Installation of a wireless update for vehicle software depending on the stress level of the driver
Patent Information
- Application Number
- PCT/EP2026/058075
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-04-16
- Filing Date
- 2026-03-23
- Publication Date
- 2026-10-22
Smart Images
Figure EP2026058075_22102026_PF_FP_ABST
Abstract
Description
[0001] Mercedes-Benz Group AG
[0002] Procedure for installing a wireless update and vehicle
[0003] The invention relates to a method for installing a wireless update in a computing unit of a vehicle as defined in the preamble of claim 1, and to a vehicle for carrying out the method.
[0004] With increasing digitalization, the proportion of computer systems in vehicles is also growing. The processing units in a vehicle, such as the electronic control units (ECUs), perform a wide variety of tasks, from engine control and providing driver assistance functions to infotainment features. To fix bugs, close security gaps, or introduce new functions, it may be necessary to update the software on a processing unit in the vehicle. In modern vehicles, this is typically done wirelessly, also known as an "over-the-air" (OTA) update. This simplifies the update process, as there is no need to visit a workshop or manually connect a computer-readable storage medium containing the update to the vehicle's onboard electronics via a suitable interface, such as a USB port or an SD card slot.
[0005] Installing a software update can restrict the availability of functions or the normal operation of the underlying processing unit. For example, the relevant processing unit must be restarted one or more times during the installation process, such as during a firmware flash. During this time, individual vehicle functions may be limited or completely unavailable. In the worst case, the vehicle may not even start. If a user wishes to interact with the vehicle during this time, it can lead to frustration.
[0006] US Patent 2015 / 0169311 A1 discloses the scheduling of automated software updates for vehicles. Software updates to be installed in a vehicle are scheduled using a scheduling manager. The timing and extent of a software update to be installed on a computing unit within the vehicle are determined based on the vehicle's behavior patterns. These behavior patterns can include a location history of places visited by the vehicle. Specifically, an update can be automatically installed if, based on historical behavioral data, it is determined that the vehicle is at home during a typical period of non-use. However, a disadvantage is that interaction with the vehicle is not possible if a user deviates from these typical behavior patterns.
[0007] Furthermore, JP2016190025 describes a system and procedure for determining personal psychological stress.
[0008] The present invention is based on the objective of providing an improved method for installing a wireless update in a computing unit of a vehicle, the implementation of which improves the user interaction between a vehicle for which the installation of the wireless update is intended and a user.
[0009] According to the invention, this problem is solved by a method for installing a wireless update with the features of claim 1. Advantageous embodiments and further developments, as well as a vehicle for carrying out the method, are described in the dependent claims.
[0010] A generic method for installing a wireless update in a vehicle's computing unit, wherein the vehicle downloads the wireless update from a central computing unit and determines a time to start an automatic installation of the wireless update during a parking period depending on external boundary conditions, is further developed according to the invention by the following method steps:
[0011] - Recording the stress level of a person driving a vehicle while driving the vehicle;
[0012] - Comparing at least the stress level measured at the end of the journey with a defined stress level threshold; and
[0013] - Failure to automatically install the wireless update or at least delaying the start of the wireless update installation during the vehicle's parking period if the measured stress level exceeds the specified stress level threshold.
[0014] The applicant has recognized the following potential problem: According to prior art methods for the automatic installation of wireless updates, these can be installed automatically when the vehicle is switched off. The vehicle typically detects that it is "switched off" when it receives a door locking command from outside. However, it can happen that the vehicle user leaves the vehicle after locking the doors, but then returns shortly afterward to retrieve an item left inside, such as a wallet, handbag, umbrella, or similar item. As mentioned earlier, installing the wireless update can temporarily disable basic vehicle functions. For example, the vehicle doors may not unlock during the wireless update installation process.Therefore, the user cannot open the doors upon reaching the vehicle. This can lead to significant frustration for the user, as they are unable to retrieve the forgotten item.
[0015] The applicant recognized that the risk of forgetting an item inside the vehicle increases with the user's stress level. According to the invention, the driver's stress level is measured, compared to a defined stress level threshold, and if the threshold is exceeded, the automatic installation of wireless updates during the parking process is either prevented or at least delayed. If the user then notices that an item has been left inside the vehicle, they can return to their vehicle, unlock it, and retrieve the item. This improves user interaction.
