METHOD FOR CLASSIFYING A DRIVER'S NON-DRIVING ACTIVITY WITH REGARD TO AN INTERRUPTIBILITY OF THE NON-DRIVING ACTIVITY WHEN THE DRIVING TASK IS REQUESTED TO TAKE OVER, AND METHOD FOR RE-RELEASING A NON-DRIVING ACTIVITY AFTER AN INTERRUPTION OF THE NON-DRIVING ACTIVITY DUE TO A REQUEST TO TAKE OVER THE DRIVING TASK

Description

State of the art

[0001] The invention is based on a device or method according to the class of the independent claims. The present invention also relates to a computer program.

[0002] To ensure the safety of automated and / or highly automated vehicles, vehicle interior sensors, such as vehicle occupant monitoring cameras, can be used. These monitoring cameras can, for example, detect the head and / or gaze direction of a vehicle occupant, track their hand gestures, and analyze their posture.

[0003] The document US 9 690 292 B1 discloses devices, systems and methods for the transition between autonomous and manual operating modes of vehicles.

[0004] The document WO 2017 / 118789 A1 discloses a method for controlling an automatic driver assistance system of a motor vehicle. Disclosure of the invention

[0005] Against this background, the approach presented here presents a method for classifying a driver's activity unrelated to the driving task with regard to the interruptibility of the activity unrelated to the driving task when the driver is requested to take over the driving task, a method for re-enabling an activity unrelated to the driving task after the activity unrelated to the driving task has been interrupted due to a request to take over the driving task, a device using this method, and finally a corresponding computer program according to the main claims. The measures listed in the dependent claims enable advantageous refinements and improvements of the device specified in the independent claim.

[0006] The approach presented here is based on the fact that the performance of a non-driving activity by a vehicle driver can be classified according to the interruptibility level of that activity in order to calculate the potential risk of driver distraction due to the non-driving activity. Furthermore, the approach presented here can be used to re-enable the non-driving activity after it has been interrupted, for example, due to a request to take over the driving task.

[0007] A method is presented for classifying a driver's non-driving activity with regard to the interruptibility of the non-driving activity when the driver is requested to take over the driving task, the method comprising the following steps: Reading in a detection signal and a first position signal from a vehicle occupant monitoring device of a vehicle, wherein the detection signal represents a detected object that serves an activity of the driver unrelated to the driving task, and wherein the first position signal represents a first position of the detected object; receiving a takeover request signal, wherein the takeover request signal represents a request from the driver to take over the driving task; determining a second position of the detected object using a second position signal, wherein the second position signal represents a final position of the object before the driver takes over the driving task and interrupts the activity unrelated to the driving task;and determining an interruptibility level with regard to an interruptibility of the non-driving activity, wherein the interruptibility level represents a degree of visual and / or cognitive capture of the driver by the non-driving activity. ;

[0008] An activity unrelated to driving tasks can be an activity that is carried out by a driver of a vehicle in addition to steering the vehicle and / or operating the vehicle's own devices. For example, an activity unrelated to driving tasks can implicitly include eating, drinking, or smoking. Furthermore, an activity unrelated to driving tasks can particularly involve the driver's engagement with modern information and communication technologies, such as mobile phones, which can distract the driver, for example through social interactions and / or playful activities. Interruptibility can be understood as a characteristic of an activity unrelated to driving tasks that indicates that the activity unrelated to driving tasks has not yet been completed and will not be continued temporarily or until a predetermined point in time.A driving task can involve operating the steering, drive, brakes, and, depending on the equipment, other components of a vehicle. Even today, the driver of a vehicle can be supported in carrying out the driving task by numerous driver assistance systems. In the future, even more extensive vehicle automation, up to and including full automation, can be expected. In automated and / or highly automated driving, the driver can be requested to take over the driving task, which can be particularly the case when the automated and / or highly automated vehicle is confronted with a traffic situation that cannot be resolved by the system itself. A vehicle occupant monitoring device can be an optical sensor device for monitoring the occupancy of a vehicle.For example, a vehicle occupant monitoring device can consist of cameras, usually infrared cameras, and / or 3D sensors, which can specifically monitor a vehicle driver and thus detect the driver's attention, thereby increasing road safety. An object can be understood, for example, as an item with which the driver of a vehicle can perform an activity unrelated to driving tasks. For example, an object for performing activities unrelated to driving tasks can be, in particular, a mobile phone.

