WO2020053170A1

WO2020053170A1 - Method for generating a collection of information relating to driving scenarios for at least one vehicle and vehicle, arrangement and system composed thereof

Info

Publication number: WO2020053170A1
Application number: PCT/EP2019/074032
Authority: WO
Inventors: Peter Schlicht; Fabian HÜGER
Original assignee: Volkswagen Aktiengesellschaft
Priority date: 2018-09-10
Filing date: 2019-09-10
Publication date: 2020-03-19
Also published as: DE102018215351A1

Abstract

The invention relates to a method for generating a collection of information relating to driving scenarios for at least one vehicle (2), comprising: - obtaining vehicle sensor data for at least one vehicle (2); - generating driving scenario data on the basis of the vehicle sensor data, which driving scenario data describe present driving scenarios; - storing the driving scenario data as a component of a collection of information relating to driving scenarios. The invention furthermore relates to a vehicle (2) , to an arrangement (3) and to a system (1) composed thereof.

Description

description

Method for generating an information collection on driving scenarios of at least one vehicle, as well as vehicle, arrangement and system consisting thereof

The invention relates to a method for generating an information collection on or, in other words, about driving scenarios of at least one vehicle, as well as a vehicle, an arrangement and a system consisting of the vehicle and the arrangement.

To validate at least partially automated driving functions for vehicles and especially highly or fully automatic driving functions (i.e. driving functions of the so-called

Autonomy levels level 3 to level 5), the driving functions are usually checked using simulations or in real operation based on driving scenarios that a vehicle typically expects as expected. The driving scenarios have so far been derived from so-called catalogs, which are based on road building regulatory standards (rules for the construction of roads, the design of intersections and bridges, markings on construction sites etc.) and

Map data (for example curve radii and rises) are based. The catalogs are created based on a set of rules based on a "top down" approach. It is also assumed that the applicable traffic regulations such as a

Maximum speed should always be observed. Furthermore, contextual factors are considered independently of one another (road users, weather, season,

Lighting situation ...)

It has been shown that with this approach, not always one for every driving situation

meaningful validation of driving functions is possible.

The task is therefore to enable improved validation of at least partially automated and in particular highly automated driving functions from level 3.

This object is achieved by a method according to claim 1, a vehicle according to claims 8 and 9, an arrangement according to claims 10 and 11 and a system according to claim 12. A basic idea of the invention is likely from a vehicle

Driving scenarios run through are not only derived from predefined rules and presumptions, but from a collection of information that is actually based on

Driving scenarios based.

For this purpose, driving scenarios run through by at least one vehicle can be carried out using

Vehicle sensors are recorded and stored as part of a collection of information. The collection of information (e.g. in the form of a database or also one

Data backends) can then be evaluated, for example in order to determine a frequency of actually occurring driving scenarios and / or driving scenario parameters. Frequently occurring driving scenarios can then be used for a validation of driving functions (for example in the context of a simulation or by re-enacting these driving scenarios in the context of a real ferry operation). The driving functions can thus be validated realistically and based on actual driving scenarios that are frequently present, which increases the validity of the validation and generally the operational safety of the vehicle.

A method is proposed in detail with:

Obtaining vehicle sensor data of at least one vehicle, the

Map vehicle sensor data in particular (at least partially external) driving conditions while the vehicle is traveling;

- Generation of driving scenario data (for this vehicle) based on the

Vehicle sensor data, which describe existing driving scenarios or driving conditions of this vehicle; and

- Saving of the driving scenario data as part of an information collection

Driving scenarios.

Receiving vehicle sensor data and / or generating driving scenario data can be vehicle-specific or vehicle-specific. For example, this can be carried out by a vehicle-based (control) device that generates driving scenario data from preferably only the corresponding vehicle. At least with the

Saving the driving scenario data can also take into account driving scenario data of different vehicles, for example from a test vehicle fleet.