[0016] Performing a proactive user prompt to initiate the wireless update installation cannot reliably prevent the situation described above, as corresponding pop-up messages on a display device in the vehicle can easily be dismissed by a stressed user. Various methods for recording and assessing a person's stress level are known from current technology. For example, the driver's stress level can be measured and assessed by the vehicle itself. Various vital parameters can be recorded for this purpose, such as heart rate, skin conductance, eye activity (especially changes in gaze direction), blink rate, and the like. Stress can also be detected through voice analysis, for example, in voice commands issued by the driver to a vehicle-integrated voice assistant.To capture these parameters, the vehicle can be equipped with a variety of sensors, such as cameras, microphones, and the like. The vehicle can also receive relevant information, allowing for an assessment of the driver's stress level, from third-party devices, such as wearables worn by the driver. These include, for example, smartwatches, fitness trackers, and similar devices. Such devices may be capable of measuring skin conductance, heart rate, performing an ECG, and so on. Both vehicle-integrated and third-party device-integrated methods for capturing and assessing stress levels can be combined.
[0017] The crucial factor is the driver's stress level when parking the vehicle, i.e., at the end of the journey. This stress level is compared to a predefined stress threshold. This threshold is typically an empirical value established by the vehicle manufacturer, above which there is a particularly high risk of items being left inside the vehicle. The predefined stress threshold can be determined empirically, for example, by conducting extensive measurement campaigns.
[0018] These empirical tests can reveal a typical waiting period required to delay the installation of the wireless update. For example, there's a typical timeframe within which a user might remember something they've left inside the vehicle. In particular, there's a correlation between the driver's stress level and the length of time the automatic installation of the wireless update is delayed. The more stressed the driver, the longer the automatic installation of the wireless update can be postponed.
[0019] Depending on the level of stress, or the extent to which the measured stress level exceeds the defined stress threshold, a probability can be calculated of how likely it is that the user will return to the vehicle to retrieve a forgotten item. Further functions or measures can then be implemented based on this probability.
[0020] An advantageous further development of the method according to the invention provides that the driver is identified and the defined stress level threshold is adjusted taking into account a user-specific stress level profile, including the historical stress level behavior of the specific person. This allows for a more reliable determination of whether or not the respective user will leave objects in the vehicle interior in the current situation. Thus, the probability that the user will return to the vehicle after parking it can be estimated more reliably.
[0021] For example, the stress level present when the vehicle is parked can be tracked based on historical parking events. This allows for a correlation to be established regarding the personal stress level at which the user will typically return to the vehicle. If, for instance, the measured stress level for historical parking events exceeds the defined stress level threshold, but the user does not return to the vehicle, the defined stress level threshold can be automatically raised.
[0022] The user's return to the vehicle can be detected by the vehicle itself or by a mobile device carried by the user, such as a smartphone. A dedicated app can be running on the smartphone to compare the location of the parked vehicle with that of the user. The current location of the mobile device can be determined, for example, using a GPS-based navigation system integrated into the device. This allows the system to recognize when the user leaves the vehicle and when they return. Ideally, the driver's behavior is evaluated during the journey, and the stress level, recorded at least at the end of the journey, is adjusted accordingly. The driver's behavior can be directly correlated with their stress level.A stressed person will typically make more frequent driving errors, react more slowly to events in the given driving situation, drive more often and at higher speeds, maintain closer distances to vehicles ahead or behind, be more likely to run a yellow light instead of stopping, and so on. Driving behavior is primarily recorded by the vehicle itself using its own sensors. If corresponding driving behavior patterns are identified, the measured stress level of the user can be increased. In particular, driving maneuvers performed by the driver shortly before parking the vehicle, such as within the last five minutes, are evaluated.
[0023] It is also conceivable to adjust the set stress level threshold depending on observed driving behavior. If the currently measured stress level suggests that the driver should be stressed, but the driver exhibits conservative, safe, or "calm" driving behavior, the set stress level threshold can be raised, particularly as a reaction to this.