[0009] Given the ongoing technical developments and innovations in the field of automated and / or autonomous driving, passengers in such automated and / or autonomous vehicles will increasingly engage in activities unrelated to their driving tasks. In automation levels two, three, and four, i.e., from partially automated driving to highly automated driving, the driver could be prompted by the vehicle to assume responsibility for driving in critical traffic situations. Various studies have revealed a tendency for drivers to resume interrupted non-driving activities after successfully assuming the driving task. This demonstrates that most drivers are at least cognitively, but often also visually, absorbed by the non-driving activity, which can negatively impact the quality of manual driving and thus overall road safety.

[0010] Numerous studies focus on the driver's ability to safely and comfortably assume control of the vehicle within a defined time window, depending on the type of non-driving activity. Performing non-driving activities while in automated driving mode has shown that these activities impact the driver mentally, cognitively, and visually, even after the driver has successfully assumed control of the vehicle. This can lead to a reduced level of attention during manual driving and thus an increased risk of accidents.

[0011] The advantages of the approach presented here therefore lie in particular in supporting a driver in concentrating on manual driving after a request to take over the driving task by taking into account non-driving task-related activities and controlling them accordingly. One goal of the approach presented here can be, in particular, to first identify the type of non-driving task-related activity that was interrupted by the driver before the takeover request and then to enable the driver to efficiently and safely resume their interrupted activities later. Furthermore, using the approach presented here, the probability of the driver being distracted by the non-driving task-related activity during manual driving of the vehicle, after a request to take over the driving task, can be determined by determining an interruptibility level of the non-driving task-related activity.

[0012] According to one embodiment, in the step of determining a final position of the object before the driver takes over the driving task and interrupts the non-driving task-related activity, the final position of the object can define a region of interest, wherein the step of reading in and / or the step of determining is carried out using the defined region of interest. Such an embodiment of the approach presented here offers the advantage that classification of a potential source of distraction during manual driving of the vehicle can be facilitated by defining a region of interest with respect to the non-driving task-related activity or the source of distraction that was determined before the request to take over the driving task.To detect and identify activities unrelated to the driving task, computer vision and / or machine learning methods can be applied to the sensor data collected by the vehicle occupant monitoring system. Various tasks can be trained by experts, for example, and learned by the system.

[0013] According to a further embodiment, a first gaze direction signal can also be read in during the reading step, wherein the first gaze direction signal represents a first gaze and / or head direction of the driver, wherein in the determining step, a second gaze and / or head direction of the driver in relation to the defined area of ​​interest is determined using a second gaze direction signal, wherein the second gaze direction signal represents a distraction of the driver by the object in the defined area of ​​interest. Such an embodiment of the approach presented here offers the advantage that the gaze direction of the driver and / or the head orientation of the driver can be used as a predictor of the driver's distraction.The distraction classification is based on determining the driver's gaze and / or head direction relative to the defined area of ​​interest, which is either relevant or not relevant for safe driving.

[0014] Furthermore, according to one embodiment, in the reading step, the detection signal and / or the first position signal and / or the first viewing direction signal can be read in by a camera unit, wherein in the determining step, the second position signal and / or the second viewing direction signal is also read in by the camera unit. Such an embodiment of the approach presented here offers the advantage that, using at least one or a plurality of camera units, or even 3D sensor devices, activities unrelated to the driving task can be detected and classified into different interruptibility levels.

