DE102022200076A1 - Adaptation of an automated motor vehicle to an environment - Google Patents

Abstract

The invention relates to a method for automated driving (240) of a motor vehicle (10) within an environment (80, 250), the motor vehicle (10) containing a control unit (12) with a driving model for automated driving using the driving model the following steps: providing (210) driving parameters (40) assigned to the environment (80, 250) for the driving model of the control unit (12), entering (110a, 220) the motor vehicle (10) into the environment (80, 250). prior or at the same time or subsequently adapting (230) the driving model on the basis of the driving parameters (40) and automated driving (240) of the motor vehicle within the environment (80, 250) using the adapted driving model.

Description

The present invention relates to the field of self-driving motor vehicles and relates to a method for the automated movement of a motor vehicle within an environment and a corresponding motor vehicle.

State of the art

For systems in motor vehicles that rely on sensing the environment, such as those present in self-driving motor vehicles, the publicly available specification ISO/PAS 21448 and the associated final draft ISO/FDIS 21448 provide guidance for determining the intended functionality or performance limitations that lead to potentially dangerous behavior of such systems. The absence of inadequate risks due to potentially dangerous behavior of a system is defined as "Safety Of The Intended Functionality" (SOTIF). The situational awareness of such systems is based on complex sensors and processing algorithms (including artificial intelligence and machine learning), which may not be able to correctly grasp all situations at all times. SOTI F puts special emphasis on the cognitive limitations under stressful conditions, i. H. under “trigger conditions” that can lead to false negative, false positive, or erroneous object properties (e.g. position error, wrong speed, etc.). A general goal is to significantly improve the safety concept or the robustness for such systems by recognizing situations in which triggering conditions and thus performance restrictions exist and mitigating them appropriately.

Methods for controlling vehicles within a specific environment are known from the prior art. So it is in the patent application DE 10 2019 208 744 A1 external control of a plurality of vehicles within a monitored environment is described. The patent application DE 11 2016 004 751 T5 discloses an autonomous driving support system that collects relevant information along a planned driving route. In the patent application DE 10 2016 102 065 A1 a method for supporting autonomous driving of a motor vehicle in a parking area is described, which uses an infrastructure map. All these systems deal with very specific problems, in none of these cases is a generally applicable method proposed that ensures the robustness and operational safety of a vehicle in specific environments.

Disclosure of Invention

According to the invention, a method for the automated driving of a motor vehicle within an environment according to claim 1 is proposed, wherein the motor vehicle contains a control unit with a driving model for automated driving using the driving model. Further advantageous configurations are specified in the dependent claims.

The method according to the invention comprises the following steps: providing driving parameters assigned to the environment for the driving model of the control unit, entering the motor vehicle into the environment with prior or simultaneous or subsequent adaptation of the driving model on the basis of the driving parameters, and automated driving of the motor vehicle within the Environment using the customized driving model.

The control unit is a system with a computing unit that has access to sensors and actuators (for example the engine control, the steering, the brake system) of the motor vehicle. The control unit can therefore evaluate sensor signals and, based thereon, control the connected actuators, for example in order to steer, accelerate or brake the motor vehicle. The driving model used by the control unit is an algorithm used by the control unit (e.g. implemented by software and/or an electrical circuit), which is used to process the sensor signals and control the actuators based thereon, possibly taking into account inputs from a human driver. responsible for.

According to the invention, it is therefore proposed to adapt a driving model for an automated motor vehicle for a specific environment on the basis of a set of parameters (driving parameters). Entry into the environment can be detected, for example, by external or on-board sensors such as light barriers, induction loops, cameras, RFID-based systems (RFID: radio-frequency identification, a technology for automatic and contactless identification and localization of objects) and/or based on the Data of a navigation system of the motor vehicle are determined. Entering an environment within the meaning of the invention means driving over an edge of the environment with the result that the motor vehicle is then located within the environment, with a motor vehicle being considered to be within an environment as soon as more than half of the vehicle (based on the length of the vehicle) is in the area. It is conceivable that the driving parameters for an environment are already Approaching the area will be provided. Such an approach can also be detected by sensors and/or established by means of a navigation system of the motor vehicle.