For example, the vehicle sensor data may include at least one of the following

Vehicle sensors are detected, a plurality of each sensor type also being present can: a light sensor, a rain sensor, a distance sensor (for example based on optical or radar-based radiation), a camera sensor, a

Geoposition sensor (GPS), an inclination sensor, an object detection sensor that performs object detection based on, for example, at least one camera sensor, image detection, a lidar sensor, a radar sensor, a

Ultrasonic sensor. Furthermore, other systems (in particular driver assistance systems and / or control or regulating systems) can also be used within the scope of the present disclosure

(for example ABS or ESP systems)) serve as vehicle sensors, for example if these signals refer to current driving situations or the road condition or also

Generate information on the (in) correct functionality of this function.

The vehicle sensor data and / or driving conditions depicted here can at least partially relate to conditions external to the vehicle, that is to say, for example, environmental conditions (weather) and static and / or dynamic traffic conditions (traffic infrastructure, road users). The driving conditions preferably do not affect

conditions specific to the vehicle type and / or conditions relating to the operating states of the vehicle, such as an engine temperature or general

Vehicle diagnostic parameters. This could reduce the informative value of the information collection or only for an application to the specific vehicle type

to qualify. Instead, driving scenarios can be understood primarily or at least partially as vehicle-independent environmental influences or driving conditions that arise due to the environmental influences, e.g. from one to be validated

Driving function must be mastered.

By mapping the driving conditions through the vehicle sensor data, e.g.

be understood that the vehicle sensor data either specify driving conditions in the sense of predetermined driving scenario parameters or contain information from which the driving conditions and / or driving scenario parameters can be determined.

In order to generate driving scenario data from the vehicle sensor data, these can be combined in accordance with predetermined regulations (for example program instructions) and / or in accordance with a predetermined data format. In particular, they can be compressed and / or converted into a desired data format. According to one variant, driving scenario data are predetermined from the vehicle sensor data

Coding regulations and in particular generated by means of a so-called hash function. For example, the driving scenario data can be in the form of several description codes and / or data records, each of which describes a single driving scenario, which is preferably composed of a plurality of driving scenario parameters. The driving scenario parameter values can be derived from the vehicle sensor data or can be contained therein.

In general, vehicle sensor data can be considered that are present and recorded within a common time period, at a common time or generally at the same time, and the driving scenario data can preferably also be determined for this time period or time. Under an existing driving scenario, a currently existing and / or related to a specific point in time or

driving scenario that is present during a specific acquisition period.

In general, the driving scenario data (or individual data records contained therein) can relate to a specific point in time or a period of time and can preferably be determined continuously, so that, for example, a type of data stream and / or an interval output of driving scenario data is generated. The driving scenario data can thus comprise several individual time-related data records or can be composed of these. Each data record can specify an individual driving scenario, which was determined on the basis of the vehicle sensor data, and optionally comprise a description code for this driving scenario.

As mentioned, a driving scenario can be defined on the basis of several driving scenario parameters or coded or described on the basis of these. Each driving scenario parameter can correspond to information of the type explained below and / or indicate or number it, wherein the information can be included in the vehicle sensor data. A single driving scenario mapped by the driving scenario data (or a single data record included therefrom) can contain values for a corresponding plurality of driving scenario parameters and / or be defined thereby. Such a compilation of values of individual driving scenario parameters for defining a driving scenario using the

Vehicle sensor data can correspond to a coding explained above or can emerge from this as a result.

In general, the vehicle sensor data and / or driving scenario data can be continuously recorded during a journey and stored, for example, at regular time intervals and / or at least when a change has been detected. It is possible that

Store driving scenario data and / or vehicle sensor data at least temporarily in a vehicle-related storage device and only afterwards, for example to transfer a vehicle-independent storage device. A transfer of the

Driving scenario data and / or vehicle sensor data can generally take place wirelessly, for example via mobile radio or a wireless Internet connection. It is preferred that the information is stored on a vehicle-independent storage device that is part of a computer system in a vehicle development center, for example.

It is preferred that the vehicle sensor data is evaluated by the vehicle itself in order to generate driving scenario data therefrom. In particular (but not exclusively) if the driving scenario data are encoded and / or compressed in accordance with predetermined regulations, the amount of data to be transmitted to a vehicle-independent device can thus be limited. This can also guarantee higher data protection in the sense of anonymization and / or deletion of personal data. Before the dates come on

Vehicle-independent devices are transferred, they can be processed in a suitable manner in the vehicle in order to take account of data protection regulations.

