DE102019217346B4 - Method for displaying information on a human-machine interface of a motor vehicle, computer program product, human-machine interface and motor vehicle - Google Patents

Abstract

Method for controlling a human-machine interface (4) of a motor vehicle (2), wherein the human-machine interface (4) has at least one display device (8, 10, 12) for displaying information (34, 36, 38, 40 , 42, 44), wherein at least one driving situation parameter is determined, which characterizes a driving situation of the motor vehicle (2), and wherein, depending on the determined driving situation parameter, a load parameter is determined by means of an algorithm (20), which represents a cognitive load of a driver (6 ) of the motor vehicle (2), so that an information density of the at least one display device (8, 10, 12) of information to be displayed by means of the display device (8, 10, 12) is adjusted depending on the load parameter, the algorithm (20) generating a data set with driving situation parameters, wherein each driving situation parameter is assigned a load parameter, the load parameters being based on at least one measured variable that characterizes the behavior of the driver, characterized in that the algorithm (20) is trained by a machine learning method by Test drives are carried out with a plurality of test drivers, with characteristic parameters being recorded and stored during the test drives to describe a driving situation, the test drivers receiving input requests to an input element of a human-machine interface during the test drives, an input duration for carrying out the input, an input error rate and/or a number of operating steps are recorded and correlated with the characteristic parameters by the algorithm (20).

Description

The present invention describes a method for displaying information on a human-machine interface of a motor vehicle, a computer program product, human-machine interfaces and a motor vehicle.

Methods for displaying information on a human-machine interface of a motor vehicle, computer program products, human-machine interfaces and motor vehicles of the type mentioned at the beginning are known in the prior art.

The amount of information a driver can grasp saliency and quickly depends on the cognitive load a driver is exposed to during a current driving situation. The cognitive load reflects how much mental strain the driver is required to fully understand a situation, especially a driving situation, and be able to react correctly. The more information a driver has to absorb, the higher his cognitive load. The more complex a driving situation is, the more the driver has to concentrate. On a simple route with little traffic, less attention is required than on a busy, winding route that is driven quickly. It is also important to avoid cognitive underloading of the driver. On a straight stretch of road with little traffic and at a constant speed, the driver is at risk of losing concentration. In the former case, the driver finds it easy to record and evaluate a variety of information from a human-machine interface. In the second case, the driver could be overwhelmed by too much information and no longer pay enough attention to the traffic. In the third case of vigilance, the driver finds it more difficult to absorb and appropriately process changing and added information.

A driving situation describes the interaction between driver, vehicle and environment.

Adaptive displays are known per se. In the simplest form, it is known to use different display modes during the day and at night, for example a night mode on a display. It is also known to vary the color of the display. Furthermore, it is known to configure the display depending on the driving mode, for example in sports cars differently for road operation and for racetrack operation. There are also variable displays in which information can be shown in different sizes. The corresponding adaptations to the displays are made either automatically, for example depending on the ambient brightness, or manually, for example when selecting a display mode or a display color.

From the DE 10 2007 058 437 A1 a method for conveying information to the driver of a motor vehicle is known, in which optical information displays are displayed to the driver using a head-up display, the overall display of the head-up display comprising a group of adjacent display fields in which information displays can be displayed , wherein each element of a set of possible information displays is assigned to one of the display fields for display, with characteristic driving situation parameters of the current driving situation being determined, with the information displays assigned to this display field being prioritized for each display field depending on the characteristic driving situation parameters, and the content of each display field in Depending on the prioritization made for this display field.

The DE 103 43 683 A1 relates to a system and method for providing information depending on the factors that stress the driver when driving the motor vehicle.

Furthermore, the revealed EP 1 330 377 B1 a method for controlling a driver assistance system, wherein a response of the system to an operating situation of a vehicle is selected depending on a driver's cognitive load.

From the US 2018/0365025 A1 an adaptive user interface is known with a user interface that can be adapted to a user experience using a learning algorithm.

The driving situation parameters known from the prior art can sometimes coincidentally correlate with the driver's cognitive load, but there are also situations in which a large amount of information would be displayed even though the driver's cognitive load is high. In these cases, the amount of information displayed could overwhelm the driver.

