DE102019201407A1 - Method and device for controlling a driver's attention - Google Patents

Abstract

The invention relates to a method for controlling the attention of a driver of an ego vehicle (1) comprising the steps: - detection of traffic objects in the environment of the host vehicle (1) by a detection device (2), - determination for the detected traffic objects whether a critical or uncritical one Driving situation exists, - division of the detected traffic objects into critical and uncritical traffic objects, based on the previous determination of the critical and uncritical driving situations, whether the at least one existing critical traffic object was perceived by the driver of the host vehicle (1), outputting at least one warning signal when a perception deficit is determined. Furthermore, the invention relates to a device for controlling attention e ines driver, an ego vehicle and a computer program.

Description

The invention relates to a method for controlling the attention of a driver of a first-person vehicle. The invention further relates to a device for controlling the attention of a driver of the first-person vehicle. The device is particularly suitable for carrying out the method. The invention further relates to a vehicle comprising such a device and a computer program.

Numerous traffic accidents result from the fact that a driver has not noticed relevant traffic objects / other road users. To increase traffic safety, today's systems can use other environmental sensors to detect other traffic objects and, if necessary, initiate emergency braking if a collision is imminent. However, this automated maneuver is not only uncomfortable for the occupants, but can also endanger traffic safety in certain situations.

Methods are known which detect a turning away of the driver's head from the traffic situation, for example with the aid of a driver camera, and warn the driver. In addition, information such as the vehicle speed or the time-to-collision for any vehicle in front can be evaluated for the activation of a warning.

The DE 102009005688 A1 discloses a method for acoustically indicating a source of danger in the vicinity of a vehicle, in particular a motor vehicle. In the method, the location of a hazard source detected by a sensor system in the surroundings of the vehicle is determined and / or provided. The sound image of an acoustic signal present in the vehicle interior is adapted such that the acoustic signal contains a virtual echo representing the location of the source of the hazard.

The DE 19734307 A1 relates to a method for monitoring and / or influencing the vehicle behavior of a driver-controlled motor vehicle, comprising the steps of: determining a line of sight of the driver of the motor vehicle by means of a line of sight determination device and providing at least one line of sight measurement value, comparing the line of sight of the driver with a preset value and / or measured value, and generating one Warning, information and / or actuator signal depending on the comparison result.

The GB 2534163 B discloses a vehicle interface device for generating an audible indication of a potential hazard, comprising: a plurality of electroacoustic transducers for generating an audio object and a processor. The processor is configured to determine the angular position of the identified hazard relative to the vehicle, generate a control signal to cause the electroacoustic transducers to generate an audio object, and modify the control signal to progressively change a perceived spatial location . The vehicle interface device may also include a display on which a visual display of the potential hazard can be generated. By progressively changing the audio object, the vehicle interface can provide a substantially continuous display of the angular position of the identified object with respect to the vehicle. The perceived spatial location of the audio object can be changed progressively to track the particular angular position of the identified object.

Starting from the prior art, it is an object of this invention to provide a method and a device for increased traffic safety.

This object is achieved by specifying a method for controlling the attention of a driver of a first-person vehicle with the features of claim 1 and specifying a device for controlling the attention of a driver of an first-person vehicle with the features of claim 18. Furthermore, the object is achieved by specifying an ego vehicle with the features of claim 20. In addition, the object is achieved by specifying a computer program with the features of claim 21.

The subclaims list further advantageous measures which can be combined with one another as desired in order to achieve further advantages.

• - Erfassen von Verkehrsobjekten im Umfeld des Egofahrzeuges durch eine Erfassungseinrichtung,
• - Bestimmen für die erfassten Verkehrsobjekte, ob eine kritische oder unkritische Fahrsituation vorliegt,
• - Einteilen der erfassten Verkehrsobjekte in kritische und unkritische Verkehrsobjekte, anhand der vorangegangenen Bestimmung der kritischen und unkritischen Fahrsituationen,
• - Erfassen der Blickrichtung des Fahrers durch eine Blickrichtungserfassungseinrichtung, bei Vorhandensein zumindest eines kritischen Verkehrsobjektes,
• - Feststellen, zumindest anhand der erfassten Blickrichtung, ob das zumindest eine vorhandene kritische Verkehrsobjekt durch den Fahrer des Egofahrzeuges wahrgenommen wurde,
• - Ausgeben zumindest eines Warnsignals bei einem festgestellten Wahrnehmungsdefizit.

