionally, the prediction can include an evaluation of further environmental information, such as weather, time of day, road type, road condition, navigation data, traffic jam data, external circumstances, etc.

[0024] Depending on the (quality of) the prediction, the history or the distance / time elapsed when operating the vehicle can also be evaluated, for example: 1. Prediction factor is high: - Evaluation of the past distance / time with fast application (x_(distance, fast), t_(cycle, fast)), 2. Prediction factor is low and / or prediction is not possible: - Evaluation of the elapsed distance / time with slow application (x_(distance, slow), t_(cycle, slow)), 3. Independent of the prediction (from a speed v) or with a low dependence on the prediction: - Evaluation of the past distance / time with instantaneous application (x_(distance, instant), t_(cycle, instant); example application: both equal to 0),

[0025] The prediction function f Praed can, for example, depend on safety factors S Live , S Schwarm and S Objekt be determined: fPraed(SLive,SSwarm,SObject).

[0026] A conceivable prediction function f Praed can be determined as follows: fPraed=SLive+SSwarm+SObject3.

[0027] Another possible prediction function f Praed can be determined as follows: fPraed=1 if SLive= 1 or SSchwarm= 1 or SObjekt=1;otherwise fPraed=0.

[0028] The evaluation function f Vergangen can depend on the prediction function f Praed (S Live ,S Schwarm ,S Objekt ) and the vehicle speed v: fPast(fPraed(SLive,SSwarm,SObject),v).

[0029] A case distinction can be made, for example, as follows:

[0030] If f Praed (S Live ,S Schwarm ,S Objekt ) ≥ k schnell , then the application can apply: xAppl=xWeg,schnell,tAppl=tTakt,schnell

[0031] The application value x_(distance, fast) can be, for example, between 15 and 50 meters, in particular between 20 and 40 meters, preferably 25 meters.

[0032] The application value t_(clock, fast) can, for example, be between 0.1 and 4 seconds, in particular between 1 and 2 seconds, preferably 1.5 seconds.

[0033] If f Praed (S Live ,S Schwarm ,S Objekt ) < k schnell , then the application can apply: xAppl=xWeg,slow,tAppl=tTakt,slow

[0034] The application value x_(distance, slow) can be, for example, between 50 and 150 meters, in particular between 85 and 125 meters, preferably 100 meters.

[0035] The application value t_(cycle, slow) can be, for example, between 4 and 10 seconds, in particular between 5 and 8 seconds, preferably 5 seconds.

[0036] If v > v_fast: v>vquick, then the application can apply: xAppl=xWeg,instant(ZB=0),tAppl=tClock,instant(ZB=0),

[0037] The evaluation of the history or the past distance and / or time can be done via x_Appl, t_Appl.

[0038] A valid past (traveled) distance x_Past can be determined as the distance that the vehicle has consistently driven in the past with valid tracks from the track model.

[0039] A valid past time t_Past can be determined as the time that the vehicle has continuously driven (or stood still) with valid tracks from the track model in the past.

[0040] Activation of the lateral guidance function can advantageously be enabled from the environment perspective if xPast≥xAppl and tPast≥tAppl, and preferably if the lane boundaries can be derived from the lane model.

[0041] In this way, a process can be provided that, from an environmental perspective, can not only release and / or activate the lateral guidance function in a stable manner (if active, then active for a correspondingly long time), but also as quickly as possible. Based on a forecast range, e.g., using live environmental data, swarm data, and possibly object data, and a dependent evaluation of the traveled (i.e., elapsed) distance and / or time, it can be predicted whether or not it makes sense to release the lateral guidance functions. This optimizes the activation behavior of the lateral guidance functions, enables high availability, and ensures a low deactivation rate.

[0042] As already indicated above, the environmental data can fundamentally comprise various sensor data that can be sensed by the vehicle itself. The environmental data, in particular the sensor data, can advantageously comprise lane data sensed by the vehicle. The lane data can be recorded, for example, by a vehicle camera. In principle, however, it is also conceivable for the lane data to be recorded by suitable optical, electromagnetic, and / or acoustic sensors. The lane data can serve to supplement, at least partially replace, and / or verify the plausibility of the swarm data received from outside. The lane data can include lane markings or lane boundaries S.