The task is therefore to further develop methods for displaying information on a human-machine interface of a motor vehicle, computer program products, human-machine interfaces and motor vehicles of the type mentioned at the beginning so that the cognitive load of a driver can be determined more directly and therefore more precisely and the human-machine interface can be configured accordingly. Another task is to to be able to take into account the individual cognitive performance of a specific driver.

The object is achieved by a method for displaying information on a human-machine interface according to claim 1, a computer program product according to the independent claim 9, a human-machine interface according to the independent claim 10, a human-machine interface according to the independent claim 11 and a motor vehicle according to independent claim 14. Further refinements and further developments are the subject of the dependent claims.

A method for controlling a human-machine interface of a motor vehicle is described below, wherein the human-machine interface has at least one display device for displaying information, wherein at least one driving situation parameter is determined, which characterizes a driving situation of the motor vehicle, and depending on of the determined driving situation parameter, a load parameter is determined by means of an algorithm, which characterizes a cognitive load of a driver of the motor vehicle, so that an information density of the at least one display device of information to be displayed by means of the display device is adjusted depending on the load parameter, the algorithm comprising a data set with driving situation parameters, wherein each driving situation parameter is assigned a load parameter, the load parameters being based on at least one measured variable that characterizes the driver's behavior.

The algorithm is trained using a machine learning method by carrying out test drives with a plurality of test drivers, with characteristic parameters being recorded and stored during the test drives to describe a driving situation. During the test drives, the test drivers receive input prompts to an input element of a human-machine interface, with an input duration for carrying out the input, an input error rate and/or a number of operating steps being recorded and correlated with the characteristic parameters by the algorithm.

By determining a driver's cognitive load and choosing the appropriate information density, it can be ensured that the driver is only shown as much information as he can process. An impairment of the driver's information processing capacity due to too much and changing information can be avoided and the driver has more mental resources free to devote to what is happening on the road.

A selection of the most relevant information can be made depending on the situation, for example navigation information if the driver is about to reach a junction, speed information if the driver is driving too fast or too slow, or a warning if, for example, there is traffic requires increased attention.

Machine learning methods make it possible to reliably assess very complex situations and, in particular, to make predictions about future cognitive loads depending on a prediction of the various driving situations. Such predictions are difficult to make with classically structured algorithms. Furthermore, algorithms of this type can be continuously improved during ongoing operation, so that the prediction of cognitive load and, depending on the design, the categorization of drivers into different user type classes continue to improve.

The algorithm can be used in the form of artificial intelligence, which has a complex filter that can continue to evolve.

In a first further embodiment, it can be provided that a behavior parameter is also determined that characterizes a current behavior of the driver, the information density of the information to be displayed on the display device being varied taking into account the determined behavior parameter.

Driver behavior is a reliable measure of driver cognitive load. This allows the information density to be better adapted to the driver's current needs.

In a further further embodiment, it can be provided that the calculation of the load parameter includes a prediction of the driver's cognitive load at a predetermined time, in a predetermined period of time and/or a predetermined distance.

In this way, a prediction can be made about the cognitive load to be expected shortly and a suitable display configuration can be selected at an early stage depending on the cognitive load to be expected. This means that even before a situation requiring attention occurs A selection of the information displayed focuses on the most important information.

In a further further embodiment, it can be provided that during normal driving operation of the motor vehicle, measured variables relating to the driver are recorded and the driver is categorized into one of several user type classes, the information density being adjusted depending on the user type class of the driver.

It is known that different drivers have different cognitive performance depending on different aspects and are stressed to different extents by different driving situations. A novice driver requires more attention and therefore cognitive effort than an experienced driver. There are also certain differences, for example age-related, in the ability to absorb and process information between different drivers. Furthermore, it is known that the duration of the journey has an influence on how attentive a driver is. Such measured variables can be recorded within the scope of this embodiment, whereby a display configuration tailored to the driver can be selected.

This embodiment can be used to provide different sets of information for different types of drivers. Some drivers can process a greater amount of information than others, which can prevent those drivers who cannot process as much information from being overloaded by the human-machine interface.

In a further further embodiment, it can be provided that the driving situation-related input variables include measured variables for the driver, the motor vehicle and/or the environment.

Environmental-related metrics can include road layout, traffic, weather, time of day and the like, driver-related metrics can include interactions with the human-machine interface, line of sight, driving behavior and the like, vehicle-related metrics can include dynamics, assistance system use and the like.

