Methods for improving the lateral and / or longitudinal guidance of a vehicle

Abstract

Method for determining a path for a vehicle (200), comprising the steps: - Retrieving (S100) configuration parameters, wherein the configuration parameters define a lateral steering, lateral guidance and / or longitudinal guidance behavior for the vehicle (200), - Determining (S102) a planned path, namely a planned lane guidance for the vehicle, based on the configuration parameters, wherein the configuration parameters are lane guidance parameters of the vehicle, wherein outputs from driver assistance systems of the motor vehicle are evaluated for determining (S102) and wherein the configuration parameters define influences of the outputs of the driver assistance systems, - Detecting (S104) a deviation of the vehicle (200) from the planned path, - Saving (S106) path parameters that represent a path changed by the deviation, - Evaluating (S108) the path parameters, - Adjusting (S110) the configuration parameters based on the evaluation, - Retrieving the adapted configuration parameters by the vehicle or by another vehicle from a remote computer system (250) and - The application of the retrieved configuration parameters by the vehicle or the other vehicle for its own lane planning, namely for planning its own lateral steering, lateral guidance and / or longitudinal guidance behavior (200).

Description

[0001] The invention relates to a method and a system for improving the lateral and / or longitudinal guidance of a vehicle.

[0002] It is known that information about the transverse steering is available in a vehicle, collected primarily by various vehicle sensors. Furthermore, it is known that a certain degree of preventive behavior, i.e., proactive planning of lateral dynamics, is possible based on this information. For example, planned vehicle behavior regarding lateral guidance can be predicted within seconds by driver assistance systems.

[0003] However, especially in the case of at least partially autonomous vehicles, depending on the actual road conditions or the state of the elements defining the lane (e.g., lane markings), a vehicle user may need to intervene in the driving process (e.g., due to an unexpected obstacle, missing / incorrect lane markings, etc.). This occurs particularly when a behavior determined or expected by the vehicle needs to be corrected or changed due to actual conditions.

[0004] If a user intervenes in the driving process of an at least partially autonomous vehicle in order to correct a driving behavior planned by assistance systems, this may be due to the fact that the driver assistance systems' assumption of the planned driving behavior is incomplete or that a condition important for planning the driving behavior was not correctly recognized.

[0005] US 2017 / 0 008 521 A1 concerns a technique for calibrating a vehicle speed.

[0006] US 2015 / 0 012 165 A1 proposes a technique for automatically avoiding irregularities in a road.

[0007] DE 10 2013 013 867 A1 concerns a vehicle that can drive using a driver assistance system, but with the possibility of driver intervention, or can drive completely autonomously in another operating mode.

[0008] The object of the invention is to improve the situation described above and to improve the safety of at least partially autonomous vehicles.

[0009] The invention provides a method and a system for this purpose according to the independent claims. Further developments of the invention are the subject of the dependent claims.

[0010] In a first aspect, a method for determining a path for a vehicle is provided, comprising the steps of: - Retrieving configuration parameters, whereby the configuration parameters define a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle; - Determining a planned path, namely a planned lane guidance for the vehicle, on the basis of the configuration parameters, wherein the configuration parameters are lane guidance parameters of the vehicle, for determining outputs of driver assistance systems of the motor vehicle are evaluated, wherein the configuration parameters define influences of the outputs of the driver assistance systems; - Detecting a deviation of the vehicle from the planned path; - Saving path parameters that reflect a path changed by the deviation, - Evaluating the path parameters, - Adjusting the configuration parameters based on the evaluation; - Retrieving the adjusted configuration parameters by the vehicle or by another vehicle from a remote computer system and - Use of the retrieved configuration parameters by the vehicle or the other vehicle for its own lane planning, namely for planning the vehicle's own transverse steering, lateral guidance and / or longitudinal guidance behaviour.

[0011] The configuration parameters define a lateral and / or longitudinal guidance behavior / lateral behavior for the vehicle and in particular an influence, preferably a weighting, of outputs from driver assistance systems that are evaluated to determine the planned path.

