Method for assessing the driving safety of a driver of a vehicle

Abstract

The invention relates to a method for assessing the driving safety (FB) of a driver of a vehicle (1), the driver of the vehicle (1) being identified as a person in the vehicle (1). According to the invention: - individual physiological and driving-specific characteristic values of the driver as well as media usage characteristic values of the driver are recorded, processed and stored over a predefinable period of time during a driving operation; - a driving safety value (FS) is determined on the basis of the recorded characteristic values; and - a recommendation or a warning is issued in the vehicle (1) depending on the determined driving safety value (FS).

Description

[0001] Procedure for assessing the driving safety of a vehicle driver

[0002] The invention relates to a method for assessing the driving safety of a driver of a vehicle, wherein the driver of the vehicle is identified as a person in the vehicle.

[0003] DE 102012 014 191 A1 discloses a method for determining the usage behavior of a motor vehicle system. The motor vehicle system offers a function usable by the driver. The method provides for sending corresponding data about the use of the function to a central computer via a communication device arranged in the motor vehicle.

[0004] DE 102021 003489 A1 discloses a method for detecting a gradually decreasing driving performance of a vehicle driver, in which an individual driving-specific behavior as well as physiological and emotional characteristics of a driver and his health data are recorded, which are combined and evaluated to determine the driving performance.

[0005] US 2016 0 046298 A1 discloses a system including a motion sensor that can essentially create a real-time map of the driver's movement while driving. Based on this data, an event that leads to a driver's distraction can be detected.

[0006] US 2021 0 398 562 A1 deals with a computer-implemented method for replaying media content, whereby the user's mental state is recorded using sensor data.

[0007] US 2022 0242 233 A1 deals with a procedure and a

[0008] Device for detecting dangerous vehicle events and encouraging the use of features that prevent the occurrence of dangerous events. US 2007 0275 770 A1 deals with a hands-free device that is attached to the sun visor of a vehicle and communicates with a remote hands-free device.

[0009] The invention is based on the object of specifying a method for assessing the driving safety of a driver of a vehicle.

[0010] The object is achieved according to the invention by a method which has the features specified in claim 1.

[0011] Advantageous embodiments of the invention are the subject of the subclaims.

[0012] A method for assessing the driving safety of a driver of a vehicle, wherein the driver of the vehicle is identified as a person in the vehicle, provides according to the invention that individual physiological and driving-specific characteristic values ​​of the driver as well as media usage characteristic values ​​of the driver are recorded, processed, and stored over a predeterminable period of time during ferry operation. Furthermore, a driving safety value is determined based on the recorded characteristic values, and depending on the determined driving safety value, a recommendation or warning is issued in the vehicle. Furthermore, driving safety is mapped as a function of vehicle and media usage characteristic values. In particular, an individual driving safety of the driver can be mapped based on the vehicle and media usage characteristic values.Based on a performed function approximation, the vehicle and media usage feature values ​​that influence the driving safety value, especially negatively, are identified. For example, methods that can identify the features that most strongly influence function values, especially for driving safety assessment, can be applied to an individual function approximation.

[0013] By applying the method, the risk of driver distraction, particularly with regard to media use, can be reduced, thus reducing the risk of an accident for the vehicle and, where applicable, for road users in the immediate vicinity of the vehicle. In particular, the risk of an accident can be significantly reduced by detecting individual vehicle and media use that increases a personal risk of being involved in and / or causing a traffic accident, since media use in the vehicle, in particular, is a potential source of distraction. Thus, by applying the method, the driver is given the opportunity to adapt their behavior and thus reduce the risk of an accident.

[0014] To this end, one embodiment of the method provides for the determination of the driver's physiological characteristic values ​​based on the driver's heart rate, respiratory rate, and skin conductance. Based on the physiological characteristic values, it can be determined, in particular, whether the driver is relatively relaxed or in a stressful situation. Particularly in stressful situations, the risk of an accident can increase because the driver, for example, reacts hectically and / or tends to overlook other road users.

[0015] In a further embodiment, the driver's driving-specific characteristic values ​​are determined based on vehicle acceleration, wheel speeds, engine speeds, driver assistance interventions, the vehicle's lateral position in its lane, and / or distances to vehicles ahead. Based on these driving-specific characteristic values, a connection can be established between driver distraction and the driver's driving style caused by the distraction. If the distance to a vehicle ahead decreases and / or the vehicle crosses a lane marking, it can be assumed that the driver is distracted, particularly by certain media usage, for example, by operating a mobile device.