[0024] A further advantageous embodiment of the method according to the invention provides that the vehicle's parking duration is recorded and the time to delay the start of the wireless update installation is determined based on the recorded parking duration. This allows it to be determined, for example, that a parking session lasts an average of one hour on Saturday morning and 12 hours on Monday evening. For instance, a required installation time might be 30 minutes. In this case, the start of the installation could be delayed by 20 to 30 minutes on Saturday, as this would most likely ensure that the installation is completed by the end of the parking session. On Monday evening, however, a longer delay, such as 2 to 6 hours, is also possible. Furthermore, it can be determined whether and after what time the user returns to the vehicle.Historical vehicle parking data reveals, for example, that the vehicle is typically parked for 8 to 12 hours. However, there have been three instances where the user returned to the vehicle after 8, 12, and 9 minutes, respectively, without driving. This suggests that an item may have been left inside the vehicle. As a result, the wireless update can be delayed for at least 12 minutes after the vehicle is parked. Advantageously, additional safety buffers are incorporated, so the automatic installation might not start until 20 or 30 minutes have passed.
[0025] In particular, the vehicle can establish a temporal and spatial reference. For example, when parked, the vehicle can determine its exact location, similar to a mobile device using a GPS-based navigation system. The current time and day of the week or calendar date are also taken into account. This allows for the determination of different typical parking durations for different locations and times. For instance, the vehicle can determine that parking in the parking lot in front of the kindergarten on Wednesday mornings typically takes 3 minutes. In this case, the automatic installation of the wireless update is prevented. By considering this temporal and spatial reference, an even more accurate probability of the user's return can be calculated.
[0026] According to a further advantageous embodiment of the method according to the invention, the processing unit reads a digital calendar of the vehicle driver and determines the time to postpone the start of the installation based on upcoming appointments in the calendar. The time until the appointment and the actual appointment duration can be evaluated separately. If an appointment is imminent, the installation of the wireless update can be postponed, as the user may need to retrieve an item from the vehicle, such as a laptop. However, if the appointment has already started and will last longer than the expected installation time, the wireless update installation can begin. The current location of the vehicle can also be compared with a location stored in the appointment.For example, it can be determined that only an intermediate stop was made on the way to the actual appointment location, thus further delaying the start of the installation.
[0027] Preferably, the computing unit estimates the expected downtime based on upcoming deadlines, compares this expected downtime with the expected installation time for parts of the wireless update, and performs a partial installation of the wireless update so that the partial installation is completed by the end of the expected downtime. The typical installation time of the wireless update can be determined by the vehicle manufacturer during the development of the wireless update, for example, using a test system. This information can be transmitted from the central computing unit to the vehicle along with the wireless update, for example, in the form of metadata. Furthermore, the wireless update can consist of several individually installable update segments.The vehicle manufacturer can estimate the required installation time for each of these update segments, even specifically for different vehicle equipment variants. The performance of a computing unit can thus affect the installation time.
[0028] The vehicle's processing unit can now estimate the duration of the current or future parking period by considering the date, time, and start and end dates in the user's calendar. This estimated parking duration can then be compared by the processing unit with the expected installation time of the respective update segments. Update segments that can be installed within the expected parking period can then be installed automatically. The system prioritizes installing as many update segments as possible within the available time. Automatic installation also occurs primarily after the start of the installation has been delayed due to exceeding the defined stress level.
[0029] According to a further advantageous embodiment of the method according to the invention, it is further provided that the vehicle performs a visual interior scan at least at the beginning of the parking period and lowers the stress level threshold upon detecting an object left inside the vehicle. This results in a higher probability that the measured stress level will exceed the stress level threshold, thereby preventing or delaying the automatic installation of the wireless update.
[0030] If objects are actually detected inside the vehicle, this also indicates that the user has likely forgotten them. Established methods can be used for the visual scanning of the vehicle interior. For example, interior scanning can be camera-based. Artificial intelligence-based algorithms can be used to analyze the corresponding camera images. For instance, machine vision algorithms, also known as computer vision, can be employed. Camera images are particularly favored when processed using convolutional neural networks (CNNs). With the help of such AI models, objects can be classified, making it possible, for example, to distinguish a football from a soccer ball or a handbag from a wallet.