[0012] Adjustment of the configuration parameters can be performed when a number of path parameters are stored, in particular when the number reaches, exceeds or falls below a threshold.

[0013] The path parameters of a variety of vehicles can be stored and evaluated. The path parameters can reflect actual transverse steering, lateral guidance, and / or longitudinal guidance behavior, or actual lateral and / or longitudinal guidance.

[0014] The path parameters can be used to store situation information, preferably a geographical position, weather information, a time of day, and / or the position of the sun. The configuration parameters can be adjusted in association with the stored situation information.

[0015] The configuration parameters can be retrieved from a remote computer system. The path parameters can be sent to the computer system. The path parameters can be saved by the computer system. The computer system can evaluate the travel parameters. The computer system can adjust the configuration parameters.

[0016] The planned path can at least partially define a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle in a certain temporal and / or spatial range, and in particular a trajectory for the vehicle.

[0017] The path parameters can be deleted, and in particular an adaptation based thereon can be reversed, if a predetermined time has passed since the acquisition, and in particular if no further path parameters are stored in the predetermined time and / or for the situation information.

[0018] The configuration parameters can be vehicle guidance parameters. The planned path can be a planned guidance path for the vehicle.

[0019] A remote computer system may adjust at least one configuration parameter when a number of changed paths for the situation information is received, in particular when the number reaches or exceeds a predetermined threshold.

[0020] Based on the retrieved configuration parameters and / or the path parameters, the vehicle can learn a behavior for a situation that can be determined by the situation information and / or the configuration parameters, in particular by using machine learning methods.

[0021] In a second aspect, a system for determining a path for a vehicle is provided, the system comprising: - a first processing unit of at least one vehicle, which is configured to determine a planned path, namely a planned lane guidance for the vehicle, based on configuration parameters retrieved from a memory, wherein the configuration parameters are lane guidance parameters and define a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle, wherein outputs of driver assistance systems of the vehicle are evaluated for the determination, and wherein the configuration parameters define influences of the outputs of the driver assistance systems, - a detection unit of the vehicle, which is configured to detect a deviation of the vehicle from the planned path and to store path parameters in the memory, which reflect a path changed by the deviation, and - a second processing unit of a remote computer system, which is configured to evaluate the path parameters and to adapt the configuration parameters on the basis of the evaluation, wherein the system is further configured to - to retrieve the adjusted configuration parameters from the remote computer system by the vehicle or by another vehicle and - to use the retrieved configuration parameters by the vehicle or the other vehicle for its own lane planning, namely for planning the vehicle's own transverse steering, lateral guidance and / or longitudinal guidance behaviour.

[0022] The system may comprise at least one vehicle and a remote computer system, wherein the vehicle may be communicatively connected to the remote computer system through a wireless communication network, and the memory and / or the second processing unit may be provided in the remote computer system.

[0023] The first processing unit, preferably in the at least one vehicle, can retrieve the configuration parameters from the remote computer system.

[0024] The detection device can send the configuration parameters to the remote computer system.

[0025] The at least one vehicle can retrieve the configuration parameters from the remote computer system. The computer system can adapt the configuration parameters based at least on path parameters received from the at least one vehicle by the computer system, preferably based on path parameters received from a plurality of vehicles.

[0026] The remote computer system can determine situation information and store it in memory in association with the path parameters and / or configuration parameters. The remote computer system can output configuration parameters to the at least one vehicle depending on the stored and acquired situation information.

[0027] The invention will now be described with reference to the figures. They show: Fig. 1 a method according to the invention, and Fig. 2 a schematic view of the system according to the invention.

[0028] The aim of the invention is to register and evaluate interventions by the user (or a driver or passenger) and to use this evaluation to improve the transverse steering, lateral guidance and / or longitudinal guidance behavior of a vehicle.