After the driving model has been adjusted, the motor vehicle drives automatically within the environment using the control unit whose driving model has been adjusted. Automated driving is to be understood here as driving with a degree of automation of at least level 1 according to the SAE J3016 standard. The invention is particularly advantageous in the case of higher degrees of automation, in particular level 4 (high automation) and level 5 (full automation), with levels 1, 2 and 3 also being conceivable. The control unit can therefore be set up to only support a human driver (level 1), but it can also be a control unit that drives the vehicle completely independently (level 5).

Adapting the driving model on the basis of driving parameters means setting the driving model taking into account information that is made available by the driving parameters. In the simplest case, specific parameters of the driving model are simply replaced by new values, namely the driving parameters.

In general, driving parameters can include, for example, basic parameters for driving (such as a maximum speed or acceleration) and/or general parameters for evaluating the sensor signals of the sensors of a motor vehicle (vehicle sensors), but also, for example, a weighting of the sensor signals of the motor vehicle relative to one another and/or in relation to the standard driving model (a general driving model that is standard, i.e. in particular outside of an environment in which special driving parameters are used). According to the invention, the evaluation and consideration of the sensor signals from sensors of a vehicle is therefore re-weighted for an environment. Such weightings do not have to relate to the electrical signals supplied directly by a sensor, but can also relate to sensor signals that are processed further electrically and/or by means of software. Different weightings of specific partial areas of a field of view (FOV) of one or more sensors of the motor vehicle can also be carried out by the driving parameters and the driving model can thus be optimized. If, for example, the field of view is subdivided into individual grid cells as such partial areas with the aid of a grid, different weightings of a sensor signal can be assigned to each grid cell. It is also conceivable that the driving parameters contain instructions to deactivate certain sensors and/or to ignore their signals, possibly also only in relation to parts of the field of vision of the motor vehicle (i.e. the field of vision in relation to the sum of all sensors in the motor vehicle). Such a deactivation can specifically also be achieved by weighting the corresponding sensor signal with the value zero. The driving parameters can also contain value ranges, for example a maximum and a minimum speed or, in the case of sensor signals, signal strengths within which the signals should be evaluated. Driving parameters can also take into account the specific properties of a sensor, i.e. in addition to the exact type, properties such as the exact model of the sensor, position and orientation.

A weighting of a sensor signal within the meaning of the invention means a weighting either in relation to other sensor signals from sensors in the motor vehicle (i.e. the weighting of the sensor signals relative to one another) and/or in relation to a certain predefined value, for example in relation to a standard value which, for example, is determined by the signal evaluation in the standard driving model of the motor vehicle could be predetermined. A weighting does not have to relate to the signals supplied by a sensor, but can also relate to further processed data. The driving model can also be used to weight certain partial areas of a field of view of one or more sensors of the motor vehicle and thus optimize a local driving model for an environment. A driving model can also take into account value ranges for relevant parameters or quantities, for example a maximum and a minimum speed or, in the case of sensor signals, signal strengths within which the signals should be evaluated.

Overall, the susceptibility of the motor vehicle to errors can also be reduced for special environments. The number of accidents is consequently reduced. Furthermore, the development time for the standard driving model can be reduced, since special cases (that is, environments that occur only rarely during the operating time of a motor vehicle) do not have to be taken into account.

However, the driving parameters do not have to directly encode technical setting instructions for the driving model, such as the weighting of sensor signals, but the actual, concrete settings of the driving model for certain environments can first be calculated on the basis of the driving parameters and/or from a local database such as a look-up table can be retrieved. This can be the case, for example, when the driving parameters made available provide general information about an environment, but this information still has to be converted into a specific configuration of the driving model that is optimized for the respective motor vehicle and the driving model. The driving parameters can, for example, provide information about the environment in a rather abstract, possibly standardized form; the translation into specific settings of the driving model can then only take place in a computing unit in the vehicle, for example in the control unit. For example, the specific sensor types installed in the vehicle could be taken into account. A mixture of specific technical setting instructions and abstract information are also conceivable as driving parameters for a driving model. Conversely, it is also possible that different driving parameters for different vehicle types, sensor types and the like are made available for an environment and then the correct driving parameters are selected on the basis of the specific motor vehicle.

