A method for displaying individual home screens on a display device of a vehicle and a corresponding display device

Abstract

The present invention relates to a method for displaying an individual home screen (14) on a display device (12), the method comprising: receiving at least a first request from a first vehicle component (28) in relation to considerations for displaying a first function symbol (16); receiving at least a second request from a second vehicle component (28) in relation to considerations for displaying a second function symbol (16); determining a context cluster (26) through current context data of a vehicle (10); determining a current first rank (30) for the first request component (28) and determining a current second rank (30) for the second request in the current context cluster (26); and comparing at least the second rank (30) with the first rank (30). and if the given first rank (30) is greater than the second rank (30), the individual home screen (14) is displayed with the first function symbol (16) having a higher priority than the second function symbol (16), and vice versa; wherein the first rank (30) is determined according to the type and frequency of interactions performed by the user in relation to the first vehicle component (28) of the context cluster (26), and the second rank (30) is determined according to the type and frequency of interactions performed by the user in relation to the second vehicle component (28) of the context cluster (26). Furthermore, the present invention relates to a display device (12).

Description

Technology Field

[0001] The present invention relates to the field of automobiles. More specifically, the present invention relates to a method for displaying an individual home screen on a display device of a vehicle, wherein at least one function symbol is displayed on the home screen. Furthermore, the present invention relates to a corresponding display device. Background Technology

[0002] In modern technology, it is known to cluster functions into different applications and display an overview of all applications to the customer, with functions hidden in sub-menus. Smartphones are an example of this. However, in vehicles, particularly motor vehicles, this is a somewhat cumbersome and highly unsafe approach. Customers must focus on driving while simultaneously looking at the touchscreen and performing various clicks before reaching the desired task.

[0003] German patent application DE 10 2013 207 081 A1 relates to a motor vehicle having an operating device and a display, comprising a display device for displaying selectable graphic objects, an input for receiving input from a user capable of selecting one of the displayed objects, and a control device for controlling the display device as a function of the received input, wherein the control device is designed to assign a specific priority value to the displayed objects as a function of the received input and to set the size of each object and / or the arrangement of the displayed objects relative to each other as a function of the priority value during the operation of the display and the operating device.

[0004] U.S. Patent Application US 2020 257 364 A1 discloses a technology for interacting with a vehicle touchscreen. According to one or more embodiments, a system is provided comprising a display control component for selecting a graphic touch control to be included in a graphic user interface for rendering on a touchscreen based on the activation of a haptic feedback mode for interfacing with the touchscreen, and a processor for executing a computer executable component stored in at least one memory, wherein the graphic touch control corresponds to a control for one or more applications or functions associated with the vehicle. The system further comprises a positioning component for determining the position of a finger on or over the touchscreen for the graphic touch control as indicated on the touchscreen, and a haptic feedback component for causing a vibration unit of the vehicle to provide vibration feedback based on a position corresponding to the graphic touch control among the graphic touch controls. The problem to be solved

[0005] The object of the present invention is to provide a method for displaying individual home screens in a more comfortable manner on a display device of a vehicle and a corresponding display device.

[0006] The present object is solved by the method and corresponding display device according to the independent claim. Advantageous embodiments are presented in the dependent claim. means of solving the problem

[0007] One aspect of the present invention relates to a method for displaying an individual home screen on a display device of a vehicle, wherein at least one function symbol is displayed on the home screen. In relation to considerations for displaying a first function symbol of a first vehicle component, at least a first request generated by a provider associated with the first vehicle component is received. In relation to considerations for displaying a second function symbol of a second vehicle component, at least a second request generated by an additional provider associated with the second vehicle component is received. A step of determining a context cluster by the current context data of the vehicle is performed. A current first rank for the first request of the first vehicle component is determined in the current context cluster, and a current second rank for the second request of the second vehicle component is determined in the current context cluster. The first rank determined for the context cluster is compared with at least a second rank determined for the context cluster. An individual home screen assigned to the display device displays a first function symbol having a higher priority than a second function symbol when a given first rank is greater than a second rank, and vice versa. A first rank is determined according to the type and frequency of interaction with a first vehicle component, and a second rank is determined according to the type and frequency of interaction with a second vehicle component. A higher priority means that, for example, if the description space is restricted to displaying only one of the first or second function symbols, the symbol with the higher priority is shown and the other symbol is not shown. In other additional embodiments, priorities are determined for third, second, and additional function symbols.

[0008] Therefore, individual in-vehicle home screens are displayed in a more comfortable manner. Consequently, the necessary interaction with the head unit is reduced, and easy access to all desired personalized features and use cases at a high level is provided as much as possible. This is accomplished particularly through the support of artificial intelligence, such as neural networks. For example, the navigation map on the home screen is combined with the so-called Magic Dock. The Magic Dock is populated with predictions from an artificial intelligence kernel and displays the most desired features at any given time based on user behavior learned within various contexts, such as time, location, and weather.