The object is achieved by specifying a method for controlling the attention of a driver of a first-person vehicle, comprising the steps:

• - Detection of traffic objects in the environment of the ego vehicle by a detection device,
• - Determine for the detected traffic objects whether there is a critical or non-critical driving situation,
• - division of the detected traffic objects into critical and uncritical traffic objects, based on the previous determination of the critical and uncritical driving situations,
• - Detection of the driver's line of sight by a line of sight detection device, at Presence of at least one critical traffic object,
• Determining, at least on the basis of the detected line of sight, whether the at least one critical traffic object was perceived by the driver of the first-person vehicle,
• - Output at least one warning signal in the event of a perceived deficit.

Here, the lack of perception means that it has been established that the driver in the first-person vehicle has not perceived the at least one critical traffic object. The perception can be designed as a visual perception.

Traffic objects can be moving objects such as vehicles hereinafter referred to as traffic vehicles, pedestrians hereinafter referred to as traffic pedestrians etc. or immovable objects such as street signs, traffic lights etc.

According to the invention, it was recognized that there is a so-called “warning dilemma” in known methods: In the case of a warning dilemma, warnings are given too often, since any turning away from the driver's gaze is classified as dangerous and the first-person vehicle / warning system has no information as to whether and how the driver does it Has perceived driving situation, i.e. which traffic objects in the current traffic situation he is most likely aware of or the first-person vehicle / warning system warns too rarely because the driver should not be irritated with unnecessary false warnings. Both cases represent a deficiency in traffic safety. This is avoided by the method according to the invention.

According to the invention, the first-person vehicle uses a detection device and preferably with the aid of a (surroundings) map to assess which traffic objects are critical to the driving situation. Critical also includes relevant. Critical or relevant traffic objects are, for example, those traffic objects whose trajectory could intersect or could have an intersection with the trajectory of the first-person vehicle. The method now compares whether the traffic objects identified as critical were viewed by the driver of the first-person vehicle or whether there is a risk that the driver might have overlooked them (lack of perception). Depending on the result, a warning signal is issued.

The “warning dilemma” is thus resolved by the method according to the invention: only a warning signal is issued if the driver has not viewed at least one critical traffic object. This avoids unpleasant emergency braking interventions. Critical situations in traffic are avoided, which leads to increased traffic safety. The "warning dilemma" is thus resolved by adaptive warnings that only occur if a perception deficit for critical traffic objects has been identified. The method thus enables a more in-depth interpretation of the traffic situation in conjunction with an assessment of driver perception for risk assessment.

It is preferred to determine whether the at least one critical traffic object was perceived, depending on a peripheral field of vision of the driver and / or the duration of the driver's gaze fixation on the at least one critical traffic object and / or the pupil narrowing of the driver's pupil the fixation on the at least one existing critical traffic object. This enables a sufficiently precise determination to be made. A constriction of the pupil can be used to assume a probability for the concentration of driver attention on the object under consideration and thus a probability for “as seen”. A probability of “as seen” can be assumed from the temporal evaluation of the fixation. By including the peripheral field of view, a “probability cone” can be defined that gives conclusions about whether a critical traffic object has been visually detected. The determination as to whether the at least one critical traffic object was perceived is preferably dependent on the peripheral field of vision of the driver and the temporal fixation period of the driver's gaze on the at least one critical traffic object and possibly additionally on the pupil narrowing of the driver's pupil in the Fixation on the at least one existing critical traffic object is determined. This enables an improved determination.

The determination as to whether the at least one critical traffic object was perceived is preferably determined as a function of at least one corresponding driver reaction of the driver of the first-person vehicle. Corresponding driver reactions are, for example, control interventions on the gas, brakes and / or steering, but also control interventions on the turn signals, horn, warning devices, lights, entertainment or media devices and / or communication devices, as well as verbal and non-verbal driver reactions and / or the interruption of non-driver activity , all of which support the evaluation of whether a driver seems to have perceived a particular traffic object because he shows a reaction to it.

The determination as to whether the at least one critical traffic object was perceived is preferably dependent on the peripheral field of vision of the driver and the temporal Duration of fixation of the driver's gaze on the at least one critical traffic object present and determined by at least one corresponding driver reaction of the driver of the first-person vehicle. This enables a very precise determination.

In a preferred embodiment, the driving situations are classified as critical or non-critical by the first-person vehicle itself. For this purpose, the first-person vehicle can have a computing unit in a driver assistance system. The division by the first-person vehicle itself ensures a quick division.

In a preferred embodiment, the division is determined as a critical or non-critical driving situation depending on a machine vehicle perception by the first-person vehicle. The mechanical vehicle perception is preferably determined by a combination of traffic objects and information from another source, in particular from an internal storage device, an external database, and / or from communication with a traffic unit, such as, for example, a smart or intelligent road vehicle. or infrastructure unit, which comprises at least one or more sensors and a communication unit, preferably, for example, in the form of a digital environment map. This enables a sufficiently precise determination.