The driving model can also be adapted taking into account the desired safety requirement level (for example according to ASIL, Automotive Safety Integrity Level). This safety requirement level can be part of the driving parameters, but can also be specified in some other way, for example by the driver. In general, in addition to the driving parameters, further information and data (for example from a navigation system of the motor vehicle) can be taken into account when adapting the driving model.

According to the invention, it was recognized that many motor vehicles are only in certain surroundings for a limited time. For example, a motor vehicle is only operated in a parking garage a fraction of the time. As a result, driving models, i.e. algorithms that are responsible for controlling automated vehicles, are typically not optimized for such special cases and in some cases are also not suitable. In order to be universally applicable, a driving model must be selected with regard to its parameters and properties in such a way that it can be used safely for any environment or at least as many different environments as possible. Environments and conditions for which a driving model was designed are referred to as the operational design domain (ODD) of the driving model. If a driving model is operated outside of its ODD, this can lead to errors and, in the worst case, to accidents.

An environment within the meaning of the invention is to be understood as a defined, arbitrarily geographically delimited partial area of the earth's surface. For example, this may be a country, city, street, intersection, parking lot, factory site (including or excluding on-site interiors of on-site buildings), the interiors of buildings such as parking garages, warehouses, and be production sites. A specific abstract geometric shape such as a circle, rectangle, or polygon can also define an environment. For example, a circle with a radius of 1 km around a defined point on the earth's surface could also define an environment.

Environments in which the invention can be used particularly advantageously are environments that lie outside the ODD of the standard driving model of an automated motor vehicle. For example, it could be the case that such a motor vehicle is in a parking garage for only 0.0001% of the operating time. It is to be expected that the standard driving model used to control the vehicle is therefore typically not optimized for the conditions that occur there, so driving in such an environment can be error-prone and use can lead to accidents.

An advantage of this invention is therefore that driving parameters can be provided for any, possibly very special, environments and the driving model of the control unit of a motor vehicle can be adapted based thereon. Different environments can have identical or different driving parameters. The driving parameters are advantageously optimized for the assigned environment. This optimization also reduces the susceptibility to errors in the standard driving model of the motor vehicle, since not all special cases have to be taken into account during its development. The driving parameters for the standard driving model can be optimized for typical scenarios and environments. The driving parameters of the model are then adjusted for special environments.

Special features of an environment that can be taken into account when designing driving parameters for this environment are, for example, the average or current traffic density, road width, lighting conditions, road surfaces, average or current weather conditions and, in particular, sensor-influencing and/or sensor-triggering properties of the environment such as reflecting Surfaces and/or potentially interfering light sources. For these points, reference is also made in particular to the publicly available specification ISO/PAS 21448 and specifically the sections 7.3.2 and 7.3.3 there or alternatively the corresponding final draft ISO/FDIS 21448.

The method according to the invention thus makes it possible to optimize a driving model of the control unit of a motor vehicle for special environments, so that it is significantly less error-prone than a driving model that is not adapted and instead tries to prepare for all conceivable situations. At the same time, not every special environment has to be taken into account during vehicle development, but dedicated driving parameters can be made available later if required, for example directly on site.