It has been shown that these parameters are suitable for predicting cognitive load for the driver. The faster the driver drives, the more attention he has to pay to the road. The more curvy a road is, the more steering input is required, which puts a cognitive strain on the driver. The more road users there are on the road, the more attention is required because the driver has to observe the behavior of other road users and, if necessary, react to them. For example, if a motor vehicle in front of the driver brakes, the driver must also brake. Monotonous driving situations, such as routes without curves with few road users, can reduce the driver's attention and concentration. This can affect the driver's reaction behavior.

In a further further embodiment, it can be provided that at least two display configurations are provided which have different information densities, with an adjustment of the information density being carried out by selecting a display configuration.

More than two different display configurations can be provided, which in some embodiments can differ in more than one parameter dimension. Thus, different display configurations may display different amounts of information, display some information in different sizes, and change the location of information, keeping in mind that a driver typically expects information to be in a particular location, such as a speedometer, rpm, range, and so on the like. Configurations are therefore conceivable in which information is picked out and displayed centrally in the appropriate size. The remaining information can be located in specific fields intended for the corresponding information.

In a further further embodiment, it can be provided that the human-machine interface has at least two display devices, on each of which a plurality of information relevant to a driver of the motor vehicle can be displayed, the algorithm being used to determine which information is displayed on which display device becomes.

Corresponding display devices can be, for example, a central information display, a cockpit display and/or a head-up display. In this way, very important information can be displayed, for example, on a display device arranged close to the driver's field of vision, for example on a head-up display.

In a further further embodiment, it can be provided that during the test drives there are additional measures for the cognitive load of the test drivers characteristic measured variables are recorded and correlated with the characteristic parameters of the driving situation, the characteristic measured variables being a vehicle position in a lane, a number of lane crossings, a duration of lane crossing, a deviation from a target speed, head movements, arm movements, steering wheel inputs, pedal inputs, and / or can include a line of sight of the driver.

Reaction time to complex tasks is a useful measure to determine a driver's cognitive load in a given situation. The more attention the driver requires for the driving activity in the current driving situation, the longer it takes him to complete the tasks. The precision of the ride, i.e. H. How exactly the driver drives within his lane, how often he crosses the lane, how long he stays outside the lane, etc. are an indication of how much cognitive strain the driver is currently under.

Monitoring the driver himself, for example using an interior camera, can also allow conclusions to be drawn about the driver's current condition. Rapid changes in gaze and stabilization inputs via the pedals and steering wheel can indicate a high visual-manual load, whereas a driver who is not heavily stressed tends to show balanced gaze and stabilization behavior. Furthermore, hectic pedal movements or steering wheel inputs can indicate a high cognitive load.

By using several different test drivers, it is also possible to better differentiate between the different user types.

A first independent subject relates to a device for controlling a human-machine interface of a motor vehicle, wherein the human-machine interface has at least one display device for displaying information, with means for determining at least one driving situation parameter that characterizes a driving situation of the motor vehicle, are provided and wherein means for determining a load parameter are provided as a function of the determined driving situation parameter by means of an algorithm, wherein the load parameter characterizes a cognitive load of a driver of the motor vehicle, wherein means for adjusting an information density of the at least one display device depending on information to be displayed by means of the display device of the load parameter are provided, wherein the algorithm comprises a data set with driving situation parameters, a load parameter being assigned to each driving situation parameter, the load parameters being based on at least one measured variable that characterizes the behavior of the driver.

In a first further embodiment, it can be provided that means for determining at least one behavioral parameter are also provided, wherein the behavioral parameter characterizes a current behavior of the driver, wherein the means for adjusting an information density are set up to reduce the information density of the information to be displayed on the display device Taking into account the determined behavioral parameter.

In a further further embodiment it can be provided that prediction means are provided for calculating the cognitive load of the driver at a predetermined time, in a predetermined period of time and/or a predetermined distance.

In a further further embodiment, it can be provided that sensors are provided for detecting measured variables relating to the driver in normal driving operation of the motor vehicle and categorization means are provided in order to categorize the driver into one of several classes, the device being set up to display the display configuration in Depending on the user type class of the driver.

In a further further embodiment, it can be provided that the driving situation-related input variables include parameters for the driver, the vehicle, the environment, the vehicle dynamics, road guidance, the traffic condition and/or the weather.