Alternatively or optionally in addition, the classification is determined as a critical or non-critical driving situation depending on a comparison of the driver's current driving behavior with a standard behavior. The standard behavior is preferably the normal driving behavior with regard to the driver's vehicle interventions in the same or a comparable traffic situation by himself and / or other reference drivers in the same or comparable traffic situation. This enables a sufficiently precise determination.

Alternatively or optionally in addition, the classification is determined as a critical or non-critical driving situation depending on a relevance measure. The relevance measure preferably comprises at least one, but preferably several, parameters from the group direction of movement of the traffic objects and / or speed of the traffic objects and / or trajectory of the traffic objects, class of the traffic objects, such as cyclists, pedestrians, vehicles, and / or the skills, in particular the acceleration capabilities of the traffic objects, or the capacity of the traffic objects, in particular the capacity of the traffic objects to change their trajectory or the speed, as well as the own movement of the first-person vehicle. This enables a sufficiently precise determination.

The classification as a critical or non-critical driving situation is preferably determined as a function of the relevance measure, the comparison of the driver's current driving behavior with the standard behavior and the machine vehicle perception by the first-person vehicle. This enables a very precise determination.

In a preferred embodiment, the warning signal is designed as an optical warning signal, the optical warning signal being designed to be displayed on a display unit of the first-person vehicle and the at least one critical traffic object, for which a perceptual deficit was identified, is identified on the display unit of the first-aid vehicle using the optical warning signal becomes. The display unit is preferably designed as a head-up display, in particular as an augmented reality head-up display. Using the head-up display, in particular the augmented reality head-up display, traffic objects for which a perception deficit has been recognized can be highlighted separately and in a targeted manner, for example with an identifier under or on the traffic object. This means that only those critical traffic objects that the driver has ignored are given a warning. This shows warnings exactly where they are relevant. By using an augmented reality head-up display, the critical traffic objects concerned can be highlighted with an augmented warning, for example a red border around the critical traffic object. This leads to faster perception by the driver of the first-person vehicle.

In a preferred embodiment, the warning signal is modulated as a function of the evaluation of the gaze direction of the driver of the first-person vehicle in relation to the at least one critical traffic object for which a perceptual deficit has been identified. If the driver's line of sight is still far away from the critical traffic object for which a deficit in perception has been identified, the warning signal can be stronger, for example. This can further increase traffic safety.

In a further preferred embodiment, the warning signal is designed as an acoustic warning signal. The acoustic warning signal is preferably output as a directed sound output which specifies to the driver the direction of the at least one critical traffic object for which a deficit in perception has been identified. Using a directed sound output, the warning can be emitted in a targeted manner “from that direction” in which the critical traffic object for which a deficit in perception was identified lies. This is a faster reaction of the driver to this critical traffic object is possible.

Preferably, the directional sound output becomes quieter and / or lower when the driver's direction of vision of the first person vehicle approaches the at least one critical traffic object, and becomes higher and / or louder when the driver's direction of vision of the first person vehicle is viewed.

A further modulation of the acoustic warning signal preferably indicates when the driver has viewed the corresponding critical traffic object.

The ego vehicle is preferably operated in an autonomous mode of operation by a controller, the control being transferred to the driver as a function of a deficit in perception of the at least one critical traffic object. In the automated driving mode, the detection device monitors the traffic objects around the ego vehicle and the current and future driving scenario. Different autonomy levels are included with autonomy.

If the driver thus observes what is happening, control can be transferred to the driver relatively late and with a simple signal, for example an acoustic, haptic or optical signal, if necessary. If the driver is completely distracted, he is quickly and purposefully prepared for the current situation in order to take over the first-person vehicle, that is, in particular, he is made aware of the critical traffic objects. This enables a targeted and supported transfer of vehicle control to the driver in the event that critical traffic objects are not perceived. A situation-specific adaptation and a time-effective control of the transfer of the vehicle control depending on the driver's perception are thus accomplished.

Furthermore, in the case of a perception deficit, the handover preferably takes place only after a predefined time and / or after the perception deficit has been remedied by the driver of the first-person vehicle.

Depending on the perception deficit, the time and duration for a handover can thus be flexibly adjusted. The time period to get used to traffic can be dynamically adjusted to the driver's perception deficit in order to initiate the handover in good time. This results in a reduction in the number of accidents in automated driving and an increase in road safety.

• - eine Erfassungseinrichtung zum Erfassen von Verkehrsobjekten im Umfeld des Egofahrzeuges,
• - eine Recheneinheit zum Bestimmen für die erfassten Verkehrsobjekte, ob eine kritische oder unkritische Fahrsituation vorliegt,
• - eine Einteilungseinheit zum Einteilen der erfassten Verkehrsobjekte in kritische und unkritische Verkehrsobjekte, anhand der vorangegangenen Bestimmung der kritischen und unkritischen Fahrsituationen,
• - eine Blickrichtungserfassungseinrichtung zum Erfassen der Blickrichtung des Fahrers,
• - eine Feststellungseinheit zur Feststellung einer Wahrnehmung zumindest eines vorhandenen kritischen Verkehrsobjektes durch den Fahrer des Egofahrzeuges, anhand zumindest der erfassten Blickrichtung,
• - ein Warnsystem zur Ausgabe zumindest eines Warnsignals bei einem festgestellten Wahrnehmungsdefizit.