A driving model used within the scope of the invention can, for example, use driving parameters that are derived from a data set (e.g. in the form of a matrix) (and in a special case correspond to this) that contains information on the performance of sensors of a motor vehicle with regard to predetermined SOTIF-relevant attributes (such as the false-negative rate or the true-positive rate of a sensor) under certain circumstances such as weather conditions in the area and special boundary conditions of the area (such as traffic density, road width, road surface). The corresponding dependencies between the circumstances of an environment and SOTIF-relevant attributes expressed by conditional probabilities can be described for each sensor by a PGM (Probabilistic Graphical Model), preferably by a Bayesian network. The marginal probabilities regarding the SOTIF-relevant attributes of interest can then form a possible data set for the driving parameters. In this case, the dataset provides the driving model with information about the reliability of the sensors under certain conditions. Based on this, in particular depending on the degree of confidence derived from the data ("confidence level") of the sensor signals, the evaluation behavior of the driving model can be adapted, for example a weighting of the sensor signals to each other and/or a general information fusion ("sensor fusion") Sensor signals are changed.

The field of view (FOV) of the motor vehicle is preferably divided into grid cells. Each grid cell can be viewed separately for each sensor and assigned its own PGM. A distinction must be made here between the structure and the parameters (the conditional probabilities) of a Bayesian network: In many cases, the structure is the same for all grid cells of a field of view, although this does not have to be mandatory (for example, the structures for parts of the distinguish field of view). However, the parameters must be determined individually for all grid cells. With such a procedure, information is obtained for each grid cell with regard to the reliability of a sensor that is used for this area of the field of view. A data set that can be used for driving parameters for a driving model then results from the total number of data sets for the individual grid cells.

Environment-specific driving parameters (that is, those associated with an environment) can be provided, for example, during the production of the motor vehicle. For example, the parameters can be stored in a memory of a computing unit of the motor vehicle and called up when required. Alternatively, it is also conceivable that the parameters are only made available to the motor vehicle later, for example on a storage medium such as a memory card. A particularly advantageous variant is when the driving parameters are transmitted to the motor vehicle via a radio link, for example from an Internet server in a cloud or by means of edge computing.

The environment is advantageously a controlled environment, in particular a controlled street, a controlled area and/or a controlled building. A controlled environment (accordingly controlled road, area, building) is to be understood as an environment in which dynamic conditions (traffic and moving objects relevant to driving safety) are fully or at least partially monitored, for example by means of sensors and/or other methods. Examples include: cameras, photodetectors, radar, lidar (light detection and ranging), ultrasound-based sensors, piezoelectric sensors, air pressure sensors, temperature sensors, RFID-based systems (RFID: radio-frequency identification), induction loops, navigation systems for those driving in the area vehicles (including transmission of the data to a central processing unit) and satellite monitoring. Typically, in a controlled environment, the properties of the environment relevant to the safe driving of a motor vehicle are also monitored (e.g. thermometer to measure temperature) and, if necessary, actively controlled (e.g. heating to control temperature), and/or these properties are predetermined by the controlled environment itself (e.g. roofing to avoid rain and consequent wetness of a road surface). Examples of properties that are relevant for safe driving of a motor vehicle are physical parameters such as temperature (e.g. of the road or the air), humidity, wind speed and brightness, but also the presence of rain. Examples of controlled environments are parking garages and parking lots with automated parking systems (AVP, Automated Valet Parking) and dedicated roads (e.g. for autonomously driving shuttle buses) with special infrastructure to support automated driving etc. Another example of a controlled building is a warehouse in which the movement of autonomously driving motor vehicles is monitored.

According to a further advantageous embodiment of the method according to the invention, the driving model is preferably adapted on the basis of the driving parameters when the motor vehicle enters the environment, ie when driving over the edge of the environment. So if the motor vehicle moves into the environment, the driving model is adapted on the basis of the driving parameters when driving over the edge of the environment. It is also conceivable for this adaptation to take place before the edge of the environment is reached or only after the edge has been traversed, ie within the environment. In other words, in such a case, the driving parameters are adjusted when the motor vehicle falls below a predetermined spatial or temporal distance from the surroundings. The spatial distance can refer to the geographic distance (i.e. straight-line distance) from a point on the motor vehicle, for example the center of the motor vehicle, to the edge of the environment (i.e. to the nearest point at the edge of the environment), but alternatively it can also be the shortest route that can be driven by the motor vehicle to the edge of the area and/or the route suggested by a navigation system of the motor vehicle to the edge of the area and/or through the planned route to the edge of the area (if the motor vehicle is moving towards the area) or the distance traveled from the edge of the environment (if the motor vehicle is already in the environment). The same applies to the time interval: This can be calculated, for example, simply from the current speed of the motor vehicle and the geometric distance to the edge of the environment, or also using the driving time predicted for the motor vehicle by a navigation system (in the event that the motor vehicle is on the environment) or the time elapsed since crossing the edge (in the case that the motor vehicle is in the environment).