[0031] Whether an object identified inside a vehicle is one that was left behind can be determined by comparing currently captured images with reference images. For example, such reference images may have been taken in the past and show an empty vehicle interior. In a currently recorded camera image, a soccer ball might then be visible on the back seat. Since this soccer ball is not present in the reference image, this can be considered an indication that an object was left behind.
[0032] Preferably, camera images depicting the vehicle interior are processed by a suitably trained AI model, whereby the AI model determines a context for an object detected in one of the camera images, in particular based on text recognition of text applied to the object. The context of the respective detected object can be used to assess the probability that the user will return to the vehicle to retrieve the object. This can be used, for example, to raise or lower the defined stress level threshold only if the object's context can be assigned to a specific context category. Further external boundary conditions, such as the season and the weather, can be used for this purpose.For example, a water bottle can be detected inside the vehicle. In summer and at high outside temperatures, such as above 25°C, the system can lower the defined stress level threshold. This increases the likelihood that the user will return to the vehicle to retrieve the water bottle. Consequently, the installation of the wireless update is prevented or delayed when the vehicle is parked.
[0033] According to a further advantageous embodiment of the method according to the invention, it is further provided that the vehicle's location is determined when it is parked, points of interest located near the parking location are identified on a digital street map, a context for said points of interest is determined, the context of the respective object is compared with the context of the points of interest, and the stress level threshold is lowered only if the context of the object and the context of the point of interest are similar or identical. For example, a sports bag on the back seat of the vehicle can be detected. Using said positioning means, the vehicle can recognize that it is near a gym. Thus, the same context is determined for both the object and the parking location, namely sports, training, gym, or similar.Determining the context for both objects and points of interest (POIs) can be achieved particularly well using large language models, also known as large language models (LLMs). A large language model can also be used to assess the similarity of contexts.
[0034] A further advantageous embodiment of the method according to the invention provides that the method is carried out for a wireless update, the installation of which leads to a failure of a door locking mechanism of the vehicle during the installation period. In other words, it is thus possible to install wireless updates while the vehicle is parked without negatively impacting the accessibility of the vehicle during the installation period. Such wireless updates can therefore be installed safely, even if the measured stress level of the driver exceeds the defined stress level threshold.
[0035] In a vehicle of this type, comprising means for measuring and / or obtaining information describing the stress level of a person driving the vehicle, a computing unit for executing a program, and means for obtaining a wireless update for the program from a central computing unit, the invention provides that the computing unit is configured to execute a method described above. The computing unit thus has access to a computer-readable storage medium containing machine-interpretable instructions, which, in one embodiment, are provided by a processor within the computing unit, causing it to execute the individual process steps of the method according to the invention. The vehicle can be any road vehicle, such as a car, truck, van, bus, or the like. Generally, it is also conceivable that it could be a watercraft, rail vehicle, or aircraft.
[0036] Further advantageous embodiments of the inventive method for installing a wireless update in a vehicle's computing unit are also evident from the exemplary embodiment, which is described in more detail below with reference to the figure.
[0037] Figure 1 shows a schematic flowchart of the inventive method for installing the wireless update in an exemplary situation.
[0038] In step 101, a father drives his child to kindergarten in his car.
[0039] In step 102, the father's stress level SL is measured during the journey, for example, using the vehicle's own sensors or a wearable device. The father then has to drive to work and is running late, causing his stress level SL to rise sharply. A processing unit in the vehicle compares the measured stress level SL with a defined stress level threshold 1, which is illustrated by a graph in Figure 1. The stress level SL is plotted against time t.
[0040] In step 103, the father and child arrive at the kindergarten. Both get out of the vehicle, and the father locks it. Because the father is so stressed, he has forgotten his child's backpack in the trunk. The subsequent steps 104 to 106 symbolize the process according to a prior art method for installing wireless updates, while steps 104* to 106* illustrate the process according to the inventive method.
[0041] In step 104, the computing unit in the vehicle starts an automatic installation process for the wireless update downloaded from a central computing unit during the journey or previously, in response to the vehicle being switched off.