The object is further achieved by specifying a device for controlling the attention of a driver of a first-person vehicle, comprising:

• a detection device for detecting traffic objects in the environment of the first-person vehicle,
• a computing unit for determining for the detected traffic objects whether there is a critical or non-critical driving situation,
• a division unit for dividing the detected traffic objects into critical and uncritical traffic objects, based on the previous determination of the critical and uncritical driving situations,
• a viewing direction detection device for detecting the driver's viewing direction,
• a determination unit for determining a perception of at least one existing critical traffic object by the driver of the first-person vehicle, based on at least the detected direction of vision,
• - a warning system for outputting at least one warning signal in the event of a perceived deficit.

The advantages of the method can also be transferred to the device. The device is particularly suitable for carrying out the method as described above.

Furthermore, the object is achieved by specifying a first-person vehicle which has a device for controlling the attention of a driver as described above. The first-person vehicle is designed in particular as a passenger car or truck.

Furthermore, the object is achieved by specifying a computer program comprising commands which cause the ego vehicle as described above to carry out the method as described above. This means that the process can also be retrofitted in all suitable vehicles.

• 1: ein erfindungsgemäßes Fahrzeug,
• 2: ein Flussdiagramm eines erfindungsgemäßen Verfahrens in einer ersten Ausgestaltung,
• 3: eine Draufsicht auf ein Head-Up-Display,
• 4: ein Flussdiagramm eines erfindungsgemäßen Verfahrens in einer zweiten Ausgestaltung,
• 5: eine beispielhafte Situation in der das erfindungsgemäße Verfahren Verwendung findet.

Further features, properties and advantages of the present invention will become apparent from the following description with reference to the accompanying figures. It shows schematically:

• 1 : a vehicle according to the invention,
• 2nd 1 shows a flow diagram of a method according to the invention in a first embodiment,
• 3rd : a top view of a head-up display,
• 4th 1 shows a flow diagram of a method according to the invention in a second embodiment,
• 5 : an exemplary situation in which the inventive method is used.

Although the invention has been illustrated and described in detail by the preferred exemplary embodiments, the invention is not restricted by the disclosed examples. Variations on this can be derived by those skilled in the art without departing from the scope of the invention as defined by the following claims.

1 shows a first-person vehicle according to the invention 1 for controlling the attention of a driver of the first-person vehicle 1 .

The first-person vehicle 1 has a detection device 2nd , in particular an environment sensor system for detecting traffic objects in the environment of the first-person vehicle 1 on.

Furthermore, the ego vehicle 1 a computing unit 3rd on. With the computing unit 3rd it is determined for the detected traffic objects whether there is a critical driving situation. The classification as a critical driving situation is advantageously dependent on a mechanical perception by the first-person vehicle 1 and a comparison of the driver's current driving behavior with a standard behavior and a relevance measure. This is in 2nd described in more detail.

Furthermore, the ego vehicle 1 a schedule line 4th to divide the detected traffic objects into critical and uncritical traffic objects based on the determination of the critical and uncritical driving situations. The schedule line unit 4th can also in the arithmetic unit 3rd be integrated. Furthermore, a viewing direction detection device is provided for detecting the driver's viewing direction. This is here as a driver camera 5 designed, which can be attached for example in or on the rear view mirror.

The first-person vehicle 1 has a detection unit 6 to determine a perception of the at least one critical traffic object by the driver of the first-person vehicle 1 , based on at least the detected line of sight. The detection unit 6 can also in the arithmetic unit 3rd be integrated.

The first-person vehicle also includes 1 a warning system 7 for outputting at least one warning signal in the event of a perceived deficit. Here, a lack of perception means that it was found that the driver in the first-person vehicle 1 no perception of the at least one critical traffic object has occurred. The perception deficit can also include a personal perception deficit, such as an inability or a restriction of the particular optical and / or acoustic perception, in particular short-sighted, far-sighted or general ametropia. The perception can be designed as a visual perception.

The warning signal is preferably designed as an acoustic or optical warning signal. The warning signal is preferably designed as an acoustic warning signal in combination with an optical warning signal. The first-aid vehicle points out the acoustic warning signal 1 prefers a surround-capable audio system 8th with at least two speakers. The ego vehicle points to the output of the optical warning signal 1 a display unit, which is preferably a head-up display 9 , in particular as an augmented reality head-up display.