[0009] For example, the display device may include a user interface background with a dynamic Global Positioning System (GPS) map. For instance, a global search function is layered at the top, along with a so-called Magic Dock at the bottom. Entertainment tiles, displayed in an expanded form, can be reduced to a minimum tile size or scaled down across the entire length of the dock when no suggestions are relevant. They may be reduced in size if they are not in use or / or if there is limited space—that is, if there is no space to display additional tiles in the display area. In terms of applications, tiles are function symbols associated with vehicle components that the user can interact with. For example, when a phone is connected, a static phone tile follows, followed by an arbitrary active use case. Examples of active use cases include an ongoing seat massage program, a phone call, or a seat heating activity, which can directly provide the user with the option to immediately engage in the ongoing activity. Finally, there are personalized tile suggestions facilitated by a new artificial intelligence algorithm that includes a subset of suggested use cases learned by the system regarding what the customer is interested in within a given contextual situation.

[0010] The display unit is configured to present a context-aware and personalized feature subset, which may be considered use cases, on the home screen. This allows for quick and easy access while limiting customer distractions throughout the driving journey. As a result, the driving environment becomes safer, and customers enjoy a more pleasant and less stressful user experience. While the use cases are a predefined set, they can be extended to additional use cases via backend or over-the-air updates. Additionally, the context features considered by the artificial intelligence are a predefined set but are also scalable over-the-air updates for the head unit / display unit.

[0011] In particular, a context cluster refers to a specific set of values ​​for each context feature data based on time, location, temperature, weather, day of the week, week, and / or other environmental conditions. A default rank is assigned to each use case to define its importance. The evolutionary rank associated with each use case changes over time as the artificial intelligence learns. Use cases will evolve different ranks for each learned context cluster. Use cases with higher ranks have higher priority than use cases with lower ranks. Use cases can also be considered as vehicle functions. These are features that the artificial intelligence kernel proposes to the customer.

[0012] According to one embodiment, requests for a first vehicle component and / or a second vehicle component are generated by a provider associated with the first and / or second vehicle components only when a predetermined rule is satisfied. For example, there may be a rule not to exceed a predetermined speed limit. For example, if the context is a parking context, the speed limit must be kept below a specific value, and a function symbol associated with Parktronic—that is, a component acting as a parking assist device—may be displayed on a display device. In particular, for example, the function of Parktronic may be used only when the vehicle is driving below a specific speed limit. This means that Parktronic is meaningful as a use case only when these requirements are met. Unlike this type of rule based on requests, some vehicle functions generate persistent requests independently of any rule, such as music streaming services, weather forecast apps, and system configuration functions. Vehicle functions generating persistent requests are always available, while rule-based requests are available only in specific situations.

[0013] The probability of using vehicle components is used to generate relevant ranks. In particular, user interaction is a major consideration influencing learning. Users interact with use cases via the home screen, for example, by providing positive feedback or temporarily discarding it, which implies, for example, negative feedback or simply not interacting with it, which can be considered neutral feedback. Furthermore, users may choose to interact with specific use cases not displayed on the home screen via the application menu, navigating the user interface by traversing the home screen and reading submenus through the application menu. The display device considers this as positive reinforcement for the use case, which will have a direct impact on the ranking. Whenever a user interacts with a use case supported by the display device, the model is trained with that data, and the corresponding kernel is updated.

[0014] In another more advanced embodiment, the first and second ranks are determined using an evolving beta distribution having an increasing or decreasing mean value, and the mean value is associated with or at least corresponds to the rank. In particular, as the kernel learns, the assumed value of the rank within each class evolves, that is, increases or decreases depending on the interaction between the vehicle component and the user through the function symbol or selection menu depicted on the display. This feature is modeled by a beta distribution signifying whether the customer engages in or does not engage in a use case, and is subsequently parameterized by combining a prior beta distribution. The beta distribution directly expresses the kernel's belief regarding the probability that the use case will be engaged by the user. As the kernel observes customer engagement or non-engagement of each use case within each cluster, this prior evolves. The evolution of the beta distribution is performed by updating the parameters α and β (mean = α / (α+β)), which are used to calculate the mean of the beta distribution, according to predefined rules based on the type of interaction. The rules are stored in a lookup table in the memory of the display device or are calculated by an appropriate algorithm. The interaction types are clicking the symbol, swiping the symbol, no interaction with the provided symbol, and / or activation through a submenu, and for each interaction, an update parameter is assigned that evolves the prior values ​​of α and β.