In a further further embodiment, it can be provided that at least two display configurations are provided which have a different information density, the means for adjusting an information density being set up to adjust the information density by selecting a display configuration.

In a further further embodiment, it can be provided that the human-machine interface has at least two display devices, on each of which a plurality of information relevant to a driver of the motor vehicle can be displayed, the algorithm being set up to determine which information is on which Display device is displayed.

In a further further embodiment, it can be provided that the algorithm is trained using a machine learning method by carrying out test drives with a plurality of test drivers, with characteristic parameters being recorded during the test drives to describe a driving situation using detection means and stored in a memory , wherein the test drivers receive input requests to an input element of a human-machine interface during the test drives through communication means, an input duration for carrying out the input being recorded by means of a clock and correlated with the characteristic parameters by the algorithm.

In a further further embodiment, it can be provided that detection means for detecting further measured variables characteristic of the cognitive load of the test drivers are provided by detection means during the test drives and computing means for correlation with the characteristic parameters of the driving situation, the characteristic measured variables being a vehicle position on a lane, a number of lane violations, a duration of lane violation, a deviation from a target speed, head movements, arm movements, steering wheel inputs, pedal inputs, and / or a direction of view of the driver.

Another independent subject relates to a computer program product, with a computer-readable storage medium on which instructions are embedded which, when executed by at least one computing unit, cause the at least one computing unit to be set up to carry out the method according to one of the preceding claims.

The method can be carried out on one or more computing units, so that certain method steps are carried out on one computing unit and other method steps on at least one further computing unit, whereby calculated data can be transmitted between the computing units if necessary.

Another independent subject relates to a human-machine interface with at least one display device and at least one computing unit as well as at least one storage medium connected to the computing unit, on which a computer program product of the type described above is stored, which can be executed by the computing unit.

A corresponding human-machine interface can have the aforementioned properties and advantages when used in a motor vehicle.

Another independent subject relates to a human-machine interface of a motor vehicle, with at least one display device for the optical representation of information relevant to a driver of the motor vehicle, at least one computing unit being provided which is connected to at least one storage medium, the computing unit being used for control the at least one display device is provided, with at least two different display configurations for operating the at least one display device being stored on the storage medium, the computing unit being connected to at least one driving situation sensor which detects measured variables relating to a driving situation, the computing unit being set up to use the Input variables of the driving situation sensor by means of an algorithm to determine at least one driving situation parameter that characterizes a driving situation of the motor vehicle, and depending on the determined driving situation parameter to determine a load parameter that characterizes a cognitive load of a driver of the motor vehicle, and depending on the load parameter to select one of the display configurations and by means of the at least one display device, the algorithm comprising a data set with driving situation parameters, each driving situation parameter being assigned a load parameter, the load parameters being based on at least one measured variable that characterizes the behavior of the driver.

The algorithm was trained using a machine learning method by carrying out test drives with a plurality of test drivers, with characteristic parameters being recorded and stored during the test drives to describe a driving situation. During the test drives, the test drivers received input prompts to an input element of a human-machine interface, with an input duration for carrying out the input, an input error rate and/or a number of operating steps being recorded and being correlated with the characteristic parameters by the algorithm.

A corresponding driving situation sensor can be, for example, a camera, a radar system, a motor vehicle-to-vehicle communication device, a motor vehicle-to-environment communication device, a navigation system, and/or a weather sensor such as a temperature and/or rain sensor.

The algorithm can be, for example, a map-based algorithm.

In a first further embodiment, it can be provided that the computing unit is connected to at least one driver monitoring sensor which detects measured variables relating to a driver, wherein the computing unit is designed to calculate the load parameter using the measured variables of the driver monitoring sensor.

A corresponding driver monitoring sensor can, for example, be a camera that is aimed at the driver and records his head movement, gestures and gaze parameters such as eyelid closing times, gaze duration, gaze direction and/or eye movements, etc. Another driver monitoring sensor can be a pure eye sensor.

In a further further embodiment, it can be provided that the computing unit is designed to monitor an interaction of the driver with the human-machine interface and is designed to record measured variables describing the interaction.

This allows input actions, e.g. operating time, operating errors, interruptions in operation, to be monitored and used in the display configuration.

In a further further embodiment, it can be provided that the computing unit is connected to at least one vehicle sensor, which detects measured variables relating to a driving state of the motor vehicle.