The first-person vehicle 1 can also be a driver assistance system 12th in which the computing unit 3rd , the detection device 2nd , the schedule line unit 4th , the detection unit 6 as well as the driver camera 5 and the warning system 7 are integrated.

2nd shows a first embodiment of the method according to the invention. Do it in one step S1 all traffic objects in a surrounding area or a detectable surrounding area of the ego vehicle 1 ( 1 ) detected. The traffic objects can be designed as traffic vehicles, ie trucks, motor vehicles, motorcycles, pedestrians, etc. The surrounding area is determined by means of the detection device 2nd ( 1 ) detected. The detection device 2nd ( 1 ) can be used as a sensor / sensors, for example in a driver assistance system 12th ( 1 ) are arranged, designed or designed as a communication device which transmits the position data / direction of travel of the traffic objects by means of car-to-car or car-to-X communication.

Then in one step S2 determines whether a critical or non-critical driving situation is present in relation to the detected traffic objects. A critical driving situation also includes a relevant driving situation. Based on the detected objects of transport, the ego vehicle 1 ( 1 ) even classified the driving situations as critical or uncritical by using the first-person vehicle 1 ( 1 ) conducts a situation assessment.

The categorization as a critical or non-critical driving situation can depend on the machine's perception of the vehicle Ego vehicle 1 ( 1 ) can be made. To create a mechanical perception of the driving environment, vehicle parameters are determined by the detection device 2nd ( 1 ) of the first-person vehicle 1 ( 1 ) are recorded, used. Furthermore, further information about the static driving environment from digital environment maps, in particular digital navigation maps, is preferably used.

A dynamic vehicle environment model is generated by the machine perception by the ego vehicle 1 ( 1 ) and all other road users are represented by individual movement models. Building on this vehicle environment model, all individual components are then placed in relation to one another to identify the driving situation (s) in order to evaluate the vehicle situation from the dependencies of the individual elements. The evaluation can also contain various possible temporal developments of the current traffic situation, which are evaluated with regard to the probability of occurrence.

The classification as a critical or non-critical driving situation can optionally be carried out additionally or alternatively depending on a comparison of the driver's current driving behavior with a standard behavior. The standard behavior corresponds to the normal behavior with regard to the driver's vehicle interventions in similar situations by himself or other reference drivers in the same or a comparable traffic situation.

The classification as a critical or non-critical driving situation can optionally be carried out in addition or alternatively depending on a relevance measure. The relevance measure includes a direction of movement of the traffic objects and / or the speed of the traffic objects and / or trajectory of the traffic objects and / or the class of the traffic objects, with a class being, for example, a cyclist class, pedestrian class, vehicle class, as well as that Own movement of the first-person vehicle 1 ( 1 ). The relevance measure is defined for each traffic object. In this way it can be recognized, for example, that oncoming traffic during normal driving roughly in the direction of the first-person vehicle at high speed 1 ( 1 ) is on the road, but this does not result in a critical driving situation. However, a critical driving situation arises when the first-person vehicle 1 ( 1 ) a left turn 13 ( 4th ). Correct driver-side perception of oncoming traffic is therefore highly relevant.

The classification as a critical or non-critical driving situation is advantageously dependent on the machine perception by the first-person vehicle 1 ( 1 ) and the comparison of the driver's current driving behavior with a standard behavior and determined by the relevance measure. Taking into account the machine perception, the norm behavior and the relevance measure, a risk assessment can be carried out and thus the critical traffic objects can be determined, which can be used as a basis for a targeted warning. The parameters of machine perception, norm behavior and relevance measure can also be weighted differently. Depending on the situation, this weighting can be predefined in the first-person vehicle 1 ( 1 ) be saved.

The traffic objects are recorded by the first-person vehicle 1 ( 1 ) now classified taking into account the machine perception, the norm behavior and the relevance measure under critical or uncritical driving situations.

Then in one step S3 a division of the detected traffic objects into critical and uncritical traffic objects based on the determination of the critical and uncritical driving situations by a division unit 4th ( 1 ) performed. Traffic objects that are assigned to an uncritical driving situation are non-critical traffic objects. Traffic objects that are assigned to a critical driving situation are critical traffic objects.

If no critical traffic object has been identified (if at least one critical traffic object does not exist), a step is taken S4 carried out and the procedure starts again.

In one step S5 If at least one critical traffic object is present, the driver's line of sight (also called eye tracking) is detected by a line of sight detection device. The eye detection device can be used as a sensor, in particular as a driver camera 5 ( 1 ) be designed.

The viewing direction can be determined, for example, by detecting a pitching movement of the driver's head and calculating the distance between the driver's eyes. However, other methods can also be used to record the viewing direction.