[0016] In a possible refinement of the method, media usage characteristic values ​​are determined based on the operation of a mobile device, the operation of the vehicle's infotainment system, and / or the volume settings of media playback in the vehicle. For example, a driver of a vehicle is more distracted from the driving situation than another driver when the media playback volume is comparatively high, so that the driving safety assessments of the two drivers differ despite the same conditions. Driver media usage is analyzed in training using computer-assisted speech analysis methods.In particular, media usage can be analyzed by means of computer-assisted speech analysis with regard to content and the moods generated in a driver based on this content in order to assess the driver's driving safety while driving the vehicle and, under certain circumstances, to alert the driver to his or her distraction due to this media usage.

[0017] For this purpose, one embodiment of the method can provide for the determination of the driver's emotions based on recorded facial, speech, and / or physical signals. Based on the detected driver emotions, a relatively reliable determination of whether the driver is distracted by media use is possible.

[0018] Furthermore, in one embodiment, the method provides that the driver is recommended an alternative media usage in the vehicle based on a driving safety value predicted at least based on the media usage signals. For this purpose, the driver's media usage is analyzed in real time, whereby the value for the driving safety assessment is determined. Depending on the driver's determined physiological values, the driver is recommended to consume, i.e., listen to, relaxing media, for example, in the form of music.

[0019] Furthermore, in a possible embodiment, the method provides that the individual physiological and driving-specific characteristic values ​​of the driver as well as the media usage characteristic values ​​of the driver recorded over the predefined period of time during ferry operation are transmitted to a central computer unit that is data-linked to the vehicle.

[0020] Alternatively or in addition to processing and storing the recorded feature values ​​in the vehicle, these can be transmitted to the central computer unit for aggregation, processing, and storage. Vehicles in a fleet transmit the recorded feature values ​​to the central computer unit in order to identify the feature values, in particular to determine factors relevant to the vehicle fleet, for example, media usage and its respective distraction. Exemplary embodiments of the invention are explained in more detail below with reference to drawings.

[0021] Showing:

[0022] Fig. 1 schematically shows a device for carrying out a method for assessing the driving safety of a driver of a vehicle and

[0023] Fig. 2 shows a schematic one-dimensional diagram of driving safety dependent on media use.

[0024] Corresponding parts are provided with the same reference numerals in all figures.

[0025] Figure 1 shows a vehicle 1 and a device for carrying out a method for the driving safety assessment FB of a driver of the vehicle 1 (not shown in detail), whereas Figure 2 shows a diagram D of a driving safety FS dependent on a media usage MN.

[0026] It is generally known that driver distraction during vehicle operation is the cause of a comparatively large proportion of traffic accidents. Media use (MN), i.e., the use of a mobile device, in particular the operation of a smartphone, the setting of podcasts, music, and / or the operation of an infotainment system of the vehicle 1, is a possible source of driver distraction.

[0027] The method described below helps the driver of vehicle 1 to identify which individual media usage MN increases their personal risk of being involved in and / or causing a traffic accident. In addition, the method gives the driver the opportunity to adapt their behavior and thus reduce the risk of an accident. The device has a control unit 2, a further control unit 3, for example of an assistance system, at least one interior microphone 4, at least one vital sensor 5, at least one interior camera 6, at least one exterior camera 7 and a central computer unit 8, to which the vehicle 1, in particular the control unit 2, is data-linked. The vehicle 1 belongs to a vehicle fleet to which other vehicles (not shown) belong, which are also data-linked to the central computer unit 8.

[0028] In particular, signals detected by the interior microphone 4, the vital sensor 5, the interior camera 6, the exterior camera 7, and the additional control unit 3 are aggregated and recorded in the control unit 2. In particular, interactions between the driver and the vehicle 1 are recorded, i.e., stored, in the control unit 2.

[0029] The driver of vehicle 1 can be identified using a variety of methods, for example, using captured image signals from the interior camera 6 and / or the exterior camera 7, and known methods, for example, using computer-aided vision, e.g., in relation to facial recognition, using a personalized vehicle key, and / or by entering a personal identification number on the vehicle side. In particular, the driver is identified as a person, so that captured signals and the values ​​and data captured based on these signals can be clearly assigned to this person. Personal information and characteristics of the driver, such as age and / or whether the driver is a smoker, can be stored or will be stored in a personal driver profile.