Such a sensor can be, for example, a pedal sensor for detecting pedal inputs, a steering wheel sensor for detecting steering inputs, a speed sensor, an acceleration sensor for detecting an acceleration and/or a cornering speed and/or a slip sensor.

In a further further embodiment, it can be provided that at least two display devices are provided, with the computing unit being set up to use the calculated cognitive load to select which of the at least two display devices displays driver-relevant information.

Such display devices can be, for example, a head-up display, a central information display and/or a cockpit display.

In a further further embodiment, it can be provided that the algorithm stored on the storage medium is a self-learning algorithm.

Self-learning algorithms are able to process very complex situations.

Depending on the design, the algorithm can also have artificial intelligence.

Another independent subject relates to a motor vehicle with a human-machine interface of the type described above.

• 1 ein Kraftfahrzeug mit einer Mensch-Maschine-Schnittstelle;
• 2A, B Anzeigevorrichtungen der Mensch-Maschine-Schnittstelle aus 1 in zwei unterschiedlichen Konfigurationen;
• 3 eine Fahrsituation, in der sich das Kraftfahrzeug aus 1 befindet;
• 4A, B Prinzipskizzen zum Training eines Algorithmus der Mensch-Maschine-Schnittstelle aus 1, sowie
• 5 eine Prinzipskizze der Entscheidungsarchitektur der Mensch-Maschine-Schnittstelle.

Further features and details emerge from the following description, in which at least one exemplary embodiment is described in detail - if necessary with reference to the drawing. Described and/or illustrated features form the subject matter on their own or in any meaningful combination, possibly also independently of the claims, and in particular can also be the subject of one or more separate applications. Identical, similar and/or functionally identical parts are provided with the same reference numerals. Show schematically:

• 1 a motor vehicle with a human-machine interface;
• 2A, B Display devices of the human-machine interface 1 in two different configurations;
• 3 a driving situation in which the motor vehicle is 1 located;
• 4A, B Principle sketches for training an algorithm for the human-machine interface 1 , as well as
• 5 a schematic sketch of the decision architecture of the human-machine interface.

1 shows a motor vehicle 2 with a human-machine interface 4 (components framed in dashed lines).

The human-machine interface 4 is an interface between the motor vehicle 2 and a driver 6 and possibly other passengers. Using the human-machine interface 4, the driver 6 can receive information from and about the motor vehicle 2 and can make control inputs for the motor vehicle 4 using suitable input means (not shown), e.g. for navigation, multimedia control, air conditioning, lighting, etc.

In the present case, the human-machine interface 4 has three displays, a head-up display 8, a cockpit display 10 and a central information display 12. The head-up display 8 displays information on a windshield of the motor vehicle 2, the cockpit display 10 is arranged in front of a steering wheel (not shown) directly in front of the driver 6 and the central information display 12 near the middle of one (in 1 (not shown) dashboard.

The human-machine interface 4 has a controller 14 with a computing unit 16 and a storage medium 18 on which an algorithm 20 is stored, which will be explained in detail below.

The controller 14 is connected to a variety of sensors, of which, for example, in 1 a radar sensor 22, a steering angle sensor 24, a traffic camera 26 and a driver monitoring camera 28 are shown. The controller 14 can be connected to a number of other sensors.

In addition to the radar sensors 22 and the traffic camera 26, a first group of such sensors can be other sensors for traffic monitoring such as lidar or communication devices such as motor vehicle-to-vehicle communication means or motor vehicle-to-infrastructure communication means, navigation systems, weather sensors and the like, all of which are for preservation to provide information about the traffic situation.

In addition to the steering angle sensor 24, a second group of corresponding sensors can be sensors for monitoring the condition of the motor vehicle, for example speed sensors, wheel speed sensors, drive sensors, pedal sensors, etc.

In addition to the driver monitoring camera 28, fatigue sensors can be used for the third group of driver monitoring sensors, but the pedal sensors and the steering angle sensor 24 already mentioned in connection with the second group can also be used for driver monitoring, since driver actions on control means of the motor vehicle 2, e.g. steering wheel, gas or Brake pedal, can be detected.

Radar sensors 22 and traffic camera 26 are used to monitor the traffic in front of the motor vehicle 2. With the help of this information, the controller 14 can make predictions about the current and future traffic situation and situation using the algorithm 20.