In one step S5 it is determined at least on the basis of the detected viewing direction whether the at least one critical traffic object was perceived.

The determination as to whether the at least one critical traffic object was perceived can preferably be made with inclusion determined by the driver's peripheral field of view. The driver's peripheral field of vision can also be viewed using the viewing direction detection device, here the driver camera 5 ( 1 ) can be determined.

The inclusion of a peripheral field of vision defines a “probability cone” that can provide conclusions as to whether a critical traffic object was perceived.

As an alternative or optionally in addition, the determination of whether the at least one critical traffic object was perceived is determined as a function of the temporal fixation period of the driver's gaze on the at least one critical traffic object. The fixation duration is preferred via the viewing direction detection device, here the driver camera 5 ( 1 ), certainly. A probability of “as seen” can be assumed from the temporal evaluation of the fixation.

In addition, the number of times the driver's gaze hits the critical traffic object and / or the frequency of how often the driver's gaze hits the traffic object within a predetermined time unit can be used to determine whether the at least one critical traffic object was perceived, flow in.

Corresponding driver reactions, such as control interventions on the gas, brakes and steering of the driver of the first-person vehicle, can also be carried out 1 ( 1 ) support the determination of whether a driver appears to have perceived the critical traffic object or not because he is responding to it.

This enables a reliable statement to be made as to whether the driver of the first-person vehicle 1 ( 1 ) not only saw the critical traffic object, but also perceived it.

If a driver's perception of the critical traffic object (s) has been determined, it starts in one step S6 the procedure again.

Regardless of a perception determined by the driver of the first-person vehicle 1 ( 1 ) However, a vehicle-side action can be carried out by the first-person vehicle, if otherwise a collision / accident by the first-person vehicle 1 ( 1 ) would be expected.

If a perception deficit is identified, it is carried out in one step S7 at least one warning signal for the (at least one) critical traffic object for which there is a lack of perception by means of a warning system 7 ( 1 ) issued. Here, a lack of perception means that it was found that the driver in the first-person vehicle 1 ( 1 ) there was no perception of the at least one critical traffic object.

The warning signal can be output as an optical warning signal. This is preferably displayed on a display device of the vehicle. The display device can be a head-up display 9 ( 1 ) be designed.

By using an augmented reality head-up display, the affected critical traffic objects, for which there is a lack of perception, can be highlighted as an augmented, optical warning signal, for example with a red border around the critical traffic object (s) in connection with, preferably, the acoustic warning signal. This creates an adaptive, situation-appropriate warning strategy that can avert unpleasant emergency braking interventions.

In one step S8 the driver's line of sight is detected again using the eye detection device. Becomes a perception of the traffic object classified as critical by the driver of the first-person vehicle 1 ( 1 ) is detected, the warning signal is ended.

In one step S9 the process is repeated at predefined time intervals as long as the driver is traveling with the first-person vehicle 1 ( 1 ) continues.

3rd shows in such a head-up display 9 in detail in the first-person vehicle 1 ( 1 ) with an indicated cockpit 10th .

All traffic objects that pass through the detection device 2nd ( 1 ) have been recognized. The traffic objects recognized as critical and subject to a lack of perception can be highlighted separately and in a targeted manner. This can be done, for example, with a marking, for example with a red border or a displayed “sickle” under the critical traffic object for which there is a lack of perception. The head-up display can do this 9 be designed as augmented reality head-up displays. The critical traffic objects, for which there is a lack of perception, are shown to the driver in his current view in such a way that he has the feeling that they are part of the environment. This can be achieved with an additional ad level.

Alternatively, only those critical traffic objects for which a There is a lack of perception in the head-up display 9 are displayed. In this example, an oncoming traffic vehicle is used as the critical traffic object 11 displayed. The head-up display can also be used here 9 be designed as an augmented reality head-up display.

The at least one critical traffic object, for which there is a lack of perception, can also be shown in a glowing or flashing manner in order to attract the driver's attention.

Alternatively or optionally in addition, the warning signal can be output as an acoustic warning signal.

The acoustic warning signal informs the driver that a relevant object has not been viewed in his field of vision. Through continuous recording and evaluation of the driver's line of sight, a corresponding modulation of the acoustic warning signal indicates whether the driver's line of sight is approaching or moving away from the critically classified traffic object for which there is a lack of perception. The acoustic warning signal becomes quieter, for example, when the viewing direction approaches the traffic object classified as critical, for which there is a lack of perception, and louder at a distance. In the case of a pulsed acoustic warning signal, the distance between two pulses can also serve as information for the driver.

The acoustic warning signal can preferably be output as a directed warning tone.