In principle, however, it is also possible to at least partially transfer the vehicle sensor data (e.g. without evaluations or driving scenario data generation taking place in the vehicle) to vehicle-independent devices. The latter can then generate at least parts of the driving scenario data from the received vehicle sensor data. This reduces the requirements e.g. to the computing power to be provided by the vehicle.

In summary, the driving scenario data generation can at least partially in the

Vehicle or external or independent of the vehicle.

In general, it can be provided to anonymize the driving scenario data and / or vehicle sensor data. This can be understood to mean the driving scenario data and / or

Editing, generating and / or transmitting vehicle sensor data in such a way that a clear conclusion about the specific vehicle, the vehicle owner or at least about a complete journey and / or a typical mode of operation of the vehicle is not possible or at least is made more difficult. This can happen, for example, in that driving scenario data and / or vehicle sensor data determined during a journey are not completely transmitted, in particular at the beginning and at the end of a journey. Additionally or alternatively, individual data records of the driving scenario data, which, for example, each indicate a recorded driving scenario, can also be transmitted contrary to the actually recorded chronological order and / or generally with a time delay. A further training provides that the information collection is evaluated and in particular is statistically evaluated. This can be done, for example, in a computer-assisted manner, preferably using a computing unit which also includes or is connected to the aforementioned storage device. The computing unit can be provided generally independently of the vehicle and in particular outside the vehicle. However, a vehicle-bound variant is also possible, in which case statically evaluated driving scenario data of the corresponding vehicle, for example to an external one

Information collection can be transferred.

The evaluation can generally comprise from the total of the present

Driving scenario data To determine which conclusions about typical, frequent and / or less frequent driving scenarios are possible. For example, the statistical evaluation can determine a frequency of driving scenarios and / or individual ones

Include driving scenario parameters. Also the determination of frequently correlating

Driving scenario parameters are possible.

In this context it can further be provided that (at least) an at least partially automated driving function (preferably from at least level 3 up to level 5) is designed based on the statistical evaluation. A partially automated driving function can be understood to mean that at least certain driving operations

maintained functions are performed automatically while others are still controlled by the driver. Laying out can be understood to mean, for example, control and / or regulating regulations or parameters for the driving function and / or

Determine activation conditions for this. If this is done on the basis of actually relevant or frequently occurring driving scenarios, as is generally made possible according to the invention, a reliable functioning of the driving function can be achieved.

The vehicle sensor data preferably comprise at least one of the following

Information and / or depict it:

- Information on a (static) transport infrastructure, especially one

Lane course, a number of lanes, the presence of an intersection and / or a predetermined type of intersection; a street type, recognized traffic signs or traffic lights;

- Information on driving conditions, in particular on the ambient light intensity (for example, can be determined using a light sensor), a precipitation intensity (for example, can be determined by Rain sensor), an ambient temperature, a road surface smoothness (for example, can be determined using an ESP assistance function);

- Information on dynamic objects, in particular on recognized vehicles or

Pedestrians, which, for example, using object detection algorithms e.g.

can be based on captured image or radar information;

- Information on the functioning of driver assistance systems (e.g. step-by-step in the sense of: works, does not work, works with restrictions, works with a low confidence value and / or a predetermined error code occurs, whereby each of the states mentioned can be assigned its own code value Activation or non-activation are recorded);

- Information about the occurrence of a predefined error scenario (in particular from

Corner cases, also called pathological cases, in which e.g. an algorithm can no longer make a reliable state determination or decision). In this case, the type of error (e.g. an algorithm outputs a specific error message) can be predefined as an error scenario, but not the cause of the error, which may is not known. If the error scenario occurs, the current data record can be the

Driving scenario data e.g. be marked appropriately to include therein

To investigate driving scenario parameter values with regard to or as a possible cause (s) of error.

The type of road, the type of intersection, traffic signs and / or traffic lights can be recognized, for example, based on image evaluation or geolocation algorithms which search image information for previously classified objects (in this case

predefined traffic signs, traffic lights, types of intersections or street types). The course of the lane and the number of lanes can be determined, for example, based on image information of the vehicle surroundings captured by a camera.