Using the steering angle sensor 24 and other sensors for monitoring the motor vehicle 2, the controller 14 can obtain information about the dynamic state of the motor vehicle 2.

With the help of the driver monitoring camera 28, it is possible to determine how busy or how distracted the driver 6 is at the moment. The driver monitoring camera 28 can, for example, detect movements and orientation of a head 30 of the driver 6 and/or his eyes. In addition to monitoring the control inputs into the motor vehicle 2 using the steering wheel and pedals, such information can allow conclusions to be drawn about the cognitive load of the driver of the motor vehicle 2.

2A, B the display devices show the 8, 10, 12 of the human-machine interface 4 1 in two different display configurations A and B.

The display devices 8, 10, 12 are arranged on a dashboard 32.

The head-up display 8 shows in the in 2A The configuration shown displays information about the current driving speed 34, navigation information 36 and lane departure warning information 38, thus providing the driver with comprehensive information.

The cockpit display 10 also shows information about the current driving speed 34 and speed information 40. In addition, multimedia information 42 is displayed.

Map information 44 is displayed on the central information display 12.

2A represents a display configuration A in which the driver's cognitive load is low and he can record and process a variety of information from the human-machine interface 4.

In display configuration B according to 2 B On the other hand, the cognitive load is higher and it is compared to the display configuration 2 A very reduced amount of information presented.

So only the navigation information 36 is shown on the head-up display 8. Only the driving speed information 34 is shown on the cockpit display 10.

In addition to the map information 44, the multimedia information 42 is also displayed on the central information display 12.

In the 2 B The configuration shown allows the driver to operate in the appropriate to quickly capture the most relevant information in the driving situation.

3 shows a driving situation in which the motor vehicle 2 is 1 located.

The motor vehicle 2 drives towards a road intersection 46. There is a lot of traffic at intersection 46. Another motor vehicle 48 is driving in front of the motor vehicle 2, another motor vehicle 50 is approaching the motor vehicle 2, the street intersection 46 is crossed by another motor vehicle 52, an oncoming motor vehicle 54 is waiting and a pedestrian 56 is walking across a zebra crossing 58.

At the time of the illustration, the cognitive load on the driver 6 of the motor vehicle 2 is only slightly increased as he is preparing to navigate at the upcoming intersection 46. The closer he gets to the road intersection 46, the higher the cognitive load becomes, since on the one hand he has to monitor the behavior of the large number of road users 48 to 56 and on the other hand he still has to navigate. In this case, the driver 6 finds it difficult to absorb all the available information regarding navigation and motor vehicle 2, which is why it makes sense to significantly reduce the number of information displayed on the human-machine interface 4 and to keep the most important information correspondingly large and present them succinctly so that they immediately catch the eye.

4A, B show principle sketches for training the algorithm 20 of the human-machine interface 4 1 .

The algorithm 20 has two different dimensions of classification, on the one hand regarding the current driving situation and on the other hand regarding the driver. Algorithm 20 is a self-learning algorithm that is developed through training using test drivers on test drives.

During the training of the algorithm 20, characteristic parameters are recorded and evaluated using a data recorder to describe a driving situation. Furthermore, other measured variables that are characteristic of driver distraction, such as eye or head movements and direction of gaze, gestures and/or pedal and steering angle settings, can be recorded. Furthermore, it can be recorded how long it takes for the driver to make a steering input, where he is in the lane, under what circumstances, how long and how often he leaves the lane and how constant the speed is maintained.

During the aforementioned test drives, the test drivers are repeatedly asked to make one or more standardized operating inputs on the human-machine interface 4. Different traffic conditions prevail during the operating inputs. The input time that a test driver needs for the required input correlates, on the one hand, with the driver's cognitive load caused by the driving situation and, on the other hand, with the driver's distraction, for example due to the request to enter information.

The measured operating times for completing the tasks are recorded and evaluated together with the characteristic parameters.

In 4A shows how different drivers are sorted into different driver type classes based on the previously described input variables for a current situation (vehicle dynamics, route, traffic, weather, human-machine interface inputs).

Experienced drivers will have fewer lane deviations in comparable driving situations and will be able to make entries into the human-machine interface 4 more quickly than less experienced drivers. The latter should be less burdened by the human-machine interface 4 than the former, for example through a simplified menu navigation. Such drivers are then sorted into a lower driver type class than experienced drivers.