In contrast to the adaptation of the driving model when reaching or at least approaching the environment, the driving model adapted on the basis of the driving parameters can of course also be brought back into a different state when leaving the environment or moving away from the environment. The adjustments that have been made could be reversed, for example, or the driving model could be adjusted again on the basis of other driving parameters. The latter could be the case, for example, when the motor vehicle enters from one environment into another to which these other driving parameters are assigned. The same options are available for undoing or readjusting the adjustments as described above for approaching the environment (although the roles of environment and area outside the environment must of course be reversed).

In order to keep the time in which the optimal driving model for the environment is not used as short as possible, the driving model is preferably adapted on the basis of the driving parameters in a period of ≦1 minute, preferably ≦30 seconds, particularly preferably ≦10 seconds and completely particularly preferably ≦1 second before or after the motor vehicle enters the environment.

Furthermore, it can be advantageous that the motor vehicle does not actively move while the driving model is being adjusted, that is to say, for example, that the engine or engines of the motor vehicle are switched off. This can ensure that the adaptation of the driving model does not lead to unwanted actions by the control unit, such as sudden steering movements or acceleration while driving. Advantageously, such a stop of the motor vehicle is not carried out solely on the basis of a planned adaptation of the driving model, but stopping the motor vehicle can be used for another reason. For example, the vehicle could come to a halt near the edge of the environment due to a traffic light set to red. This forced stop could then be used to adapt the driving model to the environment.

It is also particularly advantageous if, when entering or before entering the environment, it is automatically checked whether there are driving parameters associated with the environment that are suitable for the driving model of the motor vehicle. Such a check can be carried out by accessing a corresponding memory in which driving parameters for surroundings are stored, for example a corresponding local database in a memory of the control unit and/or by accessing a database server to which the motor vehicle is connected via a mobile phone connection, and which, for example, may be part of infrastructure within or near the environment. In the vicinity of the surroundings means a distance (straight-line distance to the nearest point of the edge of the surroundings) of preferably ≦500 m, particularly preferably ≦100 m, very particularly preferably ≦10 m wise be done in good time so that the driving parameters can be loaded from a database server before entering the area. Such an automated check can ensure that the driver of the motor vehicle does not have to carry out such a check manually and that the driving model can be adjusted in good time using the driving parameters at a desired point in time until the surroundings are reached.

It is generally advantageous if the motor vehicle has a communication interface for wireless data reception and the driving parameters are provided wirelessly, for example by a database server accessible via the Internet, and the driving parameters are received wirelessly through the communication interface and then passed on to the control unit. It is thus possible for the motor vehicle to use the communication interface to establish a wireless connection, for example to a database server, and from there to download current versions of the driving parameters for a specific environment to a memory in the vehicle, for example the control unit. This can ensure that the vehicle has driving parameters that are suitable for the environment, even if these were not yet available at the time the vehicle was produced, for example.