[0030] It is thus possible to determine individual physiological and driving-specific characteristic values / data in vehicle 1 based on recorded signals from an above-mentioned sensor system installed in vehicle 1 and to process these in control unit 2.

[0031] For example, physiological characteristic values ​​(Adata), such as a heart rate, a respiratory rate, a skin conductance, etc., can be determined based on detected signals from at least one vital sensor 5. Alternatively or additionally, these characteristic values ​​(Adata) can be determined based on detected signals from optical and / or electronic sensors in the vehicle 1 or in the vehicle seat. Driving-specific, in particular driving-dynamic characteristic values ​​(Adata) relating to the driver can be determined based on detected signals from acceleration sensors of the vehicle 1 and other sensors, based on whose detected signals, for example, wheel speeds and / or engine speeds, can be determined. Based on the detected and processed signals, lateral acceleration, longitudinal acceleration, speed distribution, maximum engine speed, and engine speed distribution can also be determined.

[0032] Driver-specific feature values ​​(A data) can also be extracted based on available information from the additional control unit 3. For example, a number of driver assistance interventions, such as initiated emergency braking, steering wheel interventions to maintain a lane, as well as information about a lateral positioning of the vehicle 1 in its lane and / or information about distances to preceding vehicles selected by the driver of the vehicle 1, can be used.

[0033] Furthermore, it is possible to detect the driver's emotions, for example, based on recorded facial, speech, and / or physical signals using a corresponding sensor system present in the vehicle 1, in particular the interior camera 6. This allows, for example, the driver's physiological stress reaction to be detected.

[0034] Characteristic values ​​and data relating to an individual driver-specific driving style or behavior can be determined using a variety of methods. For example, a classification of driving behavior can be performed based on recorded sensor signals.

[0035] It is also conceivable that emotional feedback from other road users in the vicinity of vehicle 1 can be used to evaluate the individual driving style or driving behavior of the driver of vehicle 1.

[0036] The method for assessing the driving safety of the driver of vehicle 1 provides that, in vehicle 1, characteristic values ​​(A data) relating to individual physiological, emotional, driver-specific driving dynamics, and driving behavior-specific information of the driver are recorded and aggregated in the vehicle's control unit 2 over a predeterminable, comparatively long period of time. For each driver of vehicle 1, a driving value for the driving safety assessment FB, i.e., a driving performance characteristic, is determined from a driver-specific feature space. This value for the driving safety assessment FB simultaneously characterizes the driver's driving style and characterizes or evaluates the safety of the driver's driving style.

[0037] Furthermore, individual vehicle and, in particular, media usage (MN) is determined based on signals recorded by the vehicle's sensors. Vehicle usage includes turning on the air conditioning, opening a sunroof, etc., and, in particular, media usage characteristic values ​​are determined.

[0038] Media usage includes the operation of a driver's mobile device, i.e. a smartphone, the setting of podcasts, music and / or the operation of an infotainment system of vehicle 1, the selected volume of media playback, etc., which are also recorded using signals from the sensors installed in vehicle 1.

[0039] To further differentiate media usage MN, it can be analyzed using methods of computer-assisted language analysis, for example with regard to content, sentiment or similar.

[0040] In a further embodiment of the method, additional personal physiological and driver-specific feature valuesAdata can be used to expand the above-mentioned feature space.

[0041] Based on vehicle usage characteristic values ​​and media usage characteristic values, an individual map of the driver's driving safety assessment FB can be created as a function of the vehicle usage characteristic values ​​and the media usage characteristic values.

[0042] For this purpose, a heuristic function approximation can be used, for example, a linear mapping of the media usage feature values ​​to a target value of the driving safety rating FB. Alternatively or additionally, function approximations can be learned or trained using machine learning methods in vehicle 1. For this purpose, the vehicle and media usage features are used as the input variable and the driving safety rating FB as the target value. Machine learning algorithms then determine the function approximation that can best map the feature values ​​to the corresponding target values.

[0043] Based on the function approximations, it is possible to identify the vehicle and media usage characteristics that negatively influence the individual driving safety assessment (FB). For example, methods that can identify the characteristics that most strongly influence function values, so-called feature analysis, can be applied to the individual function approximation.