Using this directional sound output, the acoustic warning signal can be emitted in a targeted manner “from the direction” of the critical traffic object for which there is a lack of perception. This has an advantage especially in complex traffic situations with multiple traffic objects. With such a directed sound output, the acoustic warning signal comes from the driver's point of view from the direction of the critical traffic object for which there is a lack of perception.

In addition or optionally in addition, the directional sound output can become higher (higher pitch), for example, when the viewing direction approaches the traffic object classified as critical, for which there is a lack of perception, and lower (lower pitch) when it is removed. As soon as the driver has viewed the corresponding traffic object, this is indicated by a further modulation of the warning tone. If the driver reacts accordingly after viewing the traffic object recognized as critical, which he had previously overlooked, for example by reducing the speed, the warning signal is preferably switched off.

The first-person vehicle advantageously has 1 ( 1 ) to output the acoustic warning signal via a surround-capable audio system 8th ( 1 ) with at least two speakers.

4th shows the method in a second embodiment. Here are the steps A1 to A4 identical to the steps S1 to S4 .

In the step A5 , which is also identical to S5, it is determined at least on the basis of the detected viewing direction whether the at least one existing critical traffic object has been perceived.

Was a driver-side perception of the critical traffic object (s) by the driver of the first-person vehicle 1 ( 1 ) is determined, and if a warning signal is already being issued, then in one step A9 the warning signal is switched off and the procedure in one step A6 started again. If a warning signal is not yet issued, the procedure is - as in 2nd described- in the step A6 started again.

If a perception deficit is identified, it is carried out in one step A7 the warning signal for the critical traffic object for which there is a lack of perception by means of the warning system 7 ( 1 ) issued.

In one step A8 starts the process at A5 again with the detection of the line of sight.

The process is continued or restarted as long as the driver is driving the first-person vehicle 1 ( 1 ) continues.

5 shows an exemplary situation in which the inventive method is used. It is based on an ego vehicle that turns left 13 the method according to the invention briefly explained by way of example.

As described above, are detected by a detector 2nd ( 1 ) of the first-person vehicle 1 the objects in the vicinity, i.e. traffic vehicles, traffic cyclists, traffic pedestrians, etc. are recorded. Sensors (environment sensors such as radar and lidar) can be used for this. The various vehicles can also - if available - via car-to-car communication with the first-person vehicle 1 communicate to transfer the relevant data of the traffic vehicles. Different traffic units or other traffic objects can also - if available - Car-to-X communication with the first-person vehicle 1 for the transmission of relevant data, in particular data relating to the environment of the traffic units or the data relating to the environment or the movement of the other traffic objects.

The computing unit 3rd ( 1 ), for example in a driver assistance system 12th ( 1 ) of the first-person vehicle 1 is present, determines for the detected traffic objects whether there is a critical driving situation.

The classification as a critical driving situation is advantageously dependent on the machine perception by the first-person vehicle 1 and / or the comparison of the driver's current driving behavior with a standard behavior and determined by the relevance measure. This determination is made by the first-person vehicle 1 himself. Based on the determination of the critical and uncritical driving situations, the detected traffic objects are then divided into critical and uncritical traffic objects.

At least here is the oncoming traffic vehicle 11 classified as a critical traffic object for which there is a critical driving situation. For this example it is assumed that this is the only critical traffic object.

Then the driver's line of sight of the first-person vehicle 1 by a line of sight detection device. Through a locking unit 6 ( 1 ) is determined based on the detected line of sight whether the oncoming traffic vehicle 11 was perceived. This is done taking into account the peripheral field of vision of the driver of the first-person vehicle 1 and depending on the duration of the fixation of the driver's gaze on the oncoming traffic vehicle 11 certainly.

If a perception deficit is determined, that is, the driver of the first-person vehicle has 1 the oncoming traffic vehicle 11 not perceived, a warning signal is issued. The warning signal can be designed acoustically and (at the same time) optically as described above.

The warning signal is switched off when it has been determined that the driver is the oncoming traffic vehicle 11 perceived.

Furthermore, the first-person vehicle can be operated in an autonomous mode by a controller. Control is now transferred to the driver as a function of a lack of perception of the at least one critical traffic object. If the driver observes what is happening and is informed about the current traffic situation, the control can be transferred to the driver relatively late and, for example, with a simple signal, for example an acoustic, haptic or optical signal, if necessary. If the driver is completely distracted, he must be prepared quickly and in a targeted manner for the current situation in order to take over the first-person vehicle, which means that he is particularly made aware of the critical traffic objects. The control can then be transferred.