A particularly advantageous variant of this is when the environment is assigned a data transmission unit (processing unit for transmitting data) and the wireless provision of the driving parameters is carried out by the data transmission unit, which is connected, for example, to a server on which the driving parameters are stored. Such a data transmission unit is preferably located in the vicinity or at least in the vicinity (preferably ≦500 m, particularly preferably ≦100 m, very particularly preferably ≦10 m away) from the surroundings. In such a case, data does not have to be transmitted via long-range mobile radio technologies ("Cellular V2X"), but can also be in the form of dedicated short-range communication (DSCR, Dedicated Short Range Communication) with WLAN-based technologies such as pWLAN (IEEE 802.11p) take place. It is conceivable, for example, that the environment is a multi-storey car park suitable for AVP and that relevant driving parameters are transmitted when entering the multi-storey car park using pWLAN or a similar standard. In general, an infrastructure that is connected to vehicles by means of vehicle-to-infrastructure communication (V2I communication, vehicle-to-infrastructure communication) is particularly advantageous for this invention, since this enables driving parameters that are currently relevant to the environment to be indicated which vehicles can be passed nearby. Such an infrastructure could also include sensors, such as cameras, that can identify vehicles so that the necessary driving parameters can be sent to them in a targeted manner. One advantage of this is that the vehicle itself or its computing units do not need to know about the existence or location and geometry of the environment; the vehicle is only informed when the driving model needs to be adjusted and is directly supplied with the relevant information (driving parameters).

It is particularly advantageous if the driving parameters used for the control unit include information for evaluating, in particular for weighting, a signal from a sensor of the motor vehicle. In this case, weighting means a weighting either in relation to other sensor signals or in relation to another value, for example in relation to a standard value. Such weighting does not have to relate to the signals supplied by a sensor, but can also relate to further processed data. The driving parameters can also be used to weight certain partial areas of a field of view of one or more sensors of the motor vehicle and thus optimize the driving model.

According to another aspect of the invention, a motor vehicle is proposed for use in a method according to one of the methods described above, the motor vehicle containing a control unit for automated driving of the motor vehicle using a driving model that is set up on the basis of information provided for an environment To be adapted to driving parameters for this environment. In this case, the control unit is a system with a computing unit that has access to sensors and actuators of the motor vehicle, wherein the control unit can contain a memory in which driving parameters can be stored. Such a motor vehicle advantageously contains a communication interface (a receiving module) for wireless data reception, which is connected to the control unit and which is set up to wirelessly receive the driving parameters and forward them to the control unit. There, the driving parameters can be stored, for example, in a memory of the control unit until they are used, or they can be used directly to adapt the driving model.

Advantages of the Invention

The invention offers many advantages. Thus, by providing environment-specific cal driving parameters for a driving model of an automated motor vehicle allows the driving model to be optimally adapted for special environments, in particular controlled environments, without these environments having to be taken into account during the original development of the vehicle. In the event that the environment provides a suitable infrastructure, the vehicle is also not dependent on its own sensors and/or data from its own navigation system, etc., in order to identify a specific type of environment at all and to suitably adapt the vehicle behavior . If the motor vehicle drives into a parking garage, for example, the vehicle could be identified, for example, by the infrastructure of the parking garage (e.g. by means of cameras) and then optimally suitable driving parameters could be transmitted directly to the motor vehicle via radio link, without the vehicle systems using the parking garage as a parking garage at all would have to identify.

Overall, the safety and robustness of a driving model for automated driving are significantly increased by the invention. These can also be increased in a targeted manner by specifying driving parameters for above-average safety-critical environments, which lead to the driving model acting particularly carefully here. Due to the evaluation of the sensor data adapted to the environment in an optimized driving model, the complexity of the necessary calculations and thus the time and resources required for their evaluation can also be reduced.

Furthermore, the invention also makes it possible to predict slower driving speeds on the route when driving through certain surroundings on the basis of the driving parameters to be used there (caused, for example, by sensor restrictions). Accordingly, the travel distance of the vehicle can be optimized taking these slower travel speeds into account.

character list

Embodiments of the invention are explained in more detail with reference to the drawings and the following description.

• 1 eine schematische Darstellung eines erfindungsgemäßen Verfahrens sowie eines erfindungsgemäßen Kraftfahrzeugs und
• 2 eine Darstellung eines erfindungsgemäßen Verfahrens als Flussdiagramm.