[0029] The vehicle can detect user intervention, in particular steering intervention and / or braking application. Detection can occur when an intervention in a vehicle operation occurs that lies outside a tolerance range. For example, the vehicle sensors, which can also include cameras, yaw rate sensors, tilt sensors, acceleration sensors, radar sensors, acceleration sensors for determining negative or positive acceleration, speed sensors, and other sensors for monitoring the vehicle's surroundings and / or vehicle behavior, can detect when the vehicle deviates from a planned path. In particular, the deviation of an actual value for the planned path from a target value for the planned path for transverse steering of the vehicle is determined. If the deviation is above a threshold value, the deviation is classified as user intervention.More deviations can be detected (e.g., simultaneously) and / or more threshold values ​​can be defined, with which a comparison with the deviation(s) is made. Position data from a positioning unit can also be recorded and used to determine a deviation. Acceleration and / or speed can be recorded, for example, to detect a deviation from a desired longitudinal guidance behavior. Using recorded accelerations, curves can be detected more effectively, for example. Deviations from (target) acceleration values ​​when negotiating curves can also be determined in this way.

[0030] For example, a point on a lane can be predetermined that the vehicle must reach within a defined period of time. If the point is not reached within the defined period of time, or if an actually reached point on the lane / roadway deviates from the predetermined point, a check is performed using at least one threshold value to determine whether the deviation is within a tolerance range. If this is not the case, it is assumed that the user has intervened.

[0031] The tolerance range can be defined, for example, by a specific angular range and / or time range. In particular, vehicle odometry can be used to determine a deviation between an actually selected path and the planned path. Map data can be used to plan the planned path or route, which can also be incorporated into the predictive planning of the planned path. In particular, a vehicle's navigation unit can plan a path based on at least one partial route from a start to a destination. The planned path can thus, in particular, be part of a planned route.

[0032] If a deviation is detected, information about it can be sent from the vehicle, for example, via a communications unit, to a remote computer system (backend). In the remote computer system, deviation information from different vehicles can then be evaluated and / or accumulated over time. In particular, parameters that determine the actual route (tuples of different values) can be transmitted to the remote computer system. The tuples thus define the actual path taken by the vehicle, i.e., the path corrected by the driver.

[0033] In the remote computer system, a large number of tuples are collected over time and across different vehicles for a deviation at a location (e.g., deviations in driving behavior within a defined geographical area). The tuples are then filtered using various filter mechanisms to take into account, for example, time, temperature, geographical position, and weather dependencies of the collected tuples.

[0034] An analysis is performed to identify similar tuples and thus determine the path taken by the most vehicles or a selection of closely spaced paths actually chosen. Each tuple specifies the reason for choosing the planned path.

[0035] In particular, lane detection, lane formation, tracking of another vehicle, e.g. when following a vehicle, or map information can influence the planning of the planned path or determine the planning of the path. These influences can be encoded in a tuple. For example, individual values ​​of a tuple can represent weightings of the influences. In particular, the parameters output by driver assistance systems are encoded in a tuple. Thus, a value in a tuple can represent a weighting that has the influence of a driver assistance system on the planning of the path. Just by analyzing the tuples that describe the planned path, it can then be determined which value, i.e. which sensor size, which driver assistance system and / or which type of lane detection had the greatest influence on the planning of the planned path.

[0036] Furthermore, the path actually chosen by the driver can be transmitted to the remote computer system. The actually chosen path can in turn be encoded in at least one tuple or a sequence of tuples. Tuples associated with the planned path can then be compared with tuples associated with the actually chosen path.

[0037] If vehicle users increasingly intervene at a point along a route, the remote system can conclude that a value in the tuples that are essential for planning the planned path at this point along the route has an excessive or dominant influence on the predetermined path. The remote computer system can thus recognize that the influence of this parameter on the planning of the planned path needs to be reduced or its weighting for determining the planned path needs to be adjusted. For example, to improve longitudinal guidance, a speed can be adjusted, particularly to keep acceleration values ​​within defined ranges, which are defined by threshold values.