[0015] According to the invention, in the step of determining the interruptibility level, the interruptibility level is determined using a mathematical model. In a further embodiment of the method, the interruptibility level can define an interruptibility of the non-driving task activity from an easy interruptibility level, which represents easy interruptibility of the non-driving task activity, to a difficult interruptibility level, which represents difficult interruptibility of the non-driving task activity. For example, if the driver is playing a game on their mobile phone and pauses the current game before putting the mobile phone aside and taking over the driving task, this interrupted non-driving task activity is considered easily interruptible because the driver can resume the game at a later time without losing any progress made.However, if, for example, the driver was texting on their mobile phone when the driver was asked to take over the driving task, they may not have enough time to properly end the conversation. In this case, the task is considered difficult to interrupt, and the driver is expected to be distracted, at least cognitively, but possibly also visually.

[0016] The invention thus offers the advantage that, using the determined interruptibility level of an activity unrelated to the driving task, a probability of the driver being distracted by the activity unrelated to the driving task is calculated.

[0017] Furthermore, the method according to the invention comprises a calculation step in which a probability of the driver being distracted by the classified non-driving activity is calculated. The probability is calculated using a prior probability based on the determined interruptibility level of the non-driving activity before the driver assumed the driving task, as well as an analysis of the driver's gaze and / or head direction after the driver assumed the driving task. The invention offers the advantage that the interruptibility level of the non-driving activity is used to assess, using a prior probability, the extent to which the driver will be distracted after the driver is requested to assume the driving task.

[0018] According to a further embodiment, the reading step and / or the receiving step and / or the determining step and / or the determining step and / or the calculating step can be carried out on a computing unit external to the vehicle and / or on a computing unit installed in the vehicle, in particular wherein the reading step and / or the determining step is carried out repeatedly. A computing unit external to the vehicle can, for example, be a cloud on which data from various servers is collected and can be retrieved online from any location. Such an embodiment of the approach presented here offers the advantage that processing data in the computing unit external to the vehicle means lower computing requirements in the vehicle itself and thus enables lower energy consumption or the possibility of using resources for other functions.In addition, the vehicle-external processing unit has greater available computing power than an in-vehicle computer. Such an embodiment of the approach presented here further offers the advantage that, upon repeated execution of the reading and / or determining steps, the classification of an activity unrelated to the driving task can be verified for plausibility and, in addition, the probability of driver distraction due to the classified non-driving activity can be calculated.

[0019] According to one embodiment, the probability of driver distraction due to the classified non-driving activity can be calculated between the repeatedly executed steps of reading and / or determining. Such an embodiment of the approach presented here offers the advantage that the accuracy of a classification of the non-driving activity can be improved by using a prior probability of driver distraction and the calculated probability of driver distraction.

[0020] A method for re-enabling an activity unrelated to the driving task after an interruption of the activity unrelated to the driving task due to a request to take over the driving task is presented, wherein the method comprises the following steps: the steps according to a method for classifying a driver's non-driving activity with regard to the interruptibility of the non-driving activity upon a request to take over the driving task; and outputting an information signal, wherein the information signal represents information regarding a temporal and / or local re-enabling of the non-driving activity.

[0021] According to one embodiment, the outputting step can further enable re-enabling an activity unrelated to the driving task while the vehicle is manually driving in an alternative execution mode using the information signal. To prevent the driver from being distracted by an activity unrelated to the driving task that is considered difficult to interrupt after the takeover request, the system can thus offer the driver information about the next available opportunity to re-enabling the activity unrelated to the driving task. Such information could, for example, be displayed on a display device of the vehicle.Depending on the interruptibility level of the non-driving activity, the system can also offer the option of re-enabling the non-driving activity during manual driving using an alternative execution mode, such as acoustically instead of visually. For example, writing a text message is permitted in SAE levels three and four, as the driver is guaranteed a specific time window to safely and comfortably take over the driving task. However, during manual driving following the takeover request, the driver is not permitted to write a text message, as text creation is a visual activity.If the vehicle is equipped with voice recognition and is connected to the driver's mobile phone, for example, via Bluetooth, the system can offer the driver the option of informing the person they were texting before the takeover request and continuing to write the text message by dictating the text message using voice recognition. Such an embodiment of the approach presented here thus offers the advantage that the driver can concentrate on manual driving while taking into account the non-driving activity, since the re-enabling of the non-driving activity can be controlled accordingly.