According to a further embodiment, the vehicle sensor data include and / or

Driving scenario data furthermore at least one of the following: location information (for example related to the location at which the vehicle sensor data was recorded and / or the

Driving scenario was present), in particular a location-based satellite-based (for example by GPS); a time information (for example based on the point in time at which the vehicle sensor data was recorded and / or the driving scenario was available), which the

Specifies the time of acquisition of the data (ie vehicle sensor data and / or driving scenario data). This can increase the information content of the vehicle sensor data and also the driving scenario data derived therefrom and / or for a plausibility check of the determined Driving scenarios are used. For example, the driving scenario data can be verified or checked for plausibility on the basis of driving scenarios that appear likely, the latter being able to be determined on the basis of map information and / or weather service information.

In order to achieve the anonymization explained above, the time information and / or the location information can preferably also be deleted again before the driving scenario data are stored as part of the information collection. In particular, this information can be used for the plausibility check explained above and, e.g. if this is positive, then deleted from the driving scenario data.

According to a further development, it is provided that the driving scenario data and / or

Vehicle sensor data are at least partially transmitted to a vehicle-independent (i.e. vehicle-external) storage device, on which the information collection is preferably also stored. As described, this can be done anonymously and / or wirelessly.

Alternative objects of the invention are described below (vehicle,

Arrangement and system). Features mentioned in this context can be the same for any of the above and below variants

Process features should be formed. In particular, these objects can generally be set up to carry out a method according to any of the above and below variants.

The invention further relates to a vehicle, in particular an automobile, for example a passenger car, with:

a control device which is set up to record vehicle sensor data generated by vehicle sensors of the vehicle and to generate driving scenario data therefrom which describe (preferably currently) driving scenarios; and

a transmission device which is set up to transmit the driving scenario data generated by the control device to a vehicle-independent storage device. So it is one in the vehicle

Driving scenario data generation.

The invention also relates to a vehicle having a control device which is set up to record vehicle sensor data generated by vehicle sensors of the vehicle; and a transmission device which is set up to detect those detected by the control device To transmit vehicle sensor data to a vehicle-independent driving scenario data generating device. It is therefore a matter of generating the driving scenario data from the vehicle.

The transmission device can be set up for wireless data transmission, for example via mobile radio or a wireless Internet connection. However, transmission by storage medium (for example a USB stick) or by means of a readout or diagnostic device is also possible, which can in each case be coupled to the transmission device.

The control device can be designed as a control device and, for example, control a process sequence of the type explained above. For this purpose, it can provide corresponding data acquisition and / or generation functions or the data required for this

Program instructions and optionally also include a (temporary) storage device. In particular, the control device can be set up for any of the driving scenario data generation and / or coding described above or below.

In order to acquire the vehicle sensor data, the control device can access a vehicle bus or another connection which carries the vehicle sensor data.

The invention further relates to an arrangement, in particular a stationary and / or vehicle-external or vehicle-independent arrangement, with:

a receiving device that is configured to receive driving scenario data generated by at least one vehicle; a storage device, which is set up to contain an information collection received by the receiving device

Save driving scenario data; and an evaluation device (for example in the form or comprising an arithmetic unit or processor unit) which is set up to:

To evaluate the collection of information, in particular to evaluate it statistically. The arrangement can be provided, for example, in the form of or as part of a computer system, in particular comprising at least one PC and / or at least one server.

Alternatively, the arrangement can include a receiving device for receiving

Include vehicle sensor data and a driving scenario data generating device which is set up to generate driving scenario data, which describe the present driving scenarios, from transmitted vehicle sensor data. This corresponds to driving scenario data generated externally by the vehicle. Finally, the invention also relates to a system comprising at least one vehicle and an arrangement according to the preceding aspects, wherein these are combined accordingly to carry out vehicle-scenario-related or vehicle-external driving scenario data generation. The system can be set up, for example, to transmit driving scenario data generated by the vehicle to the arrangement, there as part of the

Store information collection and finally statistically evaluate the latter, in particular in such a way that frequently recurring driving scenarios and / or

Parameter correlations can be determined.