[0043] As already indicated above, the environmental data can include various external data that can be obtained and / or received by the vehicle. The environmental data, in particular the external data, can advantageously include swarm data that can be provided, for example, by an external backend unit, by other road users, and / or infrastructure users. The swarm data can be used to supplement, at least partially replace, and / or verify the plausibility of the lane data recorded by the vehicle.

[0044] As already indicated above, the environmental data can include object data recorded by the vehicle. Advantageously, the object data can be checked (as part of the prediction) to determine whether an object, which is considered, for example, for a longitudinal guidance function of the vehicle, is moving within the lane boundaries according to the created lane model. The object data can thus enable an additional check to determine whether the lane model is plausible from the environmental perspective and can be safely used for lateral guidance of the vehicle.

[0045] As already indicated above, the environmental data can include various external environmental information, in particular navigation data, weather data, and / or traffic data, which can be provided, for example, by external services. This various external environmental information can also be used to enable additional verification of whether the lane model is plausible from an environmental perspective and can be safely used for lateral guidance of the vehicle.

[0046] Furthermore, it can be provided that when creating the prediction regarding the safety of the lane model, a prediction function is created that can depend on various safety factors of the environmental data. For example, it can be provided that the prediction function can take into account threshold values, e.g., hard threshold values, for the safety factors of the environmental data in order to provide a clear decision (yes or no) regarding the safety of the lane model for the lateral guidance function. In principle, however, it is also conceivable that the prediction function can be formed as a continuous function in order to provide a graded decision (e.g., in %) regarding the safety of the lane model for the lateral guidance function.

[0047] Furthermore, it can be provided that different safety factors of the environmental data are taken into account when creating the prediction. For example, the safety factors can be evaluated equally. However, it is also possible for the safety factors to be considered in a weighted manner.

[0048] For example, it is conceivable that the prediction confirms the safety of the lane model if at least one safety factor of the environmental data is likely to enable the safe provision of the lateral guidance function. Thus, it can be considered that the lane model can be assessed as safe if at least one source can enable the reliable detection of lane markings or lane boundaries, e.g., if the vehicle can sense the lane data itself and / or if the swarm data is plausible and predictive.

[0049] For a simple evaluation, it can be provided that the prediction, in particular the prediction function, provides at least one prediction factor: - a first prediction factor, for example high, e.g. a one, whereby in particular the evaluation of the history is carried out with first, preferably fast, application values, - a second prediction factor, e.g. low, e.g. zero, whereby in particular the evaluation of the history is carried out with second, preferably slow, application values.

[0050] Advantageously, the evaluation of a history above a threshold for vehicle speed can be performed independently of the prediction. This allows for the fact that above a threshold for vehicle speed, the vehicle is likely to be traveling on a highway, where reliable detection of lane markings or lane boundaries is highly likely.

[0051] Furthermore, it can be advantageous if the evaluation of a history above a threshold for vehicle speed can be performed using third, preferably instant, application values. For simplicity, instant application values can be set as follows: xDistance,instant=0,tClock,instant=0.

[0052] Advantageously, the evaluation of a history below a threshold for the vehicle speed can be carried out depending on the prediction. This allows the usability of the lateral guidance function to be expanded at low speeds.

[0053] On the one hand, the evaluation of the history can be carried out with first, preferably fast, application values if the prediction provides a first prediction factor, for example high, e.g. a one.

[0054] On the other hand, the evaluation of the history can be carried out with second, preferably slow, application values if the prediction provides a second prediction factor, for example low, e.g., zero.

[0055] In principle, it can be provided that the evaluation of the history during operation of the vehicle provides different application values depending on the prediction and / or the vehicle data: - first, e.g. fast, application values, if the prediction confirms the reliability of the lane model, especially if the vehicle speed is below a threshold value, - second, e.g. slow, application values, if the prediction does not confirm the safety of the lane model, in particular if the vehicle speed is below a threshold value, - third, preferably instant, application values, especially when the vehicle speed is above a threshold value.