QUOTES INCLUDE IN THE DESCRIPTION

This list of documents listed by the applicant has been generated automatically and is only included 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 102009005688 A1 [0004]
• DE 19734307 A1 [0005]
• GB 2534163 B [0006]

Claims (21)

Method for controlling the attention of a driver of an ego vehicle (1), characterized by the steps: - detection of traffic objects in the environment of the ego vehicle (1) by a detection device (2), - determining for the detected traffic objects whether a critical or non-critical driving situation is present, - Division of the detected traffic objects into critical and non-critical traffic objects, based on the previous determination of the critical and uncritical driving situations, - detection of the driver's line of sight by a line of sight detection device when at least one critical traffic object is present, - determining, at least based on the detected line of sight, whether that at least an existing critical traffic object was perceived by the driver of the first-person vehicle (1), - outputting at least one warning signal in the event of a perceived deficit. Method for controlling the attention of a driver of an ego vehicle (1) Claim 1 , characterized in that the determination of whether the at least one critical traffic object was perceived is determined as a function of a peripheral field of vision of the driver and / or the temporal fixation period of the driver's gaze on the at least one critical traffic object. Method for controlling the attention of a driver of an ego vehicle (1) Claim 1 or 2nd , characterized in that the determination of whether the at least one critical traffic object was perceived is determined as a function of at least one corresponding driver reaction of the driver of the first-person vehicle (1). Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the driving situations of the first-person vehicle (1) are classified as critical or non-critical. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the division is determined as a critical or non-critical driving situation depending on a machine vehicle perception by the first-person vehicle (1). Method for controlling the attention of a driver of an ego vehicle (1) Claim 5 , characterized in that the mechanical vehicle perception is determined by a combination of traffic objects detected by the detection device (2) and information from another source. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the classification as a critical or non-critical driving situation is determined as a function of a comparison of the driver's current driving behavior with a standard behavior. Method for controlling the attention of a driver of an ego vehicle (1) Claim 7 , characterized in that the norm behavior is the normal driving behavior with regard to the vehicle interventions of the driver in the same or a comparable traffic situation by himself and / or other reference drivers in the same or a comparable traffic situation. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the classification is determined as a critical or non-critical driving situation depending on a relevance measure. Method for controlling the attention of a driver of an ego vehicle (1) Claim 9 , characterized in that the relevance measure comprises at least one parameter from the group direction of movement of the traffic objects and / or speed of the traffic objects and / or trajectory of the traffic objects and / or class of the traffic objects as well as the own movement of the first-person vehicle (1). Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the warning signal is designed as an optical warning signal, the optical warning signal being designed to be displayed on a display unit of the first-person vehicle (1), and at least that An existing critical traffic object for which a perceptual deficit has been identified is identified on the display unit of the first-person vehicle (1) using the optical warning signal. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the warning signal as a function of the evaluation of the viewing direction of the driver of the first-person vehicle (1) in relation to the at least one critical traffic object for which a Perception deficit was recognized, is modulated. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the warning signal is designed as an acoustic warning signal. Method for controlling the attention of a driver of an ego vehicle (1) Claim 13 , characterized in that the acoustic warning signal is output as a directed sound output which specifies to the driver the direction of the at least one critical traffic object for which a deficit in perception has been identified. Method for controlling the attention of a driver of an ego vehicle (1) Claim 14 , characterized in that the directional sound output when the driver of the first-person vehicle (1) approaches the at least one critical traffic object is quieter and / or lower and when the driver's line of vision of the first-vehicle (1) approaches the at least one critical traffic object higher and / or gets louder. Method for controlling the attention of a driver of a first-person vehicle (1) according to one of the preceding claims, characterized in that the first-person vehicle (1) is operated in an autonomous mode by a controller, and a transfer of control to the driver as a function of a perception deficit of the at least one existing critical traffic object. Method for controlling the attention of a driver of an ego vehicle (1) Claim 16 , characterized in that the handover in the event of a perception deficit takes place after a predefined time and / or after the perception deficit has been remedied by the driver of the first-person vehicle (1). A device for controlling the attention of a driver of a first-person vehicle (1), comprising: - A detection device (2) for detecting traffic objects in the environment of the first-person vehicle (1), - a computing unit (3) for determining for the detected traffic objects whether there is a critical or non-critical driving situation, - a division unit (4) for dividing the detected traffic objects into critical and non-critical traffic objects, based on the previous determination of the critical and non-critical driving situations, a viewing direction detection device for detecting the driver's viewing direction, - a detection unit (6) for determining a perception of at least one critical traffic object by the driver of the first-person vehicle (1), based on at least the detected line of sight, - A warning system (7) for outputting at least one warning signal in the event of a perceived deficit. Device for controlling the attention of a driver of an ego vehicle (1) Claim 18 to perform a method according to one of the Claims 1 to 17th . First-person vehicle (1), which has a device for controlling the attention of a driver according to one of the Claims 18 or 19th having. Computer program comprising commands that cause the ego vehicle (1) of the Claim 20 , the method according to one of the preceding Claims 1 to 17th executes.

Images