Method, control unit and vehicle for preventing and combating fatigue in a driver

Abstract

The invention relates to a method, a control unit and a vehicle (10) for preventing and combating fatigue in a driver (12), in which, based on a generic and / or personalized baseline, existing and / or emerging fatigue of the driver (12) is detected during the journey and measures to reduce and / or combat it are applied, wherein, prior to the journey, a route with regard to break times and / or monotonous route guidance as well as recommendations regarding eating and sleeping habits are transmitted to the driver (12).

Description

[0001] The invention relates to a method, a control unit and a vehicle for preventing and combating fatigue in a driver, in which, based on a generic and / or personalized baseline, existing and / or emerging fatigue of the driver is detected during the journey and measures to reduce and / or combat it are applied, wherein, prior to the journey, a route with regard to break times and / or monotonous route guidance as well as recommendations regarding eating and sleeping habits are transmitted to the driver.

[0002] Driver fatigue is a safety-relevant problem because it slows reaction times, impairs judgment, and thus significantly reduces driving ability. Symptoms of fatigue can appear during a journey, but some drivers continue to operate their vehicles even when already fatigued. This can lead to accidents due to impaired reaction time or microsleep. Driving under the influence of fatigue can therefore be a serious criminal offense, considered a negligent cause of an accident. The number of unreported cases is likely high.

[0003] Methods and devices are known in the art that detect a driver's fatigue and indicate the need for a break by means of a break recommendation. However, these warnings are often ignored. Furthermore, fatigue detection is currently limited to a few indicators and evaluation methods.

[0004] US patent 2021 / 0129748A1 describes a method that uses machine learning to provide an evaluation algorithm for detecting signs of fatigue and then recommending or providing measures to reduce them. These algorithms can also be personalized. However, they still only offer responses to the onset of fatigue.

[0005] It would be desirable, however, to optimize an upcoming journey in advance, based on knowledge about driver fatigue during or after driving, in such a way as to delay or prevent the onset of fatigue. This is one of the objectives of the present invention.

[0006] The object of the invention is achieved by a method, a control unit, and a vehicle according to the independent claims. Further preferred embodiments of the invention result from the remaining features mentioned in the dependent claims.

[0007] A method according to the invention for preventing and combating driver fatigue comprises at least the following steps: • Detection and identification of a vehicle driver, • If the driver is unknown: Start an application that detects emerging and / or existing driver fatigue based on a generic baseline; otherwise: Start an application that detects emerging and / or existing driver fatigue based on a personalized and / or generic baseline. • continuous monitoring of the driver during the journey, • Upon detection of existing and / or emerging fatigue: initiating measures to reduce and / or combat it, and is characterized by the fact that, in order to prevent and / or delay the onset of driver fatigue in the long term, a journey is planned and optimized in advance, based on the personalized baseline and / or the generic baseline. • the route is optimized with regard to the necessity, timing and / or duration of breaks and / or the occurrence of monotonous sections of the route, and / or • Recommendations regarding nutrition and / or sleep habits before the journey will be given to the driver.

[0008] The procedure is designed to prevent driver fatigue and to combat it when it occurs. The procedure is carried out in a vehicle or in conjunction with a vehicle, as will be explained in more detail below. The term "driver" will occasionally be used synonymously with "operator" in the following text.

[0009] The detection and management of fatigue should take place while driving. For this purpose, the driver should be detected and identified. This detection also includes driver identification. This can be done in a known manner using a vehicle camera and subsequent image analysis, but also using a personalized vehicle key, a mobile device carried by the driver, and / or wearables.

[0010] If the driver is unknown, meaning no data is available about them, an application is launched that detects emerging and / or existing driver fatigue based on a generic baseline. Such an application refers to software or a program provided to implement the procedure, containing instructions or program code whose execution by a processing unit triggers the execution of the procedure. The detection or non-detection of the driver can be understood as the trigger for launching the application. Alternatively, manual launch of the application can be provided.