[0038] In particular, the vehicle plans the planned path after accessing information from the remote computer system. The remote computer system can, in particular, be a cluster, cloud, and / or server system. The information from the remote computer system is accessed primarily when the vehicle retrieves current information about a planned route. In particular, information based on tuples already transmitted from other vehicles to the remote computer system can also be retrieved here. Accordingly, for the planning of the planned path, a weighting for the influence of the driver assistance systems or sensors, at least in a subsection of the planned path, can be adjusted based on the information retrieved from the remote computer system.

[0039] The following example illustrates this: An at least partially autonomous vehicle is on a planned route to a destination. The planned route is determined by the vehicle, for example, by accessing map data. Various driver assistance systems determine a lane for the vehicle. The sensors of the driver assistance systems scan the surface of a roadway in front of the vehicle in the direction of travel and, for example, by detecting lane markings, identify a lane in which the vehicle should stay.

[0040] To plan the planned path, the vehicle also accesses information from the remote computer system and determines the planned path based on this information. Once the planned path or trajectory has been determined, the vehicle also sends information concerning the planned path to the remote computer system. The information concerning the planned path reflects, in particular, the course of the planned path.

[0041] While the vehicle can also consider real-time traffic information when planning the planned path, unknown events can still occur that make following the planned path impossible or necessitate a change to the path. For example, the planned path may be blocked, for example, by construction work, a broken-down vehicle, or an object that has entered the path, for example, due to a storm. To continue driving at such a location, user interaction with the vehicle is necessary. This can then prevent an accident or the vehicle from stopping.

[0042] The vehicle now sends odometric data to the remote computer system. The remote computer system can now detect a deviation between the actually selected path and the planned path, at least based on the information concerning the planned path and / or odometric data.

[0043] Based on the odometric data, which in particular describe a deviation in the lateral steering, lateral and / or longitudinal guidance, the remote computer system can then adjust the data used to plan the planned path and, in particular, adjust the weighting of the parameters. This occurs in particular when a corresponding deviation in the lateral steering, lateral and / or longitudinal guidance occurs frequently in a geographical area of ​​a route, and in particular when a number of deviations are above or below at least one predetermined threshold value, or reach at least one threshold value.

[0044] Depending on the detected deviation of the actually selected path from the predetermined path, at least one parameter for planning the planned path can be changed.

[0045] For this purpose, relevant information can be sent back to the vehicle. This information can also be contained in map data, or map data can be enriched with the relevant information, such as how the parameter weighting for planning the predetermined path should be selected at specific locations on a map. In this way, the information important for path planning, or a change in the parameter weighting, can be transmitted to the vehicle either proactively via the backend or the remote computer system and / or by updating the map data. The vehicle can adjust the parameters at the appropriate location according to the weighting and select the planned path accordingly.

[0046] Another parameter that can be evaluated could be the vehicle's inclination or yaw rate. Pre-processing of the parameters can be performed in the vehicle. In this respect, only information about a deviation and the parameters used to specify this deviation or the predetermined path can be sent to the remote computer system. Here, the weighting of the parameters used to determine the predetermined path is important. Sensor inputs can include LIDAR, radar, and / or optical sensors. Overall, all vehicle odometry data can, at least theoretically, be evaluated as parameters, and information can be sent to the remote computer system based on these parameters.

[0047] Quality values ​​can also be provided for the individual parameters, so that an evaluation can be carried out in advance and especially in the vehicle as to how good the input parameters were for determining the predetermined path.

[0048] This means that an action performed by a user in one vehicle does not have to be performed by other users of other vehicles. In particular, the data showing a user's intervention and thus a changed path is accumulated and evaluated and, in particular via the configuration parameters, made available to other vehicles. This means that the other vehicles or the vehicle that retrieves the adjusted configuration parameters from the remote computer system and uses them for lane planning or for planning the transverse steering behavior, lateral guidance and / or longitudinal guidance can demonstrate adjusted and, in particular, improved driving behavior based on the evaluated path parameters. User intervention to cope with the respective situation is then no longer necessary. In this respect, reactions to specific situations can be learned through the evaluation of the path parameters.