[0015] In another embodiment, the first rank and the second rank are determined at a predetermined time step and / or comparison is performed at a predetermined time step and / or the step of determining the current context cluster by the vehicle's context data is performed frequently at a predetermined time step. This is also referred to as a timer check. If the use case passes the setup check, the timer check occurs. The timer serves two purposes. It ensures that the home screen is not updated too frequently, and that an AI check is performed at specific intervals when the context changes, even if a new use case is not provided. If the timer check passes and an update to the home screen can be considered, a request or prediction call is made to the context clusterizer in the ranking algorithm.

[0016] According to another embodiment, the display of the first function symbol and / or the second function symbol is suppressed based on input by the user of the vehicle. This may be considered as a list of denials. In particular, a time check is performed on this list of denials. The user has an option to deny specific use cases, which means the user can request that such use cases not be shown again. This list of denials can be reset only when all history and learning are reset. Use cases swiped in the user interface are denied, that is, ignored during the journey. Use cases permanently denied in the settings will not be displayed as long as the setting item is turned off. In an additional embodiment, the display of the first function symbol and / or the second function symbol is suppressed as long as the rank of the request for the relevant vehicle component falls below a predefined threshold. The threshold prevents the display of symbols that are not interesting to the user and avoids unnecessarily distracting the driver.

[0017] In another embodiment, each context cluster containing one or more types of vehicle component requests refers to a subset of context features. In a ranked cluster, vehicle component requests represent the functionality of a component that can be automatically requested by a provider or initiated by a user. These context features may be weather, vehicle navigation status, speed, or time. In particular, these contexts may be assigned to environmental conditions. Specifically, the display device receives input and learns from various sources. The primary inputs are environmental context features such as time, location, and weather. However, other inputs include behavioral contexts such as navigation status, seat heater usage, and passengers in the vehicle. As the user interacts with various use cases, the display device learns which contexts are important for which use cases and learns appropriate rank values ​​for each set of contexts. These specific individualized sets of contexts are called clusters and form the basis upon which ranking algorithms can be applied. Within each cluster, individual use case rankings are continuously updated and stored to produce accurate predictions based on various contextual situations.

[0018] According to another embodiment, a context cluster is established by training, preferably by training a neural network or other known machine learning techniques known from the latest technology. In particular, therefore, the display device includes a neural network or other machine learning techniques. Therefore, individual training of the display device can be realized.

[0019] In another embodiment, within each context cluster, the ranking of vehicle component requests is stored and evolved individually. In particular, the same request may be implemented in multiple context clusters. A context cluster refers to a subset of context features, such as time and location. One context cluster may refer to morning time and home location, while another may refer to lunchtime and office location. In both cases, it is the same request, for example, a phone call system or a time scheduler notification.

[0020] In another additional embodiment, multiple context clusters are arranged in a cluster ensemble, with each cluster ensemble being processed in a separate kernel in parallel with other cluster ensembles. In particular, to improve overall accuracy and the resulting user experience, an ensemble of learning kernels is deployed in parallel with the entire model running on a single vehicle. Each kernel of the controller configured by the display device operates simultaneously on different subsets of context features, and the results are combined into a list of final use case suggestions. This ensemble model not only addresses the problem of simultaneously handling missing context features, such as glitches where input signals are not transmitted or are lost, but also enables the pre-training of a general base model suitable for most / all users, thereby generating immediately desirable and intuitive suggestions in the new user profile lifecycle.

[0021] The kernel consists of cluster ensembles that are trained and constructed in parallel. Each cluster ensemble is composed of clusters based on a subset of context features assigned to the cluster ensemble. For example, one cluster ensemble might construct clusters based on time and location context features, another based on day of the week and location, and an additional cluster ensemble based on weather, temperature, location, etc. The clustering constructed for each of these cluster ensembles is independent of one another. The clustering of context clusters and the ranking of requests associated with vehicle components are performed simultaneously and in parallel, independent of other cluster ensembles, through training on each data point. Therefore, the context clusters and rankings of requests associated with vehicle components formed in each cluster ensemble can be completely different from one another. For example, in the case of parking, requests for vehicle components such as Parktronic and related symbols will have different rankings. For example, for these vehicle components in clusters of different cluster ensembles, one cluster ensemble rank can be 0.7 and the other cluster ensemble rank can be 0.3.

[0022] By reasoning based on the given context, one can find the appropriate or at least the closest cluster within each cluster ensemble, and based on this, find the rank within that cluster. Then, for a k-cluster ensemble, each rank will yield the k ranks from other cluster ensembles.

[0023] To calculate the final ranking for each request for a vehicle component, it is necessary to average the rankings across different cluster ensembles. In a further embodiment, since each cluster ensemble has an associated weight, this will be a weighted average that gives higher weight to the rankings by cluster ensembles with higher importance. Preferably, these weights are adjusted based on an evaluation of large-scale simulation data (including more than 800 data sets) generated based on different story scenarios, and the weights that result in the best average performance across all these data sets are selected based on accuracy metrics.