[0011] The application is designed to detect signs of driver fatigue and provide suitable measures to reduce or combat it. For unknown drivers, a generic baseline is used. A generic baseline is a dataset containing signs of fatigue and measures to combat them, incorporating research findings and / or (anonymized) customer data. It thus represents an average of the individuals in the research and / or customer groups who provided their data. This data is processed in such a way that it allows for a reliable assessment of emerging and / or existing driver fatigue. The generic baseline can also include data reflecting the effectiveness of applied measures against driver fatigue.A generic baseline can also be divided into subgroups, for example, groups of people with one or more common characteristics. Grouping can be based on factors such as gender, age, place of residence, or similar criteria. Such subdivision and application of a suitable subgroup of the generic baseline based on the knowledge or identification of the driver should be included in the use of the generic baseline and does not constitute personalization. However, a generic baseline with subgroups can form the basis for personalizing a baseline.

[0012] If, however, the driver of the vehicle is known, a personalized and / or a generic baseline can serve as the basis for fatigue detection. The generic baseline corresponds to the one described above. A personalized baseline additionally or instead of generic data includes personal data for fatigue detection, countermeasures, and optionally their effectiveness, which were recorded and analyzed during previous journeys. A personalized baseline is preferably developed and updated based on a previously generic baseline. In the method according to the invention, either a personalized baseline, a generic baseline, or a combination of both can be used. This can be done automatically or according to the driver's specifications. The application corresponds to the one described above; it simply uses a different, an extended, or the same database.

[0013] The generic and / or personalized baseline can, of course, only be supplemented with customer data or personal data if the driver or the vehicle manufacturer's customers consent to the transfer and use of the data. Providing customer data for the personalized and / or generic baseline can, for example, supplement it by providing additional signs of fatigue, the frequency of known signs of fatigue, or combinations of signs of fatigue, as well as the effect of measures taken to combat fatigue.

[0014] The driver is continuously monitored during the journey and analyzed for signs of fatigue based on a personalized baseline, if available, and / or a generic baseline. This can be done using an evaluation algorithm that simply checks for the presence of fatigue signs based on the generic and / or personalized baseline. Preferably, however, a model has been trained using the personalized and / or generic baseline—that is, a machine-learned model has been created and is provided in the application to detect and evaluate the emerging signs of fatigue. Continuous monitoring can be performed continuously or at predefined intervals. Monitoring is carried out using sensors in the vehicle interior, as will be explained further with regard to optional configurations.

[0015] If the presence of existing fatigue is detected, for example, because the driver entered the vehicle already tired, and / or the onset of driver fatigue is recognized, measures to reduce and / or combat the fatigue are triggered so that they can act upon the driver. Such measures can include, for example, a warning message or an adjustment of the vehicle's interior temperature. These measures are explained in more detail in connection with an optional embodiment of the method according to the invention. In any case, they include actions that alert the driver to their state of fatigue and / or act upon them in such a way that the fatigue is reduced and / or eliminated. The measures themselves and the information on when each measure is appropriate are stored in the generic and / or personalized baseline.

[0016] The method is characterized by its ability to slow down or prevent the onset of fatigue, and preferably its continued existence, before the journey begins, thus avoiding or delaying it in the long term. This is achieved by planning and optimizing the journey in advance, based on a generic and / or personalized baseline.

[0017] A driver provides information about a planned trip, for example, via a vehicle-linked application on their mobile device or through a user interface in the vehicle. If the fatigue detection application has access to the driver's calendar, such advance planning can also be done using calendar entries. Depending on the planning time, various measures, including longer-term ones, can be suggested to prevent and / or delay the onset of fatigue. This should incorporate data from the personalized and / or generic baseline, from which factors influencing the development of fatigue can be derived.

[0018] The route itself can first be planned and optimized to minimize the number of monotonous sections and, if necessary, to choose a more attractive route, as it is known that uniform sections can be tiring. If such sections are unavoidable or no reasonable alternative route can be found, they can be interrupted with a break.