[0049] The remote computer system can use machine learning algorithms for this purpose. However, it is also possible for learning to take place within the vehicle. In particular, the method according to the invention can be carried out entirely within a vehicle. The learning can be implemented, in particular, using neural networks.

[0050] The idea of ​​the invention is therefore to improve lane detection or a fusion of influences on lateral and / or longitudinal guidance using data collected while a vehicle is driving, for example in a remote computer system (backend). Areas of application for the invention are partially automated, highly automated, and fully automated driving functions that perform or influence lateral and / or longitudinal guidance of a vehicle. The goal is to improve the functions of lateral and / or longitudinal guidance by registering and aggregating incorrect behavior and retroactively reducing or correcting incorrect behavior through evaluation. For this purpose, incorrect behavior and / or the vehicle's environment are to be learned. Learning can also take place online or offline in the vehicle; however, aggregation and evaluation preferably take place offline in the remote computer system.The vehicle and the computer system can each have a processing unit for evaluating data and performing machine learning processes. The data aggregated in the backend can then be used to provide improvements on the recognition function side (lane detection), fusion side (road model), or customer function side.

[0051] Road crests and dips can be learned and better managed. Driven curves can be learned; in the event of a failure or error, a default behavior can be learned. Retrospective determination of corrective information is possible: Automated curves can be retrospectively compared with user interventions (driver function). This allows error points and corrections to be learned.

[0052] For example, if the lane-finding algorithm mistakenly sets its course on a tarmac joint, the vehicle may continue driving straight ahead even though it should be cornering. The driver notices the error and intervenes in the steering. The error is detected from the planned path or trajectory and the actual path or trajectory. The remote computer system can then evaluate whether errors have occurred frequently at specific locations / geographical positions. Accordingly, an adjusted set of configuration parameters can be transmitted to the vehicle to prevent the error from occurring in the future.

[0053] It is also possible to detect a lane marking type (overridable / non-overridable) and transmit the corresponding information to the remote computer system in order to adjust the configuration parameters evaluated for the vehicle's transverse steering, lateral guidance, and / or longitudinal guidance behavior accordingly. This can, in particular, improve the detection of lane marking types.

[0054] By transferring path parameters, it is also possible to learn entrances and exits (especially on motorways) in order to complete or correct map material, for example.

[0055] Aggregation in the remote computer system is advantageous for detecting the frequency of changed paths. For example, it is possible to distinguish between location-independent and location-dependent events, or to detect correlations with situational information, such as weather data, time of day, position of the sun, etc.

[0056] Fig. 1 illustrates the flow of the method described herein in a diagram. In step S100, configuration parameters are retrieved. The retrieval occurs in particular by a vehicle and by the remote computer system. For this purpose, a communication unit of the vehicle is preferably used, which transmits a corresponding query, for example based on a route defined by a navigation unit, via a communication network. The query can be initiated by a processing unit of the vehicle that is connected to the navigation unit, is part of the navigation unit, or is identical to the navigation unit. The configuration parameters determine parameters that determine the transverse steering, lateral guidance, and / or longitudinal guidance behavior of the vehicle.

[0057] In step S102, the planned path is determined based on the configuration parameters. This can be done by the vehicle's processing unit. In step S104, a deviation from the planned path is detected. This can occur, in particular, if the vehicle is already following the planned path.

[0058] In a step S106, path parameters representing a path modified by the detected deviations are stored. The path parameters can be stored in a memory of the vehicle and / or in a memory of the remote computer system. The vehicle and / or the computer system can be equipped with a corresponding storage unit. To store the path parameters in the computer system, the vehicle's processing unit can transmit the path parameters to the computer system using the communication unit. This is preferably done wirelessly via the communication network.