[0024] In another embodiment, at least one static function symbol of an additional vehicle component is additionally displayed on the home screen, and the static function symbol is displayed in a non-adjustable manner or at least in an adjustable manner. For example, the static function symbol may be a navigation card that is always visible. In particular, the first function symbol and the second function symbol may be displayed in an emphasized manner. Therefore, functions such as navigation or phone calls are important functions and are always displayed. The static function symbol of the additional vehicle component is resized so that a symbol associated with an additional vehicle function requested by a provider having a rank higher than a predefined threshold may be displayed in a display area with limited space. In this case, the static function symbol is reduced to create an area where additional function symbols, such as the first and / or second function symbols, can be displayed. Function symbols next to the static function symbol are displayed according to priority until the display area is completely occupied, and additional function symbols are rejected.

[0025] Another additional aspect of the present invention relates to a display device of a vehicle for displaying an individual home screen, wherein at least one function symbol is displayed on the home screen, the display device comprises at least one electronic computing device, and the display device is configured to perform a method according to the aforementioned aspect. In particular, the present method is performed by the display device.

[0026] Another additional aspect of the present invention relates to a vehicle, particularly a motor vehicle, comprising a display device. The display device comprises, for example, electronic means and additional electronic means for performing a method, such as a processor integrated circuit.

[0027] Further advantages, features, and details of the present invention are derived from the drawings as well as the following description of preferred embodiments. Features and combinations of features mentioned in the foregoing description, as well as features and combinations of features mentioned in the following description of the drawings and / or illustrated only in the drawings, may be used in other combinations or alone without departing from the scope of the present invention, in addition to the combinations shown. Brief explanation of the drawing

[0028] Novel features and characteristics of the present disclosure are set forth in the appended claims. The accompanying drawings, which are incorporated into and constitute part of the present disclosure, serve to illustrate exemplary embodiments and, together with the description, explain the principles disclosed. The same numbering is used throughout the drawings to refer to similar features and components. Some embodiments of the system and / or method according to the embodiments of the present technical gist are now described below merely as examples and with reference to the accompanying drawings. The drawing is shown below: FIG. 1 is a schematic side view according to an embodiment of a motor vehicle having an embodiment of a display device; FIG. 2 is a schematic flowchart according to an embodiment of the method; FIG. 3 is two schematic diagrams according to an embodiment for determining weights; Figure 4 is an example of evolving the rank of a request for a vehicle component. In drawings, elements of the same type or elements having the same function are indicated by the same reference numeral. Specific details for implementing the invention

[0029] In this document, the word "exemplary" is used to mean "serving as an example, illustration, or illustration." Any embodiment or implementation of the subject matter described herein as "exemplary" must not be interpreted as being more desirable or advantageous than other embodiments.

[0030] Although various modifications and alternative forms are possible for the present disclosure, specific embodiments are illustrated in the drawings by way of example and will be described in detail below. However, the present disclosure is not intended to be limited to the specific forms disclosed, but rather should be understood to encompass all modifications, equivalents, and substitutions that fall within the scope of the present disclosure.

[0031] The terms “include,” “comprising,” or any other variation thereof are intended to encompass non-exclusive inclusion, so that a configuration, apparatus, or method comprising a list of components or steps may not include only said components or steps, but may include other components or steps not explicitly listed or not unique to such configuration, apparatus, or method. In other words, one or more elements in a system or apparatus preceded by “include” or “comprising” do not, without further restriction, exclude the existence of other elements or additional elements in the system or method.

[0032] In the following detailed description of embodiments of the present disclosure, reference is made to the accompanying drawings, which illustrate specific embodiments that form part of the present disclosure and in which the present disclosure can be practiced. These embodiments are described in sufficient detail to enable those skilled in the art to practice the present disclosure, and it should be understood that other embodiments may be used and modifications may be made without departing from the scope of the present disclosure. Therefore, the following description should not be taken in a limiting sense.

[0033] FIG. 1 is a schematic side view according to an embodiment of a motor vehicle (10) including a display device (12). In particular, the display device (12) is intended to display an individual home screen (14) that is shown in an enlarged view. At least one function symbol (16) is shown on the home screen, and the display device (12) includes at least one electronic computing device (18). In particular, the electronic computing device (18) may include artificial intelligence, in particular at least one kernel (20) and / or a plurality of kernels (20).

