Universal key implementation method and device based on AI learning

Abstract

The invention provides a universal key implementation method and device based on AI learning, and the method comprises the steps: recognizing a current driving scene according to collected scene data, and obtaining a user operation habit set in the current driving scene from a scene habit library; according to the user operation habit set in the current driving scene and the user behavior data collected after entering the current driving scene, predicting the next vehicle function to be executed by the user, and mapping the universal key to the predicted vehicle function; and for each driving scene, extracting user operation habits in the corresponding driving scene according to the user behavior data collected in the same driving scene duration, and updating the user operation habits to the scene habit library. According to the application, the mapping relation between the universal key and the vehicle function is flexibly set in combination with the current driving scene and the operation habit of the user, and the scene habit library is continuously updated in an AI learning mode, so that the function operation complexity is not increased while the function operation requirement of the user is met.

Description

Technical Field

[0001] The present application relates to the field of intelligent vehicle computer technology, and specifically to a method and device for implementing a universal button based on AI learning. Background Art

[0002] With the development of intelligent automobiles, more and more functions are designed into automobiles, resulting in more and more complex and numerous functions of automobile computers. The traditional method of using limited physical buttons to control various functions of automobile computers is obviously unable to meet the needs. Automobile manufacturers need to continuously design more levels of central control menus to layout and control many functions.

[0003] However, there are safety risks in allowing the driver to take his hands off the steering wheel to tap on the screen menu while driving. Therefore, the control of the vehicle functions cannot be completely separated from physical buttons, and some key functions still need to be implemented with the help of physical buttons.

[0004] In related technologies, car manufacturers have designed custom buttons to allow users to map frequently used functions to physical buttons for quick control. This design allows users a certain degree of freedom, which to some extent alleviates the problem of a large number of functions in car computers but a small number of physical buttons. However, this solution has obvious limitations. As car functions continue to increase, the number of frequently used functions will also increase. Fewer custom buttons will not be able to meet users' functional operation needs, while more custom buttons will lead to a sharp increase in the complexity of functional operation and a sharp decline in user experience. Summary of the Invention

[0005] The present application provides a method and device for realizing a universal key based on AI learning, which can solve the technical problem existing in the prior art that it is impossible to meet the functional operation requirements of users while avoiding increasing the complexity of functional operations.

[0006] In a first aspect, an embodiment of the present application provides a method for implementing a universal key based on AI learning, the method comprising:

[0007] Identify the current driving scenario based on the collected scenario data and obtain the user operation habit set under the current driving scenario from the scenario habit library;

[0008] Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the car function the user will perform next and maps the universal button to the predicted car function.

[0009] For each driving scenario, based on the user behavior data collected during the same driving scenario, the user operation habits in the corresponding driving scenario are refined and updated to the scenario habit library.

[0010] Furthermore, in one embodiment, the step of predicting the vehicle-mounted function that the user will next execute based on the user's operation habit set in the current driving scenario and the user behavior data collected after entering the current driving scenario, and mapping the universal button to the predicted vehicle-mounted function includes:

[0011] Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the multiple car functions that the user will perform next.

[0012] If the predicted multiple car functions match any car function group defined by the user, the universal button is mapped to the corresponding car function group;

[0013] If the predicted multiple car machine functions do not match all car machine function groups defined by the user, the universal button is mapped to the first predicted car machine function.

[0014] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0015] The prediction accuracy of the corresponding driving scenario is calculated based on the predicted vehicle computer functions during each driving scenario and the vehicle computer functions actually performed by the user;

[0016] If the prediction accuracy of the current driving scenario is greater than or equal to the accuracy threshold, the universal button is enabled;

[0017] If the prediction accuracy of the current driving scenario is less than the accuracy threshold, the universal button is disabled.

[0018] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0019] Before the vehicle leaves the factory, a corresponding set of user operating habits is preset for one or more specified driving scenarios and stored in a scenario habit library.

[0020] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0021] After the user binds the vehicle, the user behavior data in one or more specified driving scenarios is collected through online simulated driving;

[0022] Based on the user behavior data collected in each driving scenario through online simulated driving, the user operation habits in the corresponding driving scenario are extracted and updated to the scenario habit library.

[0023] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0024] After the user binds the vehicle, the user's operating habits in one or more specified driving scenarios are collected through questionnaires and updated to the scenario habit library.

[0025] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0026] If the user operation habit set for any driving scenario in the scenario habit library is empty, then after experiencing the corresponding driving scenario and parking, collect user behavior data for the corresponding driving scenario through online simulated driving;

[0027] Based on the user behavior data in the corresponding driving scenarios collected through online simulated driving, the user operation habits in the corresponding driving scenarios are refined and updated to the scenario habit library.