[0059] In step S108, the path parameters are evaluated. This evaluation can be performed by the vehicle and / or the remote computer system. A statistical, heuristic, historical, selective, and / or predictive evaluation can be performed. In particular, the evaluation is performed by a second processing unit of the remote computer system based on cumulative data from a plurality of vehicles.

[0060] In step S110, the configuration parameters that will be retrieved from a vehicle in the future are adjusted based on the evaluation of the path parameters. In particular, the weighting of the parameters that influence or determine the transverse steering, lateral guidance, and / or longitudinal guidance behavior of a vehicle is changed.

[0061] Fig.Figure 2 illustrates a system 20 as described herein. Schematically shown is a vehicle 200 with a first processing unit 201. The first processing unit 201 serves, in particular, to retrieve the configuration parameters from a memory that may be provided in the vehicle 200 or in a remote computer system 250. Preferably, the configuration parameters are retrieved using a communication unit 202 through wireless communication between the vehicle 200 and the remote computer system 250. Accordingly, the remote computer system 250 may have a corresponding communication unit 251.

[0062] The first processing unit 201 then determines a planned path for the vehicle 200 based on the configuration parameters.

[0063] A detection unit 203 is used to detect a deviation, in particular in the transverse steering, lateral guidance, and / or longitudinal guidance behavior of the vehicle, from a behavior determined by the configuration parameters and the planned path. In particular, the detection unit 203 checks whether parameters determined by planning the planned path lie within a tolerance range relative to the actually determined parameters. This can be done, in particular, based on defined threshold values ​​for a deviation. The detection unit 203 further determines the path parameters that reflect the actual or changed path. The detection unit 203 can also send the path parameters to the remote computer system 250.

[0064] A second processing unit 252 can be provided in the remote computer system 250. However, the second processing unit 252 can also be identical to the processing unit 201. The second processing unit 252 then evaluates the path parameters that represent the changed path and, based on the evaluation, determines how the configuration parameters are to be changed or adjusted. Accordingly, the second processing unit 252 can also change the configuration data stored in the memory. As already mentioned, the configuration data preferably specifies a weighting of settings for the transverse steering, lateral guidance, and / or longitudinal guidance behavior of a vehicle.

[0065] It should be understood that the system may include more than one vehicle 200. The at least one vehicle 200 may retrieve the configuration parameters from the remote computer system 250. The configuration parameters may be adjusted based at least on path parameters, but also based on other data received from the at least one vehicle.

[0066] The computer system 250 can also determine situation information and store it in the memory in association with the path parameters and / or configuration parameters. The computer system 250 can output the situation information and the configuration parameters to the at least one vehicle depending on the stored and detected situation information.