[0019] Alternatively or additionally, the need for breaks, their number, duration, and / or timing (i.e., the "when") can be planned based on one or both baselines. For example, the baselines can be used to determine after what driving time a driver typically becomes tired—or, in the case of a personalized baseline, specifically when the driver has usually become tired. Based on this, it can then be determined how many breaks the journey should be divided into, and preferably when and / or how long these breaks should be.

[0020] Planning and optimizing a journey can also include recommendations for nutrition and / or sleep habits before the trip. The aim is to provide drivers with information on how to improve their fitness to drive by getting enough sleep before setting off and / or avoiding foods that negatively affect their driving ability and, optionally, their sleep before the journey. Foods that promote good driving ability can also be recommended. For example, it is known that caffeine and alcohol can impair sleep, potentially causing the driver to tire more quickly while driving. Similarly, "heavy" meals, especially fatty and difficult-to-digest foods, eaten before bed reduce sleep quality and can lead to premature fatigue during the journey if consumed immediately beforehand.This and other information can be provided based on the generic baseline, and, if a personalized baseline exists and contains relevant information (generic and / or personalized), also based on that. Such recommendations can be particularly effective if the trip is planned well in advance to allow for their implementation. The recommendations can be delivered via the same method used for planning.

[0021] Once the driver is in their vehicle and ready to begin the planned journey, the previously generated planning data is retrieved and made available for use. For example, the route can be transmitted to the vehicle's navigation system and driver monitoring can be initiated.

[0022] As previously explained, the driver is continuously monitored for signs of fatigue during the journey, and countermeasures are initiated as needed. This is done within the framework of the inventive method, taking into account the planned and optimized journey. For example, if a break is scheduled and signs of fatigue appear early on, the timing of the break can be adjusted, or measures to reduce fatigue can be initiated in order to implement a (later) planned break.

[0023] Planning and optimizing a journey with regard to route and / or breaks, and / or providing recommendations for pre-journey behavior, makes it possible to delay or even combat the onset of fatigue during a trip. This can improve or extend the driver's fitness to drive, at least with regard to fatigue symptoms. Since fatigue is frequently a problem on longer journeys, the proposed method can counteract the onset of fatigue both before and during the journey.

[0024] A first embodiment of the method according to the invention consists in the driver actively transmitting feedback about their state of fatigue while driving. If the driver notices that they are becoming tired while driving, they can transmit this information via a user interface, according to this embodiment. This can be a textual input, an input via voice control, and / or a "report button" on the user interface. In any case, the vehicle user actively communicates any emerging and / or existing fatigue.

[0025] As a result, measures can be directly triggered to alleviate or combat fatigue. Furthermore, this information can be used to expand the generic and / or personalized baseline or to verify existing data. Personally perceived fatigue is generally associated with objectively recognized signs of fatigue, which are / were already recorded within the framework of the inventive method. Through the active reporting of fatigue by the driver, the perceived fatigue can be compared with the recognized signs of fatigue, thereby confirming the association of fatigue symptoms with the driver's fatigue level, both generically and / or personally. Optionally, the transmission of the driver's perceived fatigue can also include a degree of fatigue.

[0026] The detection of emerging and / or existing fatigue while driving can be achieved using at least one camera, at least one thermal imaging camera (also known as a FIR camera), radar, at least one microphone, and / or a wearable device belonging to the driver. This detection is therefore passive, relying on sensors inside the vehicle.

[0027] The driver's face can be observed using one or more cameras, and the captured images can be analyzed with regard to eye opening, blink rate, drooping eyes, eye rubbing, and / or increased yawning. This can be done using known image analysis methods, the results of which are incorporated into the inventive method. A thermal imaging camera can detect a drop in body temperature, which also indicates fatigue. A camera can be, in particular, an RGB camera, a NIR camera, a ToF camera (Time of Flight), or a stereo camera.