In 4B shows how an analysis of the driver's cognitive load is carried out based on the same input variables depending on the current position, the current route section, the traffic, the weather, etc.

This means that, on the one hand, a preselection of the display configurations A, B that come into question can be carried out and, on the other hand, a correlation can be made between the driving situation and the cognitive load. From this, the algorithm 20 used in the motor vehicle 2 can be developed, which decides on the information displayed on the displays 8, 10, 12 of the human-machine interface 4.

5 shows a schematic sketch of the decision architecture of the human-machine interface 4.

During later driving, the same input variables are used as during training according to 4A, B recorded and evaluated using the artificial intelligence algorithm 20.

In addition to the current inputs, a so-called digital electronic horizon is created, i.e. a prediction about the driving situation that will prevail in the near future. The horizon can cover a distance of, for example, 50 to 200 m or 1000 m or a period of, for example, 5 to 30 seconds. The corresponding input variables can be dynamically predicted via the digital electronic horizon (e.g. using ADASIS).

On the one hand, the algorithm 20 calculates from the input variables which driver type class the current driver belongs to. On the other hand, the algorithm 20 makes a prediction about what cognitive load will predominate in the upcoming section of the route. This information is transferred to the controller 14 of the human-machine interface 4, which selects a corresponding display configuration A, B from it.

Depending on the user type, for example, three cognitive performance levels (3 high - 2 medium - 1 low) could be mapped onto five display levels (4 max - 3 high - 2 medium - 1 low - 0 minimal). For a driver with performance level 3 high, load levels 3-2-1 can be assigned to display levels 4-3-2, for example, a driver with performance level 2 can be assigned display levels 3-2-1 and a driver with performance level 1 can be assigned display levels 2-1 -0.

In alternative embodiments, instead of artificial intelligence, a map can be created, according to which a corresponding display configuration A, B is selected depending on the input variables.

Although the subject matter has been illustrated and explained in detail by exemplary embodiments, the invention is not limited by the disclosed examples and other variations can be derived therefrom by those skilled in the art. It is therefore clear that a large number of possible variations exist. It is also to be understood that exemplary embodiments are merely examples and should not be construed in any way as limiting the scope, application, or configuration of the invention. Rather, the preceding description and the description of the figures enable the person skilled in the art to concretely implement the exemplary embodiments, whereby the person skilled in the art can make a variety of changes, for example with regard to the function or the arrangement of individual elements mentioned in an exemplary embodiment, with knowledge of the disclosed inventive concept, without To leave the scope of protection defined by the claims and their legal equivalents, such as further explanation in the description.

Reference symbol list

2

motor vehicle

4

Human-machine interface

6

driver

8th

Head-Up Display

10

Cockpit display

12

central information display

14

steering

16

Computing unit

18

storage medium

20

algorithm

22

Radar sensor

24

Steering angle sensor

26

Traffic camera

28

Driver monitoring camera

30

Head

32

dashboard

34

Driving speed information

36

Navigation information

38

Lane departure warning information

40

Speed information

42

Multimedia information

44

Card information

46

street intersection

48, 50, 52, 54

motor vehicle

56

pedestrian

58

Zebra crossing

AWAY

Display configuration

Claims (14)