[0034] In particular, the home screen may display, for example, at least one static function symbol (22), which may be, for example, a navigation screen and / or a tile for phone calls. In particular, a set of functions is provided to the user (24) of the motor vehicle (10) and / or the display device (12), and the static function symbol (22), which is an active tile that is always displayed when the user (24) performs a specific task as well as proposed functions based on artificial intelligence learning, is provided to the entertainment module, for example, the static phone module. A continuous user interface background shown as a dynamic global positioning (GPS) map is illustrated in FIG. 1. A global search function is layered on top, for example, and a so-called magic dock, which is, for example, a first function symbol (16) or a second function symbol (16), is layered on the bottom.

[0035] Entertainment tiles displayed in an expanded form may be reduced to a minimum tile size when no suggestions are involved, or reduced across the entire length of the dock or nearly across it. When a phone is connected, a static phone tile follows, followed by any active use case. Examples of active use cases include an ongoing seat massage program, a phone call, or a seat heating activity, and may provide the user with the option to immediately engage in the ongoing activity. Finally, there are personalized tile suggestions facilitated by an electronic computing device (18) containing a subset of suggested use cases that the system has learned the user (24) is interested in in a given contextual situation.

[0036] In particular, a context cluster (26) (Fig. 2) refers to a specific set of values ​​of context features based on other environmental conditions, such as time, location, and weather, vehicle, and navigation status. The context cluster (26) is arranged within a cluster ensemble (62) of a learning kernel (20), and multiple cluster ensembles (62) operate in parallel within the learning kernel. A context feature subset (33) is assigned to each cluster ensemble, and the context cluster (26) within each cluster ensemble (62) is learned based on this context feature subset (33). In this context, donating means that a provider, which may be, for example, a vehicle component (28) (Fig. 2) or an associated electronic module of a vehicle component assigned to a display device, informs the electronic computing device (18) that the use case(s) should be considered as being displayed on the home screen (14) upon request. Some examples of vehicle components include an energizing coach, seat heater, parking assistance system, telephone, massage, or navigation system. The contribution may be permanent for the entire duration of driving or limited to a specific time by specific rules. Vehicle components (28) or related electronic modules, which may also be called providers, contribute to the electronic computing device (18) for consideration of their use cases. They may contribute or withdraw the use cases at any time. Each application has one provider for its set of use cases. Each application may have a rank (30). In particular, a default rank (30) is assigned to each use case to define the importance of the use case. The rank (30) changes over time as the electronic computing device (18) learns. Use cases will evolve different ranks (30) for each learned context cluster (26). Use cases with higher ranks have higher priority than use cases with lower ranks.The vehicle component (28) has the option to remove the contribution to be considered on the home screen (14). This withdrawal typically occurs when the specific requirements for the use case to be performed depend on, for example, when the vehicle transmission changes from park to drive, and the trunk open tile may be withdrawn. The use case may also be considered as a function of the vehicle (10). These are functions that the electronic computing device (18) intends to offer to the user (24).

[0037] FIG. 2 is a schematic flowchart according to an embodiment of the method. In particular, FIG. 2 illustrates that a method for displaying individual home screens (14) includes the step of receiving at least a first request from a first vehicle component (28) in relation to considerations for displaying a first function symbol (16). Furthermore, the step of receiving at least a second request from a second vehicle component (28) in relation to considerations for displaying a second function symbol (16) is performed. A context cluster (26) is determined by the current context data of the vehicle. A current first rank (30) for the first request of the first vehicle component (28) and the current context cluster (26) is determined, and a current second rank (30) for the second request of the second vehicle component (28) in the current context cluster (26) is determined. The first rank (30) and the second rank (30) are compared in a comparison module (34). Individual home screens (14) are displayed with a first function symbol (16) having a higher priority if the given first rank (30) is greater than the second rank (30), and vice versa. If only one function symbol can be displayed due to limited space, this symbol with a lower priority is ignored.

[0038] The first rank (30) is determined according to the number and type of interactions of use related to the first vehicle component (28) of the context cluster (26), and the second rank (30) is determined according to the number and type of interactions of use related to the second vehicle component (28) of the context cluster (26).

[0039] In particular, the flow of the electronic computing device (18) is illustrated in FIG. 2. First, a use case must be donated through an electronic module associated with a vehicle component (28). Each application, also known as a provider, donates a designated use case to the electronic computing device (18) to be considered for the home screen (14), specifically to the kernel (20). The provider may choose to donate the use case at startup and leave it for consideration at all times, or donate it based on a designated rule and withdraw the use case when the rule no longer applies. An example of the latter could be the Parktronic Plus (PTS) use case. Since the function can only be used when the vehicle (10) is driving below a certain speed limit, it is reasonable for the provider to donate the use case only when this requirement is met.

[0040] When a use case is provided to the electronic computing device (18), the kernel (20) first checks the customer settings (36). Through the customer settings (36) or user settings, the profile-specific user (24) can select the use cases that the system wants to learn about and show to the home screen (14) and the use cases that it wants to ignore. The initial settings (36) cover use cases for phone, navigation, convenience, vehicle, in-vehicle office, and online voice services.