[0056] A corresponding computer program product provides a further aspect of the invention, comprising instructions that, when executed by a computer, cause the computer to perform the method, which can be carried out as described above. The computer program product can achieve the same advantages as those described above in connection with the method according to the invention. These advantages are incorporated herein by reference.

[0057] A corresponding control unit provides a further aspect of the invention. A computer program in the form of code can be stored in a memory unit of the control unit. When the code is executed by a computing unit of the control unit, the program performs a method that can proceed as described above. The control unit can achieve the same advantages as described above in connection with the method according to the invention. These advantages are incorporated herein by reference in their entirety.

[0058] The control unit can be implemented at least partially in a control unit for the steering gear and / or at least partially in a central control unit of the vehicle.

[0059] Furthermore, the invention provides a vehicle having a control unit which can be designed as described above.

[0060] Further advantages, features, and details of the invention will become apparent from the following description, which describes embodiments of the invention in detail with reference to the drawings. The features mentioned in the claims and in the description may be advantageous individually or in any combination. They show schematically: Fig. 1 an example of the procedure, Fig. 2 an exemplary evaluation of a live lane detection, and Fig. 3 a possible implementation of the procedure.

[0061] The Fig. 1 - 3 serve to provide, in particular release and / or activate, a lateral guidance function Q of a vehicle F.

[0062] The method can be used in particular when the vehicle F is in operation or, in other words, when the vehicle F is moving.

[0063] As the Fig. 1, the procedure comprises the following steps: - Capturing (i.e. sensing, querying and / or receiving) environmental data UD, where the environmental data UD can include, for example, sensed track data SP and / or received swarm data SW, - Detecting (i.e. sensing, querying and / or receiving) vehicle data FD, preferably a vehicle speed v, - Creating a track model SM to provide the transverse function Q depending on the environmental data UD, wherein the lane model SM may comprise track sections with associated lane markings, lane boundaries S and / or trajectory data that can be used for the lateral guidance of the vehicle F along the track sections, - Creating a prediction P about the safety of the lane model SM when providing the lateral guidance function Q, where the prediction P can evaluate the safety of the lane model SM, e.g. using a prediction function f Praed which can, for example, simply affirm or deny the safety of the lane model SM or which can provide a specific value, for example between 1 and 100%, - Evaluation of a history x_Past, t_Past when operating the vehicle F depending on the prediction P and / or the vehicle data FD, in particular the vehicle speed v, e.g. using an evaluation function: fPast(fPraed(SLive,SSwarm,SObject),v), where the history x_Past, t_Past can include distances / times driven, which can be evaluated depending on the prediction P and / or vehicle speed v, whereby it can be predicted whether the vehicle F can be safely and stably steered laterally based on its history x_Past, t_Past, the current speed v and the prediction P, - Operating, in particular enabling and / or activating, the lateral guidance function Q depending on the evaluation, in particular depending on the evaluation function: fPast(fPraed(SLive,SSwarm,SObject),v).

[0064] As the Fig. 1 suggests, environmental data UD can first be recorded, including e.g. swarm data SW, track data SP, object data OD, environmental information Ul, etc.

[0065] Based on the environmental data UD, a look-ahead area or a lane model SM can be created, which can include route sections with associated lane markings, lane boundaries S and / or trajectory data that can be used for the lateral guidance of the vehicle F along the route sections.

[0066] Furthermore, the vehicle speed v can be checked.

[0067] Furthermore, a prediction P can be created, e.g. whether the lane model SM is valid and reliable, e.g. whether the lane model SM can generate two lane boundaries S that can be used for safe and stable lateral guidance of the vehicle F.

[0068] The prediction P can include, for example: - Evaluation of track data SP (see Fig. 2), which can be detected by live lane detection by the vehicle F itself, e.g. by a vehicle camera, - Assessment of plausibility and forecast of swarm data SW, and if necessary - Evaluation of object data OD, e.g. data about a vehicle driving ahead, which can be used for longitudinal guidance.