Claims

[1] A method for determining a path for a vehicle (200), comprising the steps of: - retrieving (S100) configuration parameters, wherein the configuration parameters define a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle (200), - Determining (S102) a planned path, namely a planned lane guidance for the vehicle, based on the configuration parameters, wherein the configuration parameters are lane guidance parameters of the vehicle, wherein for determining (S102) outputs of driver assistance systems of the motor vehicle are evaluated and wherein the configuration parameters define influences of the outputs of the driver assistance systems, - detecting (S104) a deviation of the vehicle (200) from the planned path, - storing (S106) path parameters that reflect a path changed by the deviation, - Evaluate (S108) the path parameters, - Adjusting (S110) the configuration parameters based on the evaluation, - retrieving the adjusted configuration parameters by the vehicle or by another vehicle from a remote computer system (250) and - use of the retrieved configuration parameters by the vehicle or the other vehicle for its own lane planning, namely for planning the vehicle's own transverse steering, lateral guidance and / or longitudinal guidance behavior (200). [2] The method of claim 1, wherein the influence comprises a weighting of the outputs. [3] Method according to claim 1 or 2, wherein the adaptation (S110) of the configuration parameters takes place when a number of path parameters is stored, in particular when the number reaches or exceeds a threshold value. [4] Method according to one of the preceding claims, wherein path parameters of a plurality of vehicles (200) are stored and evaluated. [5] Method according to one of the preceding claims, wherein the path parameters represent an actual transverse steering, lateral guidance and / or longitudinal guidance behavior. [6] Method according to one of the preceding claims, wherein situation information, preferably a geographical position, weather information, a time of day, and / or sun position information, is stored with the path parameters and the adaptation of the configuration parameters takes place in association therewith. [7] Method according to one of the preceding claims, wherein the retrieval (S100) of the configuration parameters is carried out by a remote computer system, the path parameters are sent to the computer system, the path parameters are stored by the computer system, the evaluation (S108) is carried out by the computer system, and / or the adaptation is carried out by the computer system. [8] Method according to one of the preceding claims, wherein the planned path at least partially defines a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle (200) in a specific temporal and / or spatial range, and in particular a trajectory for the vehicle (200). [9] Method according to one of the preceding claims, wherein the configuration parameters are deleted, and in particular an adaptation (S110) based thereon is reversed if a predetermined time has passed since the detection, and in particular if no further configuration parameters are stored in the predetermined time and / or for the situation information. [10] Method according to one of the preceding claims, wherein a remote computer system adapts at least one configuration parameter when a number of changed paths for the situation information is received, in particular when the number reaches a predetermined threshold. [11] Method according to one of the preceding claims, wherein the vehicle (200) learns a behavior for a situation determined by the situation information and / or the configuration parameters on the basis of the retrieved configuration parameters and / or the path parameters, in particular by using machine learning methods. [12] A system (20) for determining a path for a vehicle (200), the system (20) comprising: - a first processing unit (201) of at least one vehicle (200), which is configured to determine a planned path, namely a planned lane guidance for the vehicle, on the basis of configuration parameters retrieved from a memory, wherein the configuration parameters are lane guidance parameters and define a transverse steering, lateral guidance and / or longitudinal guidance behavior for the vehicle (200), wherein outputs from driver assistance systems of the vehicle are evaluated for the determination (S102) and wherein the configuration parameters define influences of the outputs of the driver assistance systems; - a detection unit (203) of the vehicle (200) which is configured to detect a deviation of the vehicle (200) from the planned path and to store path parameters in the memory which reflect a path changed by the deviation, and - a second processing unit (252) of a remote computer system (250) which is configured to evaluate the path parameters and to adapt the configuration parameters on the basis of the evaluation, wherein the system (20) is further configured to - to retrieve the adjusted configuration parameters from the remote computer system (250) by the vehicle or by another vehicle, and - to use the retrieved configuration parameters by the vehicle or the other vehicle for its own lane planning, namely for planning the own transverse steering, lateral guidance and / or longitudinal guidance behavior of this vehicle (200). [13] The system (20) of claim 12, wherein the system comprises at least the vehicle (200) and a remote computer system (250), wherein the vehicle (200) is communicatively connected to the remote computer system (250) through a wireless communication network, and the memory and / or the second processing unit (252) is provided in the remote computer system (250). [14] System according to claim 12 or 13, wherein the first processing unit (201), preferably in the at least one vehicle (200), is arranged to retrieve the configuration parameters from the remote computer system (251). [15] System according to one of claims 12 to 14, wherein the detection device (203) sends the configuration parameters to the remote computer system (250). [16] System according to one of claims 12 to 15, wherein the at least one vehicle (200) is configured to retrieve the configuration parameters from the remote computer system (250) and wherein the computer system (250) is configured to adapt the configuration parameters at least based on path parameters received from the at least one vehicle (200) by the computer system (250). [17] System according to one of claims 12 to 16, wherein the remote computer system (250) is configured to determine situation information and to store it in association with the path parameters and / or configuration parameters in the memory and to output configuration parameters to the at least one vehicle (200) depending on a stored and a detected situation information.

Images