[0041] If a use case passes the configuration check, a timer (38) check occurs. The timer (38) serves two purposes. It ensures that the electronic computing device (18) check is performed at specific intervals when the home screen (14) is not updated too frequently and when use cases are not donated but the context changes. If the timer (38) check passes and an update to the home screen (14) can be considered, a request or prediction call is made to the context clusterer and the ranking algorithm. The clusterer determines which of the learned environmental context categories is most appropriate for the current conditions and, combined with an internal ranking algorithm, provides priority feedback on which use cases should be presented for viewing. Only donated use cases are considered. The clusterer receives input and learns from various sources. The primary inputs are environmental context features such as time, location, and weather. However, other inputs include behavioral contexts such as navigation status, phone usage, or passengers in the vehicle. As the user (24) interacts with various use cases, the electronic computing device (18) learns which context is important for which use case and learns appropriate rank values ​​for each set of contexts. These specific individualized sets of contexts are called 26 clusters and form the basis on which a ranking algorithm can be applied. Within each cluster, individual use case rankings are continuously updated and stored to produce accurate predictions based on various contextual situations.

[0042] After the clusterizer is executed, a check is performed on the denial list (40). The user (24) has the option to deny specific use cases, which means the user can request that these use cases not be shown again. In an exemplary embodiment, this denial list (40) may be reset only when all history and learning are reset. Use cases swiped on the display device (12) are ignored only during the travel period, which means a check is performed on the temporary ignore list (42). Use cases permanently denied in the settings will not be shown as long as the setting item is turned off.

[0043] If the use case passes the veto list check, a final comparison (44) is made between the rank (30) of the use case and the default rank as a predefined threshold given by the static entertainment use case. If the rank of the donated additional use case is equal to or greater than the static default rank, it is displayed to the user (24) on the home screen (14) along with a feature symbol. If the rank is lower than the static default rank, it will not be displayed and the entertainment module will remain extended.

[0044] In particular, user interaction is a major consideration affecting the learning process. The user (24) interacts with use cases through the home screen (14), for example, by providing positive feedback and temporarily discarding it, which means, for example, negative feedback or simply not interacting with it, which can be considered neutral feedback. Furthermore, the user (24) may choose to interact with specific use cases not displayed on the home screen (14) through the application menu, thereby navigating the display device (12) by reading through sub-menus. The electronic computing device (18) also considers this so-called Long-Way Access as positive reinforcement for the use case, which directly affects the learned rank within the relevant context cluster. In other words, whenever a vehicle component requested by, for example, the head unit's menu or sub-menu is activated, this request will be ranked higher, and even if the relevant symbol is not displayed on the display, the rank will be higher. Once the rank is sufficiently high, the relevant symbol will appear on the display, providing improved usability. Whenever the user (24) interacts with a use case supported by the electronic computing device (18), the model is trained with the corresponding data and the electronic computing device (18) is updated.

[0045] Rank (30) is a fundamental core to the decision-making process within the electronic computing device (18). Each use case starts with an initial rank (30), which can be considered a cold-start prior. Rank (30) defines the relative importance of a use case compared to other use cases. The learned rank (30) will increase as the user (24) engages in the use case more frequently. The rate at which the estimated rank (30) can change will depend on the type of interaction. The electronic computing device (18) learns that the assumed value of rank (30), specifically the prior, evolves within each cluster. These quantities are modeled by the beta distribution (46) illustrated in FIG. 3, which signifies whether the user (24) engages in the use case or not. The beta distribution directly represents the kernel's belief regarding the probability that the use case will be engaged by the user (24). As the kernel (20) observes user participation or non-participation for each use case within each cluster, these priors evolve.

[0046] FIG. 3 is a two-dimensional diagram for determining ranks. The illustrated beta distribution represents the distribution of the probability that a specific task will be performed. The top panel of FIG. 3 illustrates three examples of beta distributions and highlights how the mean of the distribution corresponds to the denormalized rank (30). These probability distributions are examples of the total information content stored in the beta parameter set for a specific use case within a specific cluster. For example, the first curve (48) represents the belief that the best guess on the estimated probability that the user (24) will interact with the use case is 70%, represented by the mean value of 0.7 shown on the x-axis, for instance. Probabilities correspond to ranks, and a higher probability implies a higher rank. The mean value is characterized by mean = α / (α+β), using α and β as beta parameters. In the case of curve (48), α = 7 and β = 3. However, the confidence in the corresponding numbers is somewhat low, which is expressed by a wide width and low height, i.e., the high variance of the first curve (48). In other words, confidence is associated with width, width is associated with the height of the graph, and the value corresponding to the height is shown on the y-axis. Alternatively, the peak of the second curve (50) (α = 70 and β = 30) has the same mean of 70%, but the confidence for that value is much stronger, which is represented by a narrower width and a higher height, i.e., a smaller variance. Finally, the third curve (52) (α = 800 and β = 800) exemplifies whether confidence is very high in a 50% probability estimate. User behavior / task history is parameterized in a beta distribution. Each tile / task has its own default prior for the beta parameter. Over time, as the user interacts, the beta parameter within the context cluster is updated, and the rank evolves accordingly.