[0022] One or more of the methods presented here can be implemented, for example, in software or hardware or in a mixed form of software and hardware, for example in a device or a control unit.

[0023] The approach presented here further provides a device configured to perform, control, or implement the steps of one of the methods presented here in corresponding devices. This embodiment of the invention in the form of a device also allows the problem underlying the invention to be solved quickly and efficiently.

[0024] For this purpose, the device can have at least one computing unit for processing signals or data, at least one memory unit for storing signals or data, at least one interface to a sensor or actuator for reading sensor signals from the sensor or for outputting data or control signals to the actuator, and / or at least one communication interface for reading or outputting data embedded in a communication protocol. The computing unit can be, for example, a signal processor, a microcontroller, or the like, wherein the memory unit can be a flash memory, an EEPROM, or a magnetic storage unit.The communication interface can be designed to read in or output data wirelessly and / or via a wired connection, wherein a communication interface that can read in or output wired data can read this data, for example, electrically or optically from a corresponding data transmission line or output it to a corresponding data transmission line.

[0025] In this case, a device can be understood as an electrical device that processes sensor signals and outputs control and / or data signals depending on them. The device can have an interface that can be implemented in hardware and / or software. In a hardware implementation, the interfaces can, for example, be part of a so-called system ASIC, which contains a wide variety of functions of the device. However, it is also possible for the interfaces to be separate integrated circuits or to consist at least partially of discrete components. In a software implementation, the interfaces can be software modules that are present, for example, on a microcontroller alongside other software modules.

[0026] Also advantageous is a computer program product or computer program with program code that can be stored on a machine-readable carrier or storage medium such as a semiconductor memory, a hard disk memory or an optical memory and is used to carry out, implement and / or control the steps of the method according to one of the embodiments described above, in particular when the program product or program is executed on a computer or a device.

[0027] Examples of the approach presented here are shown in the drawings and explained in more detail in the following description. It shows: Fig. 1 a block diagram of a device for classifying a driver's activity unrelated to the driving task with regard to the interruptibility of the activity unrelated to the driving task upon a request to take over the driving task, as well as a device for re-enabling an activity unrelated to the driving task after an interruption of the activity unrelated to the driving task due to a request to take over the driving task, according to an embodiment; Fig. 2 a schematic view of an interior situation of a vehicle for using a method for classifying a driver's non-driving task activity with regard to an interruptibility of the non-driving task activity when a request to take over the driving task is made according to an embodiment; Fig. 3 a flowchart of an embodiment of a method for classifying a driver's non-driving task activity with regard to the interruptibility of the non-driving task activity upon a request to take over the driving task, as well as a subsequent method for re-enabling a non-driving task activity after an interruption of the non-driving task activity due to a request to take over the driving task according to an embodiment; and Fig. 4 a flowchart of an embodiment of a method for classifying a driver's non-driving task activity with regard to the interruptibility of the non-driving task activity upon a request to take over the driving task, and a subsequent method for re-enabling a non-driving task activity after an interruption of the non-driving task activity due to a request to take over the driving task according to an embodiment.

[0028] In the following description of advantageous embodiments of the present invention, the same or similar reference numerals are used for the elements shown in the various figures and having a similar effect, whereby a repeated description of these elements is omitted.