Method for controlling a human-machine interface (4) of a motor vehicle (2), wherein the human-machine interface (4) has at least one display device (8, 10, 12) for displaying information (34, 36, 38, 40 , 42, 44), wherein at least one driving situation parameter is determined, which represents a driving situation of the Motor vehicle (2), and wherein, depending on the determined driving situation parameter, a load parameter is determined by means of an algorithm (20), which characterizes a cognitive load of a driver (6) of the motor vehicle (2), so that an information density of the at least one display device (8, 10, 12) of information to be displayed by means of the display device (8, 10, 12) is adapted depending on the load parameter, the algorithm (20) comprising a data set with driving situation parameters, a load parameter being assigned to each driving situation parameter, the load parameters being based on based on at least one measured variable that characterizes the driver's behavior, characterized in that the algorithm (20) is trained by a machine learning method by carrying out test drives with a plurality of test drivers, with characteristic driving situations being used to describe a driving situation during the test drives Parameters are recorded and stored, with the test drivers receiving input prompts to an input element of a human-machine interface during the test drives, an input duration for carrying out the input, an input error rate and / or a number of operating steps being recorded and with the characteristic parameters by the algorithm (20) can be correlated. Procedure according to Claim 1 , wherein a behavior parameter is also determined which characterizes the driver's current behavior, the information density of the information to be displayed on the display device (8, 10, 12) being varied taking into account the determined behavior parameter. Procedure according to Claim 1 or 2 , wherein the calculation of the load parameter includes a prediction of the cognitive load of the driver (6) at a predetermined time, in a predetermined period of time and / or a predetermined distance. Method according to one of the preceding claims, wherein in normal driving operation of the motor vehicle (2), measured variables relating to the driver (4) are recorded and the driver (4) is categorized into one of several classes, the information density depending on the user type class of the driver (6) is adjusted. Method according to one of the preceding claims, wherein the driving situation-related input variables include measured variables from the environment, driver (6) and/or motor vehicle (2). Method according to one of the preceding claims, wherein at least two display configurations (A, B) are provided which have a different information density, the information density being adjusted by selecting a display configuration (A, B). Method according to one of the preceding claims, wherein the human-machine interface (4) has at least two display devices (8, 10, 12), each of which displays a plurality of information (34) relevant to a driver (4) of the motor vehicle (2). , 36, 38, 40, 42, 44) can be displayed, the algorithm (20) being used to determine which information (34, 36, 38, 40, 42, 44) is displayed on which display device (8, 10, 12). becomes. Method according to one of the preceding claims, wherein during the test drives further measured variables characteristic of the cognitive load of the test drivers (4) are recorded and correlated with the characteristic parameters of the driving situation, the characteristic measured variables being a vehicle position in a lane, a number of lane violations , a duration of lane crossing, a deviation from a target speed, head movements, arm movements, steering wheel inputs, pedal inputs, and / or a direction of view of the driver. Computer program product, with a computer-readable storage medium (18) on which instructions are embedded which, when executed by at least one computing unit (16), cause the at least one computing unit (16) to be set up to carry out the method according to one of the preceding to carry out claims. Human-machine interface (4) with at least one display device (8, 10, 12) and at least one computing unit (16) and at least one storage medium (18) connected to the computing unit (16) on which a computer program product is stored Claim 9 is stored, which can be executed by the computing unit (16). Human-machine interface of a motor vehicle (2), with at least one display device for the optical representation of information relevant to a driver (4) of the motor vehicle (2), at least one computing unit (16) being provided, which is connected to at least one storage medium (18 ) is connected, wherein the computing unit (16) is provided for controlling the at least one display device (8, 10, 12), with at least two different display configurations (A, B) on the storage medium (18) for operating the at least one display device (8 , 10, 12) are stored, the computing unit (16) being connected to at least one driving situation sensor (22, 26) which detects at least one measurement variable relating to the driving situation, the computing unit (16) being set up to use the input variables of the driving situation sensor (22, 26). an algorithm (20) to determine at least one driving situation parameter that characterizes a driving situation of the motor vehicle (2), and depending on the determined driving situation parameter to determine a load parameter that characterizes a cognitive load of a driver (6) of the motor vehicle (2), and in Depending on the load parameter, one of the display configurations (A, B) is selected and displayed by means of the at least one display device (8, 10, 12), the algorithm (20) comprising a data set with driving situation parameters, a load parameter being assigned to each driving situation parameter, the load parameters being on Based on at least one measured variable that characterizes the behavior of the driver, characterized in that the algorithm (20) has been trained by a machine learning method by carrying out test drives with a plurality of test drivers, with the purpose of describing during the test drives Parameters characteristic of a driving situation have been recorded and stored, the test drivers receiving input prompts to an input element of a human-machine interface during the test drives, an input duration for carrying out the input, an input error rate and / or a number of operating steps being recorded and have been correlated with the characteristic parameters by the algorithm (20). Human-machine interface Claim 11 , wherein the computing unit (16) is connected to at least one driver monitoring sensor (28) which detects measured variables relating to a driver (4), wherein the computing unit (16) is designed to calculate the load parameter using the measured variables of the driver monitoring sensor (28). Human-machine interface Claim 12 , wherein the computing unit (16) is designed to monitor an interaction of the driver (6) with the human-machine interface (4) and is designed to record measured variables describing the interaction. Motor vehicle with a human-machine interface (4) according to one of the Claims 10 until 13 .

Images