[0042] In step 105, the father realizes he left his child's backpack in the trunk. In step 106, he returns to his vehicle to retrieve it. Because the wireless update is in progress, the vehicle cannot process the door unlock command the father now enters. The trunk remains locked, forcing the father to wait until the wireless update is complete before he can get the backpack. This causes significant frustration and delays his arrival.
[0043] According to the inventive method, however, in step 104* the processing unit compares the measured stress level SL, at least when the vehicle is parked (i.e., when the vehicle doors are locked), with the defined stress level threshold 1. Since the stress level SL exceeds the stress level threshold 1, the processing unit prevents the automatic installation of the wireless update during the current parking process. At the very least, however, the processing unit delays the automatic installation for a defined period of time. This period can be determined, for example, by taking into account historical parking durations and / or the user's historical behavior. Since the processing unit recognizes a situation in which a relatively short parking duration is expected, the automatic installation is prevented.
[0044] In step 105*, the father notices that he has forgotten the backpack and goes back to the vehicle to get it.
[0045] Since the processing unit omits the wireless update, the father can open the trunk and retrieve the backpack in step 106*. User interaction between the driver and the vehicle has thus been improved.
Claims
Mercedes-Benz Group AG Patent claims 1. A method for installing a wireless update in a vehicle's computing unit, wherein the vehicle downloads the wireless update from a central computing unit and sets a time to start an automatic installation of the wireless update during a parking period depending on external boundary conditions, characterized by the following process steps: - Recording the stress level (SL) of a driver while driving the vehicle; - Comparing at least the stress level (SL) measured at the end of the journey with a defined stress level threshold (1); and - Failure to automatically install the wireless update or at least delaying the start of the wireless update installation during the vehicle's parking period if the measured stress level (SL) exceeds the specified stress level threshold (1).
2. Method according to claim 1, characterized by the fact that the person driving the vehicle is identified and the specified stress level threshold (1) is adjusted taking into account a user-specific stress level profile, including a historical stress level behavior of the specific person.
3. Method according to claim 1 or 2, characterized by the fact that The driving behavior of the person driving the vehicle is evaluated during the journey, and the level of the stress level (SL) recorded at least at the end of the journey is adjusted taking the driving behavior into account.
4. Method according to one of claims 1 to 3, characterized by the fact that The vehicle's parking duration is recorded, and the time to delay the start of installation is determined based on the recorded parking durations.
5. Method according to any one of claims 1 to 4, characterized by the fact that The computing unit reads a digital calendar of the vehicle driver and determines the time to delay the start of the installation depending on upcoming appointments in the calendar.
6. Method according to claim 5, characterized by the fact that The computing unit estimates an expected shutdown duration taking into account the upcoming deadlines, compares the expected shutdown duration with an expected installation duration for parts of the wireless update, and performs a partial installation of the wireless update so that the partial installation is completed at the end of the expected shutdown duration.
7. Method according to any one of claims 1 to 6, characterized by the fact that the vehicle performs a visual interior scan at least at the beginning of the parking period and lowers the stress level threshold (1) if an object left inside the vehicle is detected.
8. Method according to claim 7, characterized by the fact that Camera images depicting the vehicle interior are processed by a suitably trained AI model, whereby the AI model determines a context for an object detected in one of the camera images, in particular based on text recognition of text applied to the object.
9. Method according to claim 8, characterized in that the location of the vehicle is determined when it is parked, points of interest located near the parking location are identified in a digital street map, a context for said points of interest is determined, the context of the respective object is compared with the context of the points of interest, and the stress level threshold (1) is only lowered if the context of the object and the context of the point of interest are similar or identical.
10. Method according to any one of claims 1 to 9, characterized by a version for a wireless update, the installation of which leads to a failure of a door locking mechanism of the vehicle during the installation period.
11. Vehicle comprising means for measuring and / or obtaining information describing the stress level (SL) of a vehicle driver, a computing unit for executing a program and means for obtaining a wireless update for the program from a central computing unit, characterized in that the computing unit is set up to execute a method according to one of claims 1 to 10.