[0028] Using radar and / or at least one microphone, breathing sounds and thus the respiratory rate can be recorded and then analyzed for signs of fatigue. Depending on the sensitivity of the aforementioned recording devices, the heart rate can also be monitored.

[0029] A wearable device belonging to the driver can also be used for the method. A wearable device is a portable device for recording bodily data, such as a suitably designed watch (smartwatch), a ring, a bracelet, or the like. If the transmission of the data as a (further) input variable into the method according to the invention is enabled, data regarding body temperature, respiration, and / or heart rate can be provided.

[0030] To detect signs of fatigue, the driver's behavior can also be analyzed, for example, their reaction time. This reaction time analysis can be compared with information gathered and evaluated by the vehicle's environmental sensors, such as the driver's behavior at traffic lights, crosswalks, or when overtaking. A comparison between the driver's gaze direction and the surroundings captured by external cameras can also be included.

[0031] According to a further embodiment of the procedure, the remaining time until the onset of critical fatigue is determined during the journey. Critical fatigue is defined as a state in which a break is unavoidable, i.e., when the number of signs of fatigue is significantly high and thus, among other things, the risk of microsleep is present, such that measures other than a sufficiently long break are no longer effective in reducing fatigue.

[0032] The onset of critical fatigue is estimated based on the generic and, where available, preferably the personalized baseline, taking into account existing signs of fatigue. This assessment is continuously adjusted based on the fatigue indicators recorded during the journey and with regard to the planned route. For example, the demands placed on the driver's abilities can also be considered, i.e., whether they can still drive this route (safely) in their current state and with further fatigue. If necessary, an alternative, less demanding route can be provided to reduce the driver's workload. Planned breaks can also be adjusted accordingly and moved to an earlier time.

[0033] The aim of this design is to assess whether and when critical fatigue occurs during the journey and, if necessary, to adjust the planning and / or to begin measures to alleviate or combat fatigue early (in good time).

[0034] A particularly good application of the method according to the invention can be achieved if the generic and / or the personalized baseline is provided by means of a machine-learned model and is adapted using data collected during the journey on the occurrence and / or existence of driver fatigue.

[0035] It has already been indicated that the baseline(s) can be used to train an artificial neural network model. The machine-learned model can then be integrated into the application for detecting emerging and / or existing driver fatigue, where it can be used to analyze incoming monitoring data for signs of fatigue.

[0036] The machine-learned model can be further improved and optimized by incorporating additional data. With the consent of the vehicle users, data on detected signs of fatigue, countermeasures taken, and, if applicable, the success (or failure) achieved can be transmitted to the application provider during or after the journey. Anonymized data can be used in a generic model for creating or updating the generic baseline. Personal, i.e., non-anonymized, data can be used in a personalized model for the personalized baseline, as well as in the aforementioned generic model. In each case, the data is used for (further) learning of the model(s). The machine learning is preferably performed by the provider, where it can also be monitored if necessary.The updated model can then be made available as part of an application update. This way, no resources are tied up in the vehicle to adjust the baseline(s).

[0037] A further embodiment of the method according to the invention exists when, in order to reduce and / or combat fatigue, taking into account the planned journey and the fatigue level of the driver, at least one of the following measures is triggered: • Adjusting the temperature in the vehicle or at least in the vicinity of the driver, • Adjusting the volume when playing media, • Provision of an activating scent, • Opening at least one window and / or sunroof of the vehicle, • Activation of a daylight lamp, • Activation of a massage function in the driver's vehicle seat, • Recommendation to take a break, • Extended recommendation for a break with information on nearby rest areas, hotels, parking lots and the like, and / or • Recommendation for adjusted caffeine consumption.

[0038] If emerging (or existing) fatigue is detected during a journey, it should be reduced and / or combated. The planned journey and the driver's existing state of fatigue should be taken into account. For example, the planned journey can be considered by checking the remaining time until a scheduled break and whether this time can be bridged using one or more of the aforementioned measures, while observing safety-relevant aspects, or whether the break should be brought forward instead. This is particularly possible in conjunction with the aforementioned approach of determining the remaining time until the onset of critical fatigue.