[0069] As the Fig. 2 suggests, the evaluation of the lane data SP (in particular the live lane detection by the vehicle F) can include, for example: - Characteristic points SP for lane markings (e.g. look-ahead points at [0s, 0.2s, 0.4s, ...] * vehicle speed) can be observed over time t.

[0070] Advantageously, it can be observed how these points SP are detected over a certain period of time t_period, whether the detection changes, and / or whether the points SP are detected at all. If each observed point SP remains within a look-ahead range within a local boundary (e.g., if y_i < y_proximity) and is measured frequently enough, live lane detection can be considered stable.

[0071] Advantageously, a safety value S_Live (e.g. in a value range [0;1]) can be introduced for the live detection, which can be higher the more the live lane detection by the vehicle F is considered stable.

[0072] If the points SP remain close (e.g. if y_i < y_proximity for i = 1, 2, ..., n; where n can describe the number of measurements) and within a certain period of time t_period depending on the x-position of the points SP sufficient measuring points n (x) are recorded (e.g. if n (x) ≥ n_threshold (x)), then the safety factor S_live can be set high (e.g. to a maximum value, e.g. to 1). For example, n_threshold(x) can decrease linearly with increasing distance (i.e. with increasing x). This allows it to be taken into account that far points SP do not need / must be measured as often as close points SP.

[0073] The assessment of the plausibility and foresight of swarm data SW can include, for example: - Swarm data SW (e.g. comprising: swarm track data and / or swarm trajectories) can be compared with the track data.

[0074] Optionally, swarm attributes such as roundabouts, construction sites, intersections, road splits, road junctions, etc. can be taken into account within a look-ahead area so that the lateral guidance function is not activated in such situations and / or the swarm data SW for such routes is marked as invalid.

[0075] Advantageously, a confidence value S_swarm (e.g., in a value range [0;1]) can be introduced to assess the plausibility and predictability of swarm data SW. High values can indicate that the plausibility and predictability of swarm data SW are good. If the plausibility is valid for a specific period t_period, e.g., within a few seconds, and the swarm data SW exhibits a comparatively long predictability (second and meter values), then the swarm data SW can be assessed as plausible, and the confidence value S_swarm can be set high (e.g., to a maximum value, e.g., to 1).

[0076] The evaluation of object data OD can include, for example: - If a control object (e.g. a vehicle that can be used for longitudinal guidance) is located between the generated lane data according to the created lane model SM (which can be based on live lane marking, swarm trajectories, swarm lane data), stable lane generation can be predicted.

[0077] Advantageously, a safety value S_Object (e.g., in a value range [0;1]) can be introduced, which can evaluate the prediction based on the object data OD. The safety value S_Object can be set high (e.g., to a maximum value, e.g., 1) if the control object is positioned in the lane and longitudinal guidance is also performed on the object.

[0078] Optionally, the prediction can include an evaluation of further environmental information, such as weather, time of day, road type, road conditions, navigation data, traffic jam data, external circumstances, etc.

[0079] Depending on the (quality of) the prediction, the history x_Past, t_Past or the past distance x_Past, time t_Past when operating the vehicle F can be evaluated, e.g. as follows: 1. Prediction factor P1 is high: - Evaluation of the past distance / time with fast application (x_(distance, fast), t_(cycle, fast)), 2. Prediction factor P2 is low and / or prediction is not possible: - Evaluation of the elapsed distance / time with slow application (x_(distance, slow), t_(cycle, slow)), 3. Independent of the prediction (from a speed v) or with a low dependence on the prediction: - Evaluation of the past distance / time with instantaneous application (x_(distance, instant), t_(cycle, instant); example application: both equal to 0),

[0080] As the Fig. 3 suggests, a prediction function f_Praed can be determined depending on different safety factors S_Life, S_Swarm, S_Object: fPraed(SLive,SSwarm,SObject).

[0081] A possible prediction function f_Praed can be determined as follows: fPraed=sLive+SSwarm+SObject3.

[0082] Another possible prediction function f_Praed can be determined as follows: fPraed=1 if SLive=1 or SSchwarm=1 or SObjekt=1;

[0083] As the Fig. 3 further suggests, the evaluation function f_Vergangen can be determined depending on the prediction function f_Praed and the vehicle speed v: fPast(fPraed(SLive,SSwarm,SObject),v).