[0029] Fig. 1 shows a block diagram of a device 100 for classifying a driver's activity unrelated to the driving task with regard to the interruptibility of the activity unrelated to the driving task upon a request to take over the driving task, as well as a device 102 for re-enabling an activity unrelated to the driving task after an interruption of the activity unrelated to the driving task due to a request to take over the driving task, according to one exemplary embodiment. According to one exemplary embodiment, the devices 100 and 102 are arranged on a vehicle 104, wherein the vehicle 104 is, for example, a highly automated vehicle 104. Additionally or alternatively, the device 100 and / or the device 102 can also be arranged on a vehicle-external computing unit.

[0030] According to one exemplary embodiment, the vehicle 104 has at least one vehicle occupant monitoring device 106 for optical-sensory monitoring of an occupancy of the vehicle 104, wherein the vehicle occupant monitoring device 106 is, for example, a camera unit 106 that can be arranged in an interior of the vehicle 104.

[0031] According to one exemplary embodiment, the device 100 comprises a reading unit 108, a receiving unit 110, a detection unit 112, a determination unit 114, and a calculation unit 116. The reading unit 108 is configured, for example, to read in a detection signal 118 and a first position signal 120 from the vehicle occupant monitoring device 106 of the vehicle 104, wherein the detection signal 118 represents a detected object that serves an activity of the driver unrelated to the driving task, and wherein the first position signal 120 represents a first position of the detected object. The reading unit 108 is further configured, for example, to read in a first gaze direction signal 122 from the vehicle occupant monitoring device 106, wherein the first gaze direction signal 122 represents a first gaze and / or head direction of the driver.According to one embodiment, the reading unit 108 is designed to repeatedly read in the detection signal 118, the first position signal 120 and the first viewing direction signal 122.

[0032] The receiving unit 110 is configured, for example, to receive a takeover request signal 124 during an automated and / or highly automated driving process of the vehicle 104. The takeover request signal 124 represents a request from the driver to take over the driving task. A takeover request for the driving task can occur, for example, when the highly automated vehicle 104 is confronted with a traffic situation that cannot be resolved by the system itself, whereby the driver should be able to take over the driving task in a timely and comfortable manner.

[0033] In response to the received takeover request signal 124, the determination unit 112 is configured, according to one embodiment, to receive a second position of the detected object using a second position signal 126 from the vehicle occupant monitoring device 106, wherein the second position signal 126 represents a final position of the object before the driver takes over the driving task and the non-driving activity is thus interrupted. Here, the determination unit 112 is further configured, by way of example, to define an area of ​​interest using the final position of the object before the driver takes over the driving task and interrupts the non-driving activity.Furthermore, the determination unit 112 is configured to determine a second gaze and / or head direction of the driver relative to the defined area of ​​interest using a second gaze direction signal 128 from the vehicle occupant monitoring device 106, wherein the second gaze direction signal 128 represents a distraction of the driver by the object in the defined area of ​​interest. According to one embodiment, the determination unit 112 is configured to repeatedly determine the second position signal 126 and the second gaze direction signal 128.

[0034] The determination unit 114 is therefore designed according to one embodiment to determine an interruptibility level 132 with regard to an interruptibility of the non-driving task activity using a classification signal 130 provided by the determination unit 112, which represents the classified non-driving task activity, wherein the interruptibility level 132 represents a degree of visual and / or cognitive capture of the driver by the non-driving task activity.Here, the interruptibility level 132 is determined by way of example using a mathematical model, wherein the interruptibility level 132 defines an interruptibility of the activity non-related to the driving task, which ranges from a slight interruptibility level, which represents an easy interruptibility of the activity non-related to the driving task, to a severe interruptibility level, which represents a severe interruptibility of the activity non-related to the driving task.