[0039] The driver's current state of fatigue is assessed based on the number of fatigue symptoms and their frequency during the current journey. This assessment can be made, particularly within a predefined timeframe, such as 30 minutes or one hour. Therefore, some of the aforementioned measures are more effective and thus more suitable for reducing fatigue than others.

[0040] The aforementioned measures are explained below; however, the order does not imply any indication of their effectiveness or prioritization. The selection of one or more of the measures depends on the driver's level of fatigue, their signs of fatigue, and, where applicable, their personalized baseline.

[0041] A first step would be to adjust the temperature in the vehicle, or at least in the driver's immediate surroundings. This primarily aims to lower the temperature. At lower temperatures, the human body reacts with increased energy consumption to maintain core temperature. This boosts metabolism, which has an activating effect. Furthermore, when the ambient temperature is lowered, less blood flows to the extremities to retain and protect the internal organs. This allows more blood, and therefore oxygen, to reach the brain, thus increasing alertness. Lowering the temperature can also activate the sympathetic nervous system, thereby raising adrenaline levels and consequently increasing alertness, while reducing fatigue.

[0042] Many passenger cars allow you to set temperature zones tailored to the driver and front passenger, enabling them to choose different ambient temperatures. This allows the area around the driver to be cooled without having to cool the entire vehicle.

[0043] If media is played while driving, it can be turned up louder to reduce driver fatigue. Scents that stimulate the nervous system in an activating way can also be used to combat fatigue. These can be provided via an air freshener. Examples include scents of citrus fruits, mint, eucalyptus, or rosemary.

[0044] Fatigue can also be caused by low oxygen levels. Therefore, opening at least one window and / or the sunroof of the vehicle can help reduce fatigue, as this allows fresh air with a high oxygen content to enter the vehicle.

[0045] Light, especially daylight with its associated wavelengths, also has an activating effect on the driver. Therefore, a so-called daylight lamp can be positioned within the driver's field of vision. This lamp can be activated to reduce or combat fatigue and its wavelength and / or intensity can preferably be adjusted to the ambient light conditions.

[0046] A massage can also have an activating effect on the driver, provided it is administered through appropriately designed vehicle seats. Such a massage can particularly stimulate reflex zones that promote alertness.

[0047] If the aforementioned measures are ineffective, or if a planned break is imminent or is deemed by the baseline to be the most foreseeable and effective measure at that moment, a break may be recommended. Such a recommendation can be communicated via voice output, text displayed on one of the vehicle's screens, displayed symbols, and / or similar means. The recommendation can be general or more specific, including information on nearby rest areas, hotels, parking lots, and similar locations where a break can be taken.Alternatively or additionally, an adjusted caffeine consumption can be recommended, which in particular specifies an amount of caffeine that is advisable at this time and with regard to the further journey, i.e. sufficient, and so that after arrival, for example, it does not disturb the night's sleep or impair the driver with undesirable side effects (racing heart, stomach problems and the like) due to an excessive amount.

[0048] Alternatively or additionally to the aforementioned measures, a power nap mode can be recommended and provided. The power nap mode involves darkening the vehicle once it has come to a standstill, for example, with electrically adjustable windows or sunshades, the closing of which is indicated to the driver. Sleep-inducing sounds can then be played and / or active noise cancellation activated to allow the driver to rest in the vehicle. After a predetermined period or based on the driver's observation, the vehicle will wake up before the driver enters deep sleep. This period can be ten to fifteen minutes.The transition to deep sleep can be recognized and estimated by breathing rate, driver movements, and similar indicators, and can therefore also be covered by one of the baselines. This short, light sleep phase reduces fatigue, allowing the driver to continue driving more alert.