[0084] A case distinction can be made, for example, as follows:

[0085] If: fPraed(SLive,SSwarm,SObject)≥kfast, then the application can apply: xAppl=xWeg,schnell,tAppl=tTakt,schnell

[0086] The application value x_(distance, fast) can be, for example, between 15 and 50 meters, in particular between 20 and 40 meters, preferably 25 meters.

[0087] The application value t_(clock, fast) can, for example, be between 0.1 and 4 seconds, in particular between 1 and 2 seconds, preferably 1.5 seconds.

[0088] If: fPraed(SLive,SSwarm,SObject)≥kfast, then the application can apply: xAppl=xWeg,slow,tAppl=tTakt,slow

[0089] The application value x_(distance, slow) can be, for example, between 50 and 150 meters, in particular between 85 and 125 meters, preferably 100 meters.

[0090] The application value t_(distance, slow) can be, for example, between 4 and 10 seconds, in particular between 5 and 8 seconds, preferably 5 seconds.

[0091] If v > v_fast: v>vquick, then the application can apply: xAppl=xWeg,instant(e.g.=0),tAppl=tClock,instant(e.g.=0),

[0092] The evaluation of the history x_Past, t_Past or the past distance x_Past, time t_Past can be done via x_Appl, t_Appl.

[0093] A valid past (traveled) distance x_Past can be determined as the distance that the vehicle F has consistently driven in the past with valid tracks from the track model.

[0094] A valid past time t_Past can be determined as the time that the vehicle F has continuously driven or stood still with valid tracks from the track model in the past.

[0095] Activation of the lateral guidance function Q can advantageously be enabled from the environment perspective if xPast≥xAppl and tPast≥tAppl, and preferably if the lane boundaries S can be derived from the lane model SM.

[0096] A corresponding computer program product, a corresponding control unit ECU and a corresponding vehicle F with a corresponding control unit ECU also represent aspects of the invention.

[0097] The control unit ECU can be provided as a central control unit or as a control unit for a steering system.

[0098] The above explanation of the embodiments describes the present invention exclusively by way of examples. Of course, individual features of the embodiments can be freely combined with one another, provided they are technically feasible, without departing from the scope of the present invention. List of reference symbols F vehicle Q Lateral guidance function UD environmental data SP track data SW swarm data OD object data UI environment information FD vehicle data v Vehicle speed v_fast threshold SM track model S Lane markings SP points for lane markings P Prediction ECU control unit x coordinate, vehicle longitudinal direction y coordinate, vehicle transverse direction f_Praed prediction function f_Past evaluation function S_Live security value S_Swarm security value S_Object security value