An embodiment of the invention will now be described with reference to the accompanying

schematic figures explained. They represent:

Fig. 1 is a schematic diagram of a system according to an embodiment of the

Invention comprising an arrangement according to the invention and a vehicle according to the invention, the system executing a method according to the invention; and

Fig. 2 is a flowchart of the method carried out by the system.

A system 1 according to the invention is shown in FIG. 1 that a vehicle 2 and a

Arrangement 3 according to an embodiment of the invention comprises. The arrangement 3 is provided independently of the vehicle (i.e. as a component not physically connected to it and / or moved together with it) and, for example, as a stationary component

Computer system realized.

The system 1 enables vehicle-bound driving scenario data generation with information-gathering storage and evaluation external to the vehicle. Alternative variants which relate to driving scenario data generation external to the vehicle have been generally described above and are not explained separately below.

The vehicle 2 comprises a plurality of vehicle sensors 10, one of which is marked on a vehicle front merely by way of example. The vehicle 2 preferably comprises at least one vehicle sensor 10 on each side (i.e. on the front and rear as well as on the two sides comprising the entrance doors), which sensor detects the surroundings and

Environmental information determined. From this, for example, an existing one Traffic infrastructure and / or dynamic objects located in the vicinity are closed (for example, to other road users). examples for

Vehicle sensors 10 and information which can be ascertained herewith have been generally discussed above and are partly explained in more detail below.

The vehicle 2 comprises a control device 12 which is connected to a vehicle bus, not shown separately. All vehicle sensors 10 directly or indirectly feed the information recorded by you into the vehicle bus (for example in the form of a digital signal or as vehicle sensor data). In other words, the vehicle sensors 10 write the acquired information onto the common vehicle bus. The control device 12 can read the corresponding ones by reading out the vehicle bus

Acquire vehicle sensor data and, if necessary, buffer them on a storage device (not shown separately).

The control device 12, which is an example of a driving scenario data generation device 13, comprises a processor unit with which the vehicle sensor data

Driving scenario data are generated. For this purpose, the vehicle sensor data are converted in accordance with a predetermined coding or data generation specification

summarized. In the example shown, this is done in accordance with the coding explained below, whereby only selected information explained below can be taken into account and the individual information can generally be combined with one another in any way (i.e. the driving scenario data can comprise any combination of individual or all of this information). Examples of code blocks associated with a single record (i.e., a single

Driving scenario) of the driving scenario data can be put together (for example according to the following order).

The following applies to the exemplary embodiment:

- Information on the static (ie immobile) traffic infrastructure is determined and encoded from the vehicle sensor data. For example, image processing uses a number of lanes (e.g. 1 - 10), an indication of the lane on which the vehicle is located (so-called ego lane), a road type from a pre-stored classification of different road types (e.g. types 1 to 6) and the existence of an intersection (yes / no, specified as 1/0) is determined. A resulting code block of the driving scenario data could be, for example, 4 2 2 1, the Individual values indicate the number of lanes (four), the ego lane (2 = second lane, e.g. from the right), the type of road (type 2), and the presence of an intersection (1 = yes).

Determined location information (for example GPS information) is taken into account. This can be done by specifying a block of code with at least two coordinate values (for example, a latitude and a longitude value).

Information on dynamic objects (i.e. objects moving actively relative to the vehicle) is taken into account, which is done, for example, based on image or radar-based object detection. For this purpose, a recognized vehicle type, a distance to this vehicle type, information on which lane this vehicle type is located, and in the same way a number and position of pedestrians can be summarized as a code block.

The code block 1 10 3 15 1 0 3 indicates, for example, that there is a type 1 vehicle (in this case a car) in a left lane (first position in the code block) and that it is located 10 m away from your own position (see second position (10) in the code block). It is also stated that there is a type 3 vehicle (in this case a motorcycle) in a right lane (third position in the code block) and that it is located 15 m away from your own position (see fourth position (15) in the code block) ). The distance considerations can e.g. in each case in or against the direction of travel. Finally, it is also stated that there is a single pedestrian to the left of the lane (fifth position in the code block) (see value 1) and that to the right of the lane (seventh position in

Block) three pedestrians are located (see value 3). Due to the value 0, there are no pedestrians on the lane itself (sixth position in the code block). As far as can be seen from the vehicle sensor data, the direction of travel of the other road users could also be specified.