[0035] According to the invention ee, the calculation unit 116 is designed to calculate a probability of the driver being distracted by the classified non-driving task activity, wherein the probability is calculated using a prior probability based on the determined interruptibility level 132 of the non-driving task activity before the driver assumed the driving task, as well as an analysis of the driver's gaze and / or head direction after the driver has assumed the driving task, by means of an analysis signal 134 provided by the determination unit 112.Here, the prior probability based on the determined interruptibility level 132 of the non-driving activity before the driver assumed the driving task can be denoted by the formula symbol P(DT ), and the analysis of the driver's gaze and / or head direction after the driver assumed the driving task can be denoted by the formula symbol P(b|DT ). The following formula thus results, for example, in the calculation of the probability of the driver being distracted by the classified non-driving activity: P D b ≈ P b D T P D T

[0036] Here, P the probability D the likelihood of driver distraction, D T the driver’s distraction by the activity unrelated to the driving task and b the observed driver behavior, in particular the driver's gaze and / or head direction.

[0037] According to one exemplary embodiment, the device 102 has an output unit 136, wherein the output unit 136 is configured, for example, to output an information signal 138, wherein the information signal 138 represents information about a temporal and / or local re-enabling of the activity unrelated to the driving task. According to one exemplary embodiment, the output unit 136 is further configured, using the information signal 138, to enable re-enabling of an activity unrelated to the driving task already during manual driving of the vehicle 104 in an alternative execution mode, for example, acoustically instead of visually.

[0038] Fig. 2 shows a schematic view of an interior situation of a vehicle for using a method for classifying a driver's non-driving task activity with regard to an interruptibility of the non-driving task activity when a request to take over the driving task is made according to an embodiment.

[0039] Here, the first 205 and the second 210 views each show a top view of a vehicle occupant 215, for example a truck driver 215, holding an object 220 in his hand, from the perspective of a camera unit arranged, for example, on the ceiling of the truck cab. The first 205 and the second 210 views were recorded by the camera unit at two different times, for example, with the second view 210 being recorded at a later time than the first view 205, for example, after receiving a request to take over the driving task.

[0040] According to one embodiment, the left side of views 205 and 210 each shows a confidence level of the pixel depth estimate, and the right side of views 205 and 210 each shows the corresponding infrared images from the same perspective. In the first view 205, using the camera unit, the object 220 is detected at a first position 225 in the hand of the driver 215 and classified as a mobile phone 220 before the takeover request for the driving task is received. The camera unit then tracks the object 220 to its final position 230, where the driver 215 puts down the mobile phone 220 to take over control of the steering wheel.

[0041] The final position 230 defines a region of interest 230, which can now be considered to estimate the probability that the driver 215 is distracted by a non-driving task related to the mobile phone 220 that affects the region of interest 230 by evaluating the gaze and / or head direction of the driver 215 relative to this region of interest 230. For example, if the driver 215 frequently looks in the direction of the region of interest 230, the system classifies the driver 215 as distracted by the object 220.

[0042] Fig. 3 shows a flowchart 300 of an embodiment of a method for classifying a driver's non-driving task activity with regard to the interruptibility of the non-driving task activity upon a request to take over the driving task, as well as a subsequent method for re-enabling a non-driving task activity after an interruption of the non-driving task activity due to a request to take over the driving task according to an embodiment.

[0043] According to one embodiment, the flowchart 300 first has a data output process 310, wherein the data relating to a detected object, a position of the object and / or a direction of gaze and / or head of the driver are captured by a camera unit arranged in an interior of the vehicle and used for optical-sensory detection of the interior and the occupancy of the vehicle, and are sent to a device for classifying an activity of a driver that is unrelated to the driving task with regard to an interruptibility of the activity that is unrelated to the driving task when a request to take over the driving task is made.

[0044] In a subsequent process step 320, the object detected by the camera unit, as well as the position of the object, are further tracked by the camera unit using optical and sensory means. In parallel, in a process step 330, the driver's gaze and / or head direction is further tracked by the camera unit using optical and sensory means.

[0045] Using the determined positions of the object and / or the determined gaze and / or head directions of the driver, according to the invention, an activity of the driver that is unrelated to the driving task is classified with great certainty in a process step 340.

[0046] The starting point 350 represents a request from the driver to take over the driving task. This can be the case, for example, when the automated and / or highly automated vehicle is confronted with a traffic situation that cannot be resolved by the system itself, whereby the driver should be able to take over the driving task in a timely and comfortable manner.