Show it:

• 1 a schematic representation of a method according to the invention and a motor vehicle according to the invention and
• 2 a representation of a method according to the invention as a flow chart.

Embodiments of the invention

In the following description of the embodiments of the invention, the same or similar elements are denoted by the same reference symbols, with a repeated description of these elements being dispensed with in individual cases. The figures represent the subject matter of the invention only schematically.

1 shows a schematic representation of an exemplary embodiment of the method according to the invention with a motor vehicle 10 according to the invention with a control unit 12 for the automated movement of motor vehicle 10 using a driving model, wherein motor vehicle 10 is set up on the basis of for an environment 80, for example a parking garage provided driving parameters 40 for this environment 80 to be adjusted. As shown in the figure, the vehicle approaches and eventually enters the environment 80 from a position 16a (arrow 110a). A standard driving model is used until a distance 82a from the environment 80 is reached. Such a distance 82a can also be zero, ie it can be identical to the edge of the area 80 .

In the figure, the settings of the driving model of motor vehicle 10 are illustrated schematically by a grid 20a, which represents the current weightings for a sensor of motor vehicle 10. The grid 20a corresponds to the field of vision of the motor vehicle 10, which was divided into cells 22a in the present case. Depending on the weighting of the sensor signal, the cells 22a are shown hatched differently in the figure, the cells 22a of the grid 20a in the figure only being provided with the reference symbol 22a in three cases purely by way of example. (The same applies to the grids 20b, 20c and their cells 22b, 22c.) If the vehicle falls below a specified spatial distance 82a from the surroundings 80, the driving model is adapted on the basis of driving parameters 40. The sensor signal weights are weighted differently across the field of view according to the effects of the environment 80 on the sensor. This is illustrated by the grid 20b with the cells 22b partially hatched differently in comparison to the grid 20a. Here, for example, individual sensor signals can be weighted more or less with regard to parts of the field of view. The driving parameters 40 can be made available, for example, by a data transmission unit 42 in the environment 80 or in the vicinity of the environment 80 which is part of an infrastructure 100 . This data can be transmitted to motor vehicle 10 (arrow 44) by radio-assisted (i.e. wireless) transmission, for example by means of WLAN Motor vehicle 10 includes a communication interface 14 for wireless data reception, which is set up to receive driving parameters 40 wirelessly and to transmit them to control unit 12 by means of a data connection 18 . Sensors of the infrastructure 100, for example cameras 50, can be used to detect motor vehicles 10 approaching the environment 80. If such a motor vehicle 10 is identified, driving parameters 40 are transmitted as described (arrow 44). The motor vehicle 10 is now driving within the environment 80 using the adapted driving model and thus also has the adapted driving model at the drawn-in position 16b.

A foreseeable or completed departure 110b of the surroundings 80 by the motor vehicle 10 can also be detected by the cameras 50, for example. The motor vehicle 10 is then instructed to switch to the previous driving model. In the figure, such a conversion takes place, for example, when a distance 82b from the environment 80 is exceeded. At position 16c, the motor vehicle 10 thus again has the driving model with the original weightings for the sensor, represented by the grid 20c with the cells 22c.

2 shows a method according to the invention in schematic form as a flow chart. Here, in step 210, driving parameters 40 for a driving model are provided to a control unit 12 of a motor vehicle 10, with driving parameters 40 being assigned to an environment 250. This provision 210 can take place, for example, by means of radio-assisted transmission, via which a connection is established to a server on which the driving parameters 40 are stored. In such a case, this server has a data transmission unit 42 for radio-assisted transmission of the driving parameters 40 to motor vehicles 10. Alternatively, it is also conceivable, for example, for the driving parameters 40 to be stored locally, ie in the motor vehicle 10, in a memory.