Claims

[1] Method for providing, in particular enabling and / or activating, a lateral guidance function (Q) of a vehicle (F), comprising: - Collection of environmental data (UD), - Recording of vehicle data (FD), in particular a vehicle speed (v), - Creating a lane model (SM) to provide the cross function (Q) depending on the environmental data (UD), - Creating a prediction (P) about the safety of the lane model (SM) when providing the lateral guidance function (Q), - Evaluating a history (x_Past, t_Past) when operating the vehicle (F) depending on the prediction (P) and / or the vehicle data (FD), in particular the vehicle speed (v), - Operating, in particular releasing and / or activating, the lateral guidance function (Q) depending on the evaluation. [2] Method according to claim 1, wherein the environmental data (UD) comprises sensor data sensed by the vehicle (F), and / or wherein the environmental data (UD), in particular the sensor data, comprise lane data (SP) sensed by the vehicle (F). [3] Method according to claim 1 or 2, wherein the environmental data (UD) comprises external data obtained and / or received by the vehicle (F), and / or wherein the environmental data (UD), in particular the external data, comprise swarm data (SW) provided by an external backend unit, by other road users and / or infrastructure users. [4] Method according to one of the preceding claims, wherein the environmental data (UD) comprise object data (OD) which are detected by the vehicle (F) and are preferably checked to determine whether an object which is taken into account, for example, for a longitudinal guidance function of the vehicle (F) is moving within lane boundaries (S) according to the created lane model (SM). [5] Method according to one of the preceding claims, wherein the environmental data (UD) comprise external environmental information (UI), in particular navigation data, weather data and / or traffic data, which are provided, for example, by external services. [6] Method according to one of the preceding claims, wherein when creating the prediction (P) about the safety of the lane model (SM), a prediction function (f_Praed) is created which depends on different safety factors (S_Life, S_Swarm, S_Object) of the environment data (UD), - in particular, wherein the prediction function (f_Praed) takes into account threshold values, e.g. hard threshold values, for the safety factors (S_Life, S_Swarm, S_Object) of the environmental data (UD) and / or wherein the prediction function (f_Praed) is formed as a continuous function, [7] Method according to one of the preceding claims, wherein different safety factors (S_Life, S_Swarm, S_Object) of the environment data (UD) are taken into account when creating the prediction (P), - in particular wherein the safety factors (S_Life, S_Swarm, S_Object) are taken into account equally and / or weighted, and / or wherein the prediction (P) confirms the safety of the lane model (SM) if at least one safety factor (S_Life, S_Swarm, S_Object) of the environmental data (UD) is expected to enable a safe provision of the lateral guidance function (Q). [8] Method according to one of the preceding claims, wherein the prediction (P), in particular the prediction function (f_Praed), provides at least one prediction factor (Pi): - a first prediction factor (P1), e.g. high, e.g. a one, whereby in particular the evaluation of the history (x_Past, t_Past) is carried out with first, preferably fast, application values (x_(Distance, fast), t_(Cycle, fast)), - a second prediction factor (P0), e.g. low, e.g. zero, wherein in particular the evaluation of the history (x_Past, t_Past) is carried out with second, preferably slow, application values (x_(Distance, slow), t_(Clock, slow)). [9] Method according to one of the preceding claims, wherein the evaluation of a history (x_Past, t_Past) above a threshold value (v_fast) for the vehicle speed (v) is carried out independently of the prediction (P), and / or wherein the evaluation of a history (x_past, t_past) above a threshold value (v_fast) for the vehicle speed (v) is carried out with third, preferably instant, application values (x_(distance, instant), t_(clock, instant)). [10] Method according to one of the preceding claims, wherein the evaluation of a history (x_Past, t_Past) below a threshold value (v_fast) for the vehicle speed (v) is carried out depending on the prediction (P), and / or wherein the evaluation of the history (x_Past, t_Past) is carried out with first, preferably fast, application values (x_(Distance, fast), t_(Clock, fast)) if the prediction (P) delivers a first prediction factor (P1), e.g. high, e.g. a one, and / or wherein the evaluation of the history (x_Past, t_Past) is carried out with second, preferably slow, application values (x_(Distance, slow), t_(Clock, slow)) if the prediction (P) delivers a second prediction factor (P0), e.g. low, e.g. a zero. [11] Method according to one of the preceding claims, wherein the evaluation of the history (x_past, t_past) during operation of the vehicle (F) provides different application values (x_apl, t_appl) depending on the prediction (P) and / or the vehicle data (FD): - first, e.g. fast, application values (x_(path, fast), t_(cycle, fast)), if the prediction (P) confirms the reliability of the lane model (SM), in particular if the vehicle speed (v) is below a threshold value (v_fast), - second, e.g. slow, application values (x_(path, slow), t_(cycle, slow)), if the prediction (P) does not confirm the safety of the lane model (SM), in particular if the vehicle speed (v) is below a threshold value (v_fast), - third, preferably instant, application values (x_(Weg, instant), t_(Takt, instant)), if in particular the vehicle speed (v) is above a threshold value (v_schnell). [12] A computer program product comprising instructions which, when the computer program product is executed by a computer, cause the computer to carry out a method according to any one of the preceding claims. [13] Control unit (ECU), comprising a computing unit and a memory unit in which a code is stored which, when at least partially executed by the computing unit, carries out a method according to one of the preceding claims. [14] Vehicle (F) comprising a control unit (ECU) according to the preceding claim.

Images