- Information about driving conditions or, in other words,

Context information captured. This affects ambient light (via light sensor),

Precipitation (via rain sensor), temperature and smoothness (via ESP driver assistance function). In addition, the time of entry is also taken into account. The code block 60 5 15 10.5 0 would then stand, for example, for a 60 percent light intensity (based on a maximum measurable light intensity), a precipitation of strength 5 (based on a previous precipitation classification, for example from 0 to 10), an outside temperature of 15 °, a time of 10.30 a.m. (given here as 10.5) and a roadway specified as not smooth (based on a previous one

Smoothness classification, for example from 0 to 10).

Further information or data mentioned in the general part could also be determined and stored as a code block, e.g. whether driver assistance systems are activated and / or functioning correctly or whether predefined error scenarios (in particular corner cases) occur.

The individual data record obtained according to the example above would then be in the form of a

Description codes for the current driving scenario are as follows: 4 2 2 1

Latitude value Longitude value 1 10 3 15 1 0 3 60 5 15 10.5 0. Each of the values represents a single driving scenario parameter value and the total value obtained or the parameter value sequence can also be referred to as a hash value.

The driving scenario data is thus composed of individual (driving scenario) data records, which in turn are each formed by a sequence of driving scenario parameter values and relate to a single recorded driving scenario. The

Driving scenario parameter values indicate the information contained in or derived from the vehicle sensor data, in particular with regard to existing driving conditions.

Such a data record can optionally also be compressed before a transmission to the arrangement 3 explained below in order to reduce the amount of data to be transferred. Furthermore, such data records can be continuously generated and transmitted by reading out or recording the vehicle sensor data after the expiry of predetermined time intervals. The data records transmitted as a whole then form the driving scenario data which are sent to the

Transfer arrangement 3 and be stored there as part of a (driving scenario) information collection. The information collection generally collects the determined

Driving scenarios and serves as a kind of knowledge database.

Vehicle 2 includes one for transmitting the driving scenario data

Transmission device 14, which transmits the data to a receiving device 16 of the arrangement 3, for example by mobile radio and thus wirelessly (see dashed line in FIG. 1). This can take place immediately after receiving a single data record of the type explained above. The arrangement 3 comprises a storage device 18, on which the received data records are stored as part of a driving scenario data information collection. An evaluation device 20, comprising at least one processor (and / or one microcontroller), also accesses the memory device 18.

Before or after the transfer, the data records can be checked for plausibility and / or anonymized. For this purpose, the control device 12 (before the transmission) or the

Evaluation device 20 (after the transfer) at least selected

Compare driving scenario parameter values with information obtained from maps or weather services. This can take place in particular taking into account the time or location information contained in the driving scenario data. However, for the purpose of anonymization, this time and / or location information can then be deleted again,

for example before the corresponding data record is saved in the information collection.

A further possibility of anonymization consists in not determining or transmitting driving scenario data for a predetermined period of time at the beginning or at the end of a journey and / or retrospectively after the end of the journey, the driving scenario data for such

Clear periods.

After the end of a journey of the vehicle 2, there are therefore preferably a plurality of individual data records in the storage device 18 relating to running scenarios. It is also possible for a plurality of vehicles 2 to transmit data records to the arrangement 3 and for the information collection stored there to be a plurality of

Combines driving scenario data of different vehicles 2. The

Information collection can be continuously supplemented over a longer period (for example, several weeks or months). Furthermore, the information collection can be statistically evaluated by the evaluation device 20, in particular by frequently

to determine recurring driving scenarios or generally a driving scenario frequency distribution. Additionally or alternatively, frequency distributions and / or frequent correlations of the individual driving scenario parameter values can also be determined. The result of this evaluation can also be stored in the memory device 18.

Thus, the storage device 18 contains information relating to actually relevant driving scenarios that are generated by the vehicle 2 (or a possible plurality thereof, for example one Test car fleet) were run through. This can be used to validate and design automated driving functions based on these driving scenarios.