[0047] Subsequently, in a process step 360, a region of interest is defined at this final position using a final position of the object before the driver assumes the driving task and interrupts the non-driving activity. Using the defined region of interest, process step 330 is repeatedly executed, and the driver's gaze and / or head direction is tracked in relation to the defined region of interest, whereby a distraction of the driver by the object in the defined region of interest is now detected.

[0048] In a decision process 370, based on an interruptibility level of the non-driving activity, the interruptibility level being determined using a mathematical model, a decision is made as to whether the non-driving activity is easily interruptible or difficult to interrupt. For example, a game that can be paused is classified as easily interruptible, whereas writing a text message is classified as difficult to interruptible, since the driver is often cognitively and / or visually absorbed by the non-driving activity.

[0049] If the activity unrelated to the driving task can be easily interrupted, the driver can either be informed in a process step 380 about the re-enabling of the activity unrelated to the driving task at a later time, or a process step 390 can follow in which no action is carried out due to the invalidity of the activity unrelated to the driving task.

[0050] If the activity unrelated to the driving task is difficult to interrupt, the driver can be offered the option in a process step 399 to re-enable the activity unrelated to the driving task in an alternative execution mode.

[0051] Fig. 4 shows a flowchart of an embodiment of a method 400 for classifying a driver's non-driving task activity with regard to the interruptibility of the non-driving task activity upon a request to take over the driving task, as well as a subsequent method 460 for re-enabling a non-driving task activity after an interruption of the non-driving task activity due to a request to take over the driving task according to an embodiment. Here, the methods 400 and 460 can be implemented using the Fig. 1 presented device for classifying a driver's activity unrelated to the driving task with regard to the interruptibility of the activity unrelated to the driving task when the driver is requested to take over the driving task, as well as the device for re-enabling an activity unrelated to the driving task after an interruption of the activity unrelated to the driving task due to a request to take over the driving task.

[0052] The method 400 initially comprises a step 410 in which a detection signal and a first position signal are read in from a vehicle occupant monitoring device of a vehicle, wherein the detection signal represents a detected object that serves an activity of the driver unrelated to the driving task, and wherein the first position signal represents a first position of the detected object. In a subsequent step 420, a takeover request signal is received, wherein the takeover request signal represents a request from the driver to take over the driving task. In response to step 420, the method 400 comprises a step 430 in which a second position of the detected object is determined using a second position signal, wherein the second position signal represents a final position of the object before the driver takes over the driving task and interrupts the activity unrelated to the driving task.In a step 440 of the method 400, an interruptibility level is determined with regard to an interruptibility of the non-driving activity, wherein the interruptibility level represents a degree of visual and / or cognitive preoccupation of the driver by the non-driving activity. Finally, the method comprises a step 450 in which a probability of driver distraction by the classified non-driving activity is calculated, wherein the probability is calculated using a prior probability based on a determined interruptibility level of the non-driving activity before the driver assumed the driving task, as well as an analysis of the driver's gaze and / or head direction after the driver assumed the driving task.

[0053] According to one embodiment, in particular step 410 and / or step 430 is repeatedly executed.

[0054] According to one embodiment, method 400 is followed by method 460, which also includes steps 410, 420, 430, 440, and 450. Finally, method 460 includes a step 465 in which an information signal is output, the information signal representing information regarding a temporal and / or local re-enabling of the activity unrelated to the driving task.

[0055] If an embodiment includes an "and / or" link between a first feature and a second feature, this should be read as meaning that the embodiment according to one embodiment has both the first feature and the second feature and according to another embodiment has either only the first feature or only the second feature.