[0047] To illustrate in detail how the prior evolves with more observations of user engagement, the bottom panel of FIG. 3 illustrates the initial prior of the fourth curve (54), which features an initial assumption of about a 45% probability that the user will engage in the use case, but the wide width of the curve shows that the kernel (20) is not very confident about this prior assumption. The distribution of the fifth curve (56) is the actual probability distribution of user engagement. Over time, as the kernel (20) observes user behavior, it begins to learn updates to the probability estimates. This appears in the sixth curve (58) and the seventh curve (60), which move closer and closer to the actual distribution to become the fifth curve (56). Depending on the type of interaction, α and β are updated. For example, an interaction resulting from a click increases α by +1 and β remains unchanged, while long-distance access increases α by +5 and β remains unchanged. A neutral interaction increases β by +1, and a swipe operation increases β by +5; in both cases, α remains unchanged. To adjust the degree of change in α and β for different applications, they can be parameterized through software.

[0048] An example of an evolution rank associated with an average value is illustrated in FIG. 4. Starting from a preliminary state (70) with an average value of 0.5 (average = 2 / (2+2)=0.5), 15 interactions with vehicle components are performed. In this case, the interactions are clicking on tiles such as radio tiles. As each interaction preliminary value is updated, α increases by +1 and β remains unchanged. In the final state (72), the current value for α is 17 and β is still 2. The average value starting from 0.5 is now 0.81 (average = (2+15) / (2+15+2+2)=17 / 21=0.81), which gives a much higher priority to requests from vehicle components than before.

[0049] To make suggestions for use cases to actually display on the home screen (14) interface, the kernel (20) calculates a probability distribution for all available use cases. If the expected probability value is lower than the media tile value, the use case is discarded; in other words, if the rank corresponding to the probability is lower than the default rank associated with the use case, the use case will be discarded. The remaining probabilities are normalized and compared to form a sorted list of suggestions. The top suggestions are displayed on the user interface, specifically on the display device (12). An example of balancing the initial rank (30) can be found by comparing the missed call use case and the Paktronic use case. Assuming user data is aggregated, the user (24) engages in each of these use cases with an equal probability of, for example, 70%. However, the probability of the Paktronic has a much larger variability, with many people engaging in the Paktronic almost every time while others rarely engage. Due to the high spread, a wide initial beta distribution is applied to the Paktronic use case, which has low confidence. On the other hand, since the majority of users (24), especially when receiving notifications, check missed calls about 70% of the time, there is a high confidence that missed calls will receive a narrow beta distribution. The confidence in the 70% value of missed calls is stronger than the confidence in the 70% value of Pakhtronic. Incidentally, this example may be, for example, the first curve (48) which is Pakhtronic and the second curve (50) for missed calls.

[0050] The rank (30) of each use case is changed based on the interaction between the display device (12) and the user and is stored independently for all clusters. Within a cluster, the rank (30) evolves individually to create a fully personalized user experience. Use case rank updates through α and β evolution can make the rank (30) larger or smaller depending on the behavior. In the absence of any additional input, the learned rank (30) eventually degrades back to the initial default value.

[0051] The combination of the clusterizer and the ranking algorithm includes a single learning kernel (20). To improve overall accuracy and the resulting user experience, an ensemble (62) of learning kernels (20) is deployed in parallel with the entire model running on a single vehicle (10). Each kernel (20) operates simultaneously on a subset of context features using a specific set of parameters, and the results are combined into a list of final use case suggestions shown by block (64). The ensemble model not only solves the problem of evolving around missing context features, such as glitches where input signals are not transmitted or are lost, but also enables the pre-training of a general base model suitable for most / all users (24), thereby generating immediately desirable and intuitive suggestions in the new user profile lifecycle.

[0052] As use cases, for example, phone calls may be presented. Other use cases may be comfort use cases such as massage, heating control, energizing comfort programs, air defense, energizing activation programs, or power naps. Other use cases may be navigation use cases, and lasts, favorites, and / or predicted destinations may be presented. Other use cases may relate to the vehicle, for example, trunk control, comfort door control, vehicle level adjustment, or Parktronic use cases. Furthermore, an in-vehicle office may be a use case that could be, for example, call logs and birthdays. Furthermore, online voice services may be a use case for, for example, food ordering. The use cases presented as such are merely examples.