[0044] In the event of individual vehicle and, in particular, media usage MN that negatively impacts driving safety FS during operation of vehicle 1, an acoustic, visual, and / or haptic warning can be issued to the driver in vehicle 1. For this purpose, current individual vehicle and media usage MN is analyzed in real time and mapped as a driving safety value FS using function approximation. If it is determined that the driving safety value FS falls below a specified threshold, the optical, acoustic, and / or haptic warning is issued in vehicle 1.

[0045] In a further embodiment of the method, alternatively or in addition to issuing the warning, options for action can be suggested to the driver which essentially do not have a negative impact on driving safety FS, but may even positively increase it.

[0046] Furthermore, the method provides that individual images and / or the underlying feature values ​​are transmitted by wireless communication K from the vehicle 1 to the central computer unit 8, which can be a so-called cloud or a data center of a vehicle manufacturer. There, the feature values ​​are aggregated and processed in order to identify the feature values ​​that are relevant for the vehicle 1 and the other vehicles in the vehicle fleet, in particular for driving safety FS when using media MN. In one exemplary embodiment, a driver is driving a comparatively high-quality luxury vehicle as vehicle 1 and switches on the infotainment system of the vehicle 1. The driver selects a podcast to play in the vehicle 1. Based on the speech analysis performed with regard to the podcast, it is recognized that it is a current stock market report from a specific company.

[0047] Based on the function approximation, a significantly worse assessment of driving safety FS is predicted for the feature values ​​"stock market report of the specific company", "increased resting heart rate" and "increased skin conductance", whereupon the driver is suggested a different podcast or relaxing music as an alternative.

[0048] In the embodiment shown in Figure 2, a diagram D is shown with a driving safety rating FB assigned to each driver when using media MN.

[0049] For a driver whose driving safety rating FB is shown by a solid line, a maximum driving safety rating FS is achieved with media usage MN with a value of zero, i.e., no media usage MN during ferry operation of vehicle 1. This maximum value is represented by a star S1.

[0050] For another driver, whose driving safety rating FB is shown by a dashed line, the driving safety value FS increases with increasing media usage value MN.

[0051] Media usage MN, in particular its characteristic value, can, for example, be the set volume of a media playback in vehicle 1.

[0052] For the driver with the driving safety rating FB shown by the solid line, simply switching on the infotainment system can lead to a decrease in the driving safety value FS. Such a circumstance can lead to an increase in the driving safety value FS for the other driver as the volume increases, leading to an increase in the driving safety FS during ferry operation of vehicle 1, as shown by an example of another star S2.

Claims

Patent claims Method for the driving safety assessment (FB) of a driver of a vehicle (1), wherein the driver of the vehicle (1) is identified as a person in the vehicle (1), characterized in that - over a predefined period of time during ferry operation, individual physiological and driving-specific characteristic values ​​of the driver as well as media usage characteristic values ​​of the driver are recorded, processed and stored, - a driving safety value (FS) is determined based on the recorded characteristic values ​​and - depending on the determined driving safety value (FS), a recommendation or a warning is issued in the vehicle (1), - where the driver's driving safety is mapped as a function of vehicle and media usage characteristic values, - wherein, based on a performed function approximation, the vehicle and media usage feature values ​​that influence the driving safety value (FS) are identified. Method according to claim 1, characterized in that physiological feature values ​​of the driver are determined based on a heart rate, a respiratory rate, and a skin conductance value of the driver. Method according to claim 1 or 2, characterized in that driving-specific feature values ​​of the driver are determined based on accelerations of the vehicle (1), wheel speeds, engine speeds, driver assistance interventions, lateral positions of the vehicle (1) in its lane and / or distances to vehicles ahead are determined. Method according to one of the preceding claims, characterized in that media usage feature values ​​of the driver are determined based on operations of a mobile terminal, operations of an infotainment system of the vehicle (1) and / or set volumes of media playback in the vehicle (1). Method according to one of the preceding claims, characterized in that media usage (MN) of the driver is analyzed using methods of computer-assisted speech analysis. Method according to one of the preceding claims, characterized in that The driver's emotions are determined based on recorded facial, speech, and / or physical signals. Method according to one of the preceding claims, characterized in that alternative media usage in the vehicle (1) is recommended to the driver based on a driving safety value predicted at least based on the media usage signals. Method according to one of the preceding claims, characterized in that the individual physiological and driving-specific characteristic values ​​of the driver, as well as the media usage characteristic values ​​of the driver, recorded over the predeterminable period of time during ferry operation, are transmitted to a central computer unit (8) data-linked to the vehicle (1).

Images