Outside the environment 250, the motor vehicle 10 is driven 200, for example, by means of a general driving model, the standard driving model. When the vehicle enters 220 the environment 250, the driving model is adapted 230 on the basis of the driving parameters 40 either beforehand, at the same time or afterwards. The motor vehicle 10 can now drive within the environment 250 with the adapted driving model (step 240). Before, at the same time as or after leaving 260 the environment 250, a renewed adaptation 270 of the driving model preferably takes place. After leaving 260 the environment 250, the original state of the driving model was thus restored; the standard driving model is used again when driving 280 .

The invention is not limited to the exemplary embodiments described here and the aspects highlighted therein. Rather, within the range specified by the claims, a large number of modifications are possible, which are within the scope of expert action.

QUOTES INCLUDED IN DESCRIPTION

This list of documents cited by the applicant was generated automatically and is included solely for the better information of the reader. The list is not part of the German patent or utility model application. The DPMA assumes no liability for any errors or omissions.

Patent Literature Cited

• DE 102019208744 A1 [0003]
• DE 112016004751 T5 [0003]
• DE 102016102065 A1 [0003]

Claims (11)

Method for automated driving (240) of a motor vehicle (10) within an environment (80, 250), wherein the motor vehicle (10) includes a control unit (12) with a driving model for automated driving using the driving model, comprising the following steps: a. Providing (210) driving parameters (40) assigned to the environment (80, 250) for the driving model of the control unit (12), b. Entry (110a, 220) of the motor vehicle (10) into the environment (80, 250) with prior or simultaneous or subsequent adaptation (230) of the driving model on the basis of the driving parameters (40) and c. automated driving (240) of the motor vehicle (10) within the environment (80, 250) using the adapted driving model. procedure after claim 1 , characterized in that the environment (80, 250) is a controlled environment (80, 250), in particular a controlled road, a controlled area and/or a controlled building. Method according to one of the preceding claims, characterized in that the adaptation (230) of the driving model on the basis of the driving parameters (40) takes place at: a. Entry (110a, 220) of the motor vehicle (10) into the environment (80, 250), or b. The motor vehicle (10) falling below a predetermined spatial or temporal distance (82a, 82b) to the environment (80, 250). Method according to one of the preceding claims, characterized in that the adaptation (230) of the driving model on the basis of the driving parameters (40) ≤ 1 minute, preferably ≤ 30 seconds, particularly preferably ≤ 10 seconds and very particularly preferably ≤ 1 second before or after entry (110a, 220) of the motor vehicle (10) into the environment (80, 250). Method according to one of the preceding claims, characterized in that the motor vehicle (10) does not actively move during the adjustment (230) of the driving model. Method according to one of the preceding claims, characterized in that when entering (110a, 220) or before entering (110a, 220) the environment (80, 250), an automated check is carried out to determine whether the driving parameters ( 40) exist that are suitable for the driving model of the motor vehicle (10). Method according to one of the preceding claims, characterized in that the motor vehicle (10) comprises a communication interface (14) for wireless data reception (70) and the provision (210) of the driving parameters (40) takes place wirelessly and the driving parameters (40) wirelessly by the Communication interface (14) received (70) and then passed on to the control unit (12). procedure after claim 7 , characterized in that the environment (80, 250) is assigned a data transmission unit (42) and the wireless provision (210) of the driving parameters (40) takes place through the data transmission unit (42), the data transmission unit (42) preferably being located in the environment (80, 250). Method according to one of the preceding claims, characterized in that the driving parameters (40) include information for evaluating, for example for weighting, a signal from a sensor of the motor vehicle (10). Motor vehicle (10) for use in a method according to any one of the preceding claims, characterized in that the motor vehicle (10) includes a control unit (12) for automated movement of the motor vehicle (10) using a driving model, which is set up on the basis of to be adapted to driving parameters (40) made available for an environment (80, 250) for this environment (80, 250). Motor vehicle (10) after claim 10 using a method according to claim 7 or 8th , characterized in that the motor vehicle (10) comprises a communication interface (14) for wireless data reception, which is connected to the control unit (12) and which is set up to receive the driving parameters (40) wirelessly and to the control unit (12) to pass on.

Images