Claims

1. Method (400) for classifying a non-driving activity of a driver (215) with regard to interruptibility of the non-driving activity in the event of a request to take over the driving task during automated and / or highly automated driving, wherein the method (400) has the following steps of: reading in (410) a detection signal (118) and a first position signal (120) from a vehicle occupant monitoring device (106) of a vehicle (104), wherein the detection signal (118) represents a detected object (220) which serves a non-driving activity of the driver (215), and wherein the first position signal (120) represents a first position (225) of the detected object (220); receiving (420) a take-over request signal (124), wherein the take-over request signal (124) represents a request for the driver (215) to take over the driving task; in response to the receiving step (420), ascertaining (430) a second position of the detected object (220) using a second position signal (126), wherein the second position signal (126) represents a final position (230) of the object (220) and defines an area of interest (230) before the driver (215) takes over the driving task and interrupts the non-driving activity, and wherein the non-driving activity of the driver (215) is classified using the positions (225, 230) of the object (220) and / or determined viewing and / or head directions of the driver (215) in the area of interest (230); and after the ascertaining step (430), determining (440) a level of interruptibility (132) with regard to interruptibility of the non-driving activity, wherein the level of interruptibility (132) is determined using a mathematical model, wherein the level of interruptibility (132) represents a degree of visual and / or cognitive appropriation of the driver (215) by the non-driving activity, characterized in that a step (450) of calculating a probability of the driver (215) being distracted by the classified non-driving activity is provided, wherein the probability is calculated using a previous probability based on the determined level of interruptibility (132) of the non-driving activity before the driver (215) took over the driving task and an analysis of the viewing and / or head direction of the driver (215) after the driver (215) has taken over the driving task.

2. Method (400) according to Claim 1, in which, in the reading-in step (410), the detection signal (118) and / or the first position signal (120) and / or the first viewing direction signal (122) is / are read in from a camera unit (106), wherein, in the ascertaining step (430), the second position signal (126) and / or the second viewing direction signal (128) is / are also read in from the camera unit (106).

3. Method (400) according to one of the preceding claims, in which the level of interruptibility (132) defines interruptibility of the non-driving activity from a slight degree of interruption, which represents slight interruptibility of the non-driving activity, to a severe degree of interruptibility, which represents severe interruptibility of the non-driving activity.

4. Method according to Claim 3, in which, in the determining step (440), a game which can be paused, as a non-driving activity, is classified as a slight degree of interruption and / or writing of a text message, as a non-driving activity, is classified as a severe degree of interruptibility.

5. Method (400) according to one of the preceding claims, in which the reading-in step (410) and / or the receiving step (420) and / or the ascertaining step (430) and / or the determining step (440) and / or the calculating step (450) is / are performed on a computing unit outside the vehicle and / or on a computing unit installed in the vehicle (104).

6. Method (400) according to Claim 5, wherein the reading-in step (410) and / or the ascertaining step (430) is / are performed repeatedly, wherein the probability of the driver (215) being distracted by the classified non-driving activity is calculated between the repeatedly performed reading-in step (410) and / or the repeatedly performed ascertaining step (430).

7. Method (460) for re-enabling a non-driving activity after an interruption of the non-driving activity on account of a request to take over the driving task, wherein the method (460) has the following steps: the steps (410, 420, 430, 440, 450) of a method (400) according to one of Claims 1 to 8; and outputting (465) an information signal (138), wherein the information signal (138) represents information relating to temporal and / or local re-enabling of the non-driving activity.

8. Method (460) according to Claim 7, in which, in the outputting step (465), it is also possible to re-enable a non-driving activity already during the manual journey of the vehicle (104) in an alternative execution mode using the information signal (138).

9. Device (100, 102) which is configured to perform and / or control the steps (410, 420, 430, 440, 450, 465) of one of the methods (400, 460) according to Claims 1 to 7 in corresponding units (108, 110, 112, 114, 116, 136).

10. Computer program which is configured to perform and / or control the steps (410, 420, 430, 440, 450, 465) of one of the methods (400, 460) according to Claims 1 to 7 in corresponding units (108, 110, 112, 114, 116, 136).

11. Machine-readable storage medium on which the computer program according to Claim 10 is stored.