[0053] The contexts considered by the kernel (20) may include local time, time boot, GPS, internal temperature, external temperature, air quality, passengers in the vehicle, speed, navigation guidance status, estimated time of arrival, driving status, seat heating status, and time since the last application use. These contexts will be updated and added.

Claims

Claim 1 A method for displaying an individual home screen (14) on a display device (12) of a vehicle (10), comprising: receiving a first request generated by a provider associated with a first vehicle component (28) in relation to considerations for displaying a first function symbol (16); receiving a second request generated by an additional provider associated with a second vehicle component (28) in relation to considerations for displaying a second function symbol (16); determining a context cluster (26) through current context data of the vehicle (10); determining a current first rank (30) for a first request of the first vehicle component (28) in the current context cluster (26), and determining a current second rank (30) for a second request of the second vehicle component (28) in the current context cluster (26); and comparing the second rank (30) determined for the context cluster (26) with the first rank (30). A method for displaying an individual home screen (14) on a display device (12) of a vehicle (10), comprising the step of displaying the individual home screen (14) assigned to the display device (12) with the first function symbol (16) having a higher priority than the second function symbol (16) when the given first rank (30) is greater than the second rank (30), and vice versa, wherein the first rank (30) is determined according to the type and frequency of interactions performed by the user in relation to the first vehicle component (28) of the context cluster (26), and the second rank (30) is determined according to the type and frequency of interactions performed by the user in relation to the second vehicle component (28) of the context cluster (26). Claim 2 A method according to claim 1, characterized in that the request for the first vehicle component (28) or the second vehicle component (28) is generated by the provider associated with the first or second vehicle component when a predetermined rule is satisfied. Claim 3 A method according to claim 1, wherein the first rank (30) and the second rank (30) are determined using an evolving beta distribution (46) having an increasing or decreasing average value, and the average value is associated with the rank. Claim 4 A method according to claim 3, characterized in that the evolution of the beta distribution (46) is performed by updating parameters α and β according to predefined rules based on the type of interaction. Claim 5 A method according to any one of claims 1 to 4, characterized in that interaction with a vehicle component includes clicking a symbol, swiping a symbol, no interaction with a provided symbol, or activation through a submenu. Claim 6 A method according to any one of claims 1 to 4, wherein the first rank (30) and the second rank (30) are determined at a predetermined time step, or the comparison is performed at a predetermined time step, or the step of determining the current context cluster (26) through the context data of the vehicle (10) is performed frequently at a predetermined time step. Claim 7 A method characterized in that, in any one of claims 1 to 4, the display of the first function symbol (16) or the second function symbol (16) is suppressed according to input by the user (24) of the vehicle (10). Claim 8 A method according to any one of claims 1 to 4, wherein the display of the first function symbol (16) or the second function symbol (16) is suppressed as long as the rank for the request of the relevant vehicle component falls below a predefined threshold. Claim 9 A method characterized in that, in any one of claims 1 to 4, each context cluster (26) containing one or more types of vehicle component requests refers to a subset of context features. Claim 10 A method according to claim 9, characterized in that, within each context cluster, the rank of a request associated with a vehicle component is stored or evolved individually. Claim 11 A method according to claim 9, characterized in that the rank of requests associated with the same vehicle component is implemented simultaneously in multiple context clusters and evolves individually. Claim 12 A method characterized in that, in any one of claims 1 to 4, the context cluster (26) is established by training. Claim 13 A method according to any one of claims 1 to 4, wherein a plurality of context clusters (26) are arranged in a cluster ensemble (62), and each cluster ensemble (62) is processed in a separate kernel (20) in parallel with other cluster ensembles (62). Claim 14 A method according to claim 13, characterized in that a context feature subset is assigned to each cluster ensemble, and a context cluster in each of these cluster ensembles refers to such context feature subset. Claim 15 A method according to any one of claims 1 to 4, wherein at least one static function symbol (22) of an additional vehicle component (28) is additionally displayed on the home screen (14), and the static function symbol (22) is always displayed in a way that cannot be adjusted or can be adjusted by reducing its size. Claim 16 A method according to claim 15, wherein the static function symbol (22) of the additional vehicle component (28) is resized when a symbol associated with an additional vehicle function requested by a provider having a higher rank than the static default rank (22) is to be displayed in a display area having a limited space. Claim 17 A display device (12) of a vehicle (10) for displaying an individual home screen (14), wherein at least one function symbol (16) is displayed on the home screen (14), the display device (12) includes at least one electronic computing device (18), and the display device (12) is configured to perform a method according to any one of claims 1 to 4.

Images