[0049] Measures that require a vehicle stop can be combined with a refueling or charging stop. At least one of the aforementioned measures should therefore be combined with a stop to refuel the vehicle or, in the case of electric vehicles, to charge it. This way, for example, the time spent on a break is used to replenish sufficient fuel or energy for the remainder of the journey, without requiring a separate stop. This can also be done with consideration and knowledge of the planned journey, allowing for further optimization of that plan.

[0050] The process is executed on or by means of a control unit on which instructions in the form of program code or the aforementioned application are stored, the execution of which serves to implement the process. The instructions can also be stored on a memory device to which the control unit can access. The control unit can thus be understood as a processing unit.

[0051] To implement the method according to the invention, the control unit is connected to the vehicle's systems that monitor the driver and execute the necessary measures, enabling the control unit to receive and evaluate data and trigger the corresponding actions. If a measure cannot be implemented by controlling actuators, instructions can be transmitted to the driver to assist in its implementation.

[0052] The control unit can be a control unit provided in the vehicle that (with) executes the method according to the invention, but also a control unit that is primarily intended for executing the method and / or is subsequently installed in the vehicle.

[0053] According to the invention, a vehicle is also provided in which the method according to the invention is applied and / or in which a control unit as described above is used.

[0054] The solution proposed according to the invention reduces and delays fatigue on long journeys by first planning a route that is optimized with regard to emerging fatigue and / or by providing instructions for preparing for the journey in advance. Optional embodiments provide approaches for reducing fatigue should it nevertheless occur and / or for optimizing the process.

[0055] Unless otherwise stated in individual cases, the various embodiments of the invention mentioned in this application can be advantageously combined with one another.

[0056] The invention is explained below using exemplary embodiments with reference to the accompanying drawings. These show: Fig. 1 a schematic representation of the method according to the invention, and Fig. 2. A sketch for the implementation of monitoring and measures against fatigue.

[0057] Fig. Figure 1 schematically illustrates the process of the method according to the invention. It is assumed that the application required to execute the method is installed in the vehicle 10 and can be executed by a control unit. For planning his upcoming journey, the driver 12 uses his smartphone or an app from the vehicle manufacturer installed on it, which is communicatively connected to the application in the vehicle 10 and can therefore exchange data with it. In this exemplary embodiment, it is assumed that a personalized baseline exists and is used together with a generic baseline for detecting and combating fatigue during driving, as well as for planning and optimizing the journey. Both baselines are stored on a memory device that the executing control unit can access.

[0058] In step S1, driver 12 of vehicle 10 plans his upcoming longer trip, which is expected to take six hours and for which he plans to arrive at 4 p.m. the following day. He enters the destination and the desired arrival time. Based on the data from the generic baseline and the personalized baseline, a route is then suggested, optimized for route planning, and the departure time and breaks are already scheduled (step 2a). The departure is planned for 9 a.m. the following day. According to the plan, a first 15-minute break is scheduled after approximately one and a half hours, followed by a 30-minute break after another one and a half hours. This is followed by another one and a half hours of driving, another 15-minute break, and the final one and a half hours of driving. The route is chosen to ensure a varied and scenic journey.

[0059] In preparation for the journey, the driver is advised to get seven to eight hours of sleep and a light breakfast before starting the trip (step 2b). These recommendations, as well as a summary of the planned route, will be displayed to the driver on their smartphone in advance.

[0060] Both the planning of the journey (step S2a) and the creation of the recommendations (step 2b, both steps also summarized as step 2) are carried out taking into account the data from the personalized baseline and the generic baseline, with the data from the personalized baseline being weighted more heavily.

[0061] The driver 12 now enters his vehicle 10 and is registered and checked (step S3) to see if he is known (decision “ID”). This is the case (reference sign “Y”). Subsequently, the application for executing the method according to the invention is started in the vehicle 10 using both available baselines, and the planning data required for the journey is retrieved (step S4b).

[0062] If the vehicle user 12 were not known (decision “ID” and branch “N”), the application would only be started and provided with the data from the generic baseline (step S4a).