FIG. 2 shows a flow diagram of the method explained above. In a step S1, vehicle sensor data are acquired with the vehicle sensors 10. In a step S2, these are transmitted to the control device 12 or detected by the latter by reading out the vehicle bus. In a step S3, an individual driving scenario data record is determined on the basis of the vehicle sensor data. This takes place after a predetermined time interval and continuously while driving. If a corresponding data record has been determined, it is transferred to the arrangement 3 in a step S4 and stored there as part of the information collection in the storage device 18.

In an optional step S5 it is checked whether a predetermined evaluation condition is fulfilled (for example the expiration of a predetermined time interval, the presence of a

Minimum number of driving scenario records or the existence of a predetermined one

User input). If this is not the case, the process goes back to step S1 in order to determine further data records. If the evaluation condition is met, on the other hand, a statistical evaluation of the information collection is carried out in step S6 by means of the evaluation device 20.

Reference list

system

vehicle

arrangement

Vehicle sensor

Control device

Driving scenario data generation device

Transmission device

Receiving device

Storage device

Evaluation device

Claims

1. Method for generating an information collection on driving scenarios at least one

Vehicle (2), with:

- obtaining vehicle sensor data from at least one vehicle (2);

- Generation of driving scenario data on the basis of the vehicle sensor data which are present

Describe driving scenarios; and

- Saving of the driving scenario data as part of an information collection

Driving scenarios.

2. The method of claim 1, further comprising:

- Statistical evaluation of the information collection.

3. The method according to claim 2,

the statistical evaluation comprising at least one of the following:

- determining a frequency of individual driving scenarios;

Determining a frequency of individual driving scenario parameters;

- Determination of correlations of individual driving scenario parameters.

4. The method of claim 2 or 3, further comprising:

- Designing an at least partially automated driving function based on the

statistical evaluation.

5. The method according to any one of the preceding claims,

wherein the vehicle sensor data comprise or depict at least one of the following information:

- Information on a traffic infrastructure, in particular on a lane course, a number of lanes, the presence of an intersection and / or a predetermined type of intersection, a type of road, recognized traffic signs or traffic lights;

- Information on driving conditions, in particular the ambient light intensity, one

Precipitation intensity, an ambient temperature, a road surface smoothness; - Information on dynamic objects, in particular on recognized vehicles or

Pedestrians;

- Information on the functioning of driver assistance systems;

- Information on the occurrence of a predefined error scenario.

6. The method according to any one of the preceding claims,

wherein the vehicle sensor data and / or driving scenario data further comprise at least one of the following:

location information, in particular location information determined by satellite;

- Time information, which indicates a time of acquisition of the data.

7. The method according to any one of the preceding claims,

wherein the driving scenario data and / or vehicle sensor data are at least partially transmitted to a vehicle-independent storage device (18).

8. Vehicle (2) with

- A control device (12) which is set up by vehicle sensors (10) of the

Vehicle (2) generated vehicle sensor data and from it

To generate driving scenario data which describe existing driving scenarios; and

- A transmission device (14) which is set up by the

Control device (12) to transmit generated driving scenario data to a vehicle-independent storage device (18).

9. Vehicle (2) with

- A control device (12) which is set up by vehicle sensors (10) of the

Detect vehicle sensor data generated by the vehicle (2); and

- A transmission device (14) which is set up by the

Control device (12) recorded vehicle sensor data to a vehicle-independent driving scenario data generation device (13).

10. Arrangement (3), with:

- A receiving device (16) which is set up by at least one

Receive vehicle (2) generated driving scenario data;

- a storage device (18) which is set up to store an information collection comprising driving scenario data received by the receiving device (16); and - An evaluation device (20) which is set up to evaluate the information collection.

1 1. arrangement (3), with:

- A receiving device (16) which is set up by at least one

Receive vehicle (2) generated vehicle sensor data;

- A driving scenario data generating device (13), which is set up from the

Vehicle sensor data to generate driving scenario data, which present

Describe driving scenarios;

- a storage device (18) which is set up to store an information collection comprising driving scenario data received by the receiving device (16); and

- An evaluation device (20) which is set up to evaluate the information collection.

12. system (1),

comprising at least one vehicle (2) according to claim 8 and an arrangement (3) according to claim 10 or at least one vehicle (2) according to claim 9 and an arrangement (3) according to claim 11.