[0063] The driver 12 is then monitored during the journey using various vehicle-integrated systems and their wearable device 28 (step S5), which transmits their data to the control unit. The monitoring systems are described below with reference to Fig. 2. The monitoring data is analyzed for signs of fatigue, and these are identified (step S6). The assessment is based on the data from both baselines, taking into account the driving time to date. Optionally, the driver can actively report fatigue (step S9), for example, via voice input.

[0064] If signs of fatigue are detected, for example, because the driver 12 repeatedly yawns and blinks frequently, the control unit initiates measures to reduce fatigue (step S7). This could involve increasing the volume of the music the driver 12 is listening to while driving, lowering the vehicle's interior temperature, or activating a daylight lamp in their extended field of vision.

[0065] The driver's fatigue level is continuously monitored and assessed, and compared with the planned route and breaks. This may involve adjusting the timing and / or duration of the planned breaks. The remaining time until critical fatigue sets in (step S10), at which point a break becomes absolutely necessary, is also determined. Since this appears likely in this example, a power nap is suggested for the planned 30-minute break.

[0066] Since half the journey will already be completed, it makes sense to use the break for a charging stop for vehicle 10. Driver 12 can then use the charging time for a power nap. To do this, they park vehicle 10 at a charging station, plug it in, and get back inside. Following a prompt on the user interface, they darken the windows and hear calming sounds, allowing driver 12 to drift off to sleep. They are woken up just before falling into a deep sleep and can use any remaining charging time needed for a walk or a coffee. Afterwards, they can continue their journey refreshed and arrive at their destination relaxed.

[0067] The driver in this example has opted in, meaning they have consented to the sharing of their data for baseline optimization. Therefore, their personalized data is incorporated into their personalized baseline in step S8 (decision "Y, pers." to the right of opt-in). If no such baseline had been created previously, this could now be done in step S8a. In step S8b (both steps S8a and S8b are also referred to as step 8), they provide their anonymized data for the generic baseline (decision "Y, gen." to the left of opt-in). The data is then transmitted to the vehicle manufacturer, and the process is terminated (step E). If no data release has occurred, the process is terminated immediately (decision "N").

[0068] The data for the baselines are transmitted to the vehicle manufacturer, who uses it to further train the machine-learned models that form the basis for the detection and evaluation of fatigue signs and then provides it in the form of an update for the application in the vehicle 10.

[0069] In Fig. Section 2 now presents a sketch of selected systems for monitoring the driver 12 and combating fatigue. The arrangement is purely schematic.

[0070] The driver 12 is seated in his vehicle seat 14 in his vehicle 10. He is being monitored with a camera 20 or a thermal imaging camera 22. The camera primarily records the driver's face 12 to detect signs of fatigue, such as yawning, blinking, or similar behaviors. The recordings from camera 20 are then analyzed. The thermal imaging camera 22 can detect a drop in the driver's body temperature 12, which would also indicate fatigue.

[0071] The driver's breathing (12) is monitored by radar (24). If the driver becomes tired, their breathing typically slows. Additionally, a microphone (26) detects breathing and yawning sounds. Furthermore, data on breathing and heart rate can be transmitted to the vehicle (10) via the wearable (28), a smartwatch.

[0072] If sufficient signs of fatigue are detected after evaluation of the recorded data, countermeasures are triggered, such as the aforementioned increase in the volume of the music played via loudspeaker 30 and / or the activation of the daylight lamp 32. Since the vehicle seat 14 also has a massage device 34, this can also be used to activate the driver 12. Reference symbol list 10 vehicles 12 drivers 14 vehicle seats 20 cameras 22 Thermal imaging camera 24 Radar 26 microphones 28 Wearable 30 speakers 32 daylight lamp 34 massage facilities S1...S10 Process steps Y decision Y, gen. decision Y, personal decision N decision ID Decision question opt-in decision question E End QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited patent literature

[0000] US 2021 / 0 129 748 A1

[0004]

Images