[0028] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0029] If the user operation habit set for any driving scenario in the scenario habit library is empty, after experiencing the corresponding driving scenario and parking, the user operation habits for the corresponding driving scenario are collected through a questionnaire and updated to the scenario habit library.

[0030] Furthermore, in one embodiment, the scene habit library is associated with the vehicle computer account, and the scene habit libraries of different vehicle computer accounts are independent of each other.

[0031] In a second aspect, an embodiment of the present application further provides a universal key implementation device based on AI learning, the universal key implementation device based on AI learning comprising:

[0032] The habit query module is used to identify the current driving scenario based on the collected scenario data and obtain the user operation habit set in the current driving scenario from the scenario habit library;

[0033] The operation mapping module is used to predict the car function that the user will perform next based on the user's operation habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, and map the universal button to the predicted car function;

[0034] The habit update module is used to extract the user operation habits in each driving scenario based on the user behavior data collected during the same driving scenario, and update them to the scenario habit library.

[0035] In this application, the current driving scene is identified based on the collected scene data, and the user operation habit set in the current driving scene is obtained from the scene habit library; based on the user operation habit set in the current driving scene and the user behavior data collected after entering the current driving scene, the car function that the user will perform next is predicted, and the universal button is mapped to the predicted car function; for each driving scene, based on the user behavior data collected during the same driving scene, the user operation habits in the corresponding driving scene are refined and updated to the scene habit library. Through this application, the mapping relationship between the universal button and the car function is flexibly set in combination with the current driving scene and the user operation habits, and the scene habit library is continuously updated through AI learning to improve the prediction accuracy. While meeting the user's functional operation needs, it will not increase the complexity of the functional operation, thereby improving the user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 This is a flow chart of a method for implementing a universal key based on AI learning in one embodiment of the present application;

[0037] Figure 2 This is a logical diagram of a method for implementing a universal key based on AI learning in one embodiment of the present application;

[0038] Figure 3 This is a schematic diagram of the functional modules of a universal key implementation device based on AI learning in one embodiment of the present application. DETAILED DESCRIPTION

[0039] In order to enable those skilled in the art to better understand the present invention, the following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0040] In order to make the objectives, technical solutions and advantages of this application clearer, the implementation methods of this application will be further described in detail below with reference to the accompanying drawings.

[0041] In a first aspect, an embodiment of the present application provides a method for implementing a universal key based on AI learning.

[0042] Figure 1 The figure shows a flow chart of a method for implementing a universal key based on AI learning in one embodiment of the present application. Figure 2 A logical diagram of a method for implementing a universal key based on AI learning in one embodiment of the present application is shown.

[0043] Reference Figure 1 and Figure 2 In one embodiment, a method for implementing a universal key based on AI learning includes the following steps:

[0044] S1. Identify the current driving scene based on the collected scene data, and obtain the user operation habit set in the current driving scene from the scene habit library.

[0045] In this embodiment, considering that user operation habits are usually closely related to driving scenarios, corresponding user operation habits are associated with different driving scenarios respectively. Various user operation habits under the same driving scenario are summarized into the user operation habit set under the driving scenario, and the user operation habit sets under multiple driving scenarios are summarized into a scenario habit library.

[0046] Specifically, scene data is collected in real time through acquisition equipment, and may include external camera images, millimeter-wave radar point clouds, vehicle control data such as accelerator and brake, vehicle posture data, weather API data, etc.

[0047] It should be noted that the total number of driving scenarios may be very large, for example, covering more than 200 climate conditions and more than 50 road types. Depending on factors such as the user's region and driving habits, some driving scenarios may not be experienced throughout the entire vehicle life cycle. Therefore, the scenario habit library does not need to store the user operation habit sets for all driving scenarios.

[0048] S2. Based on the user's operation habit set in the current driving scenario and the user behavior data collected after entering the current driving scenario, predict the car function that the user will perform next, and map the universal button to the predicted car function.

[0049] In this embodiment, a designated physical button is used as a universal button, which can be mapped to different on-board functions as needed. Unlike custom buttons in related technologies, the on-board function mapped to the universal button is not manually set by the user. Instead, it flexibly changes based on the driving scenario, the user's operating habits during the driving scenario, and changes in user behavior after entering the driving scenario. After the universal button is mapped to a predicted on-board function, pressing the universal button will execute the corresponding on-board function.

[0050] Specifically, user behavior data is used to provide feedback on the user's status and operations. It is collected in real time through acquisition equipment and can include in-vehicle camera data, touch screen operation trajectories, voice command text, intonation feature data, typical operation sequence timestamps, etc.

[0051] For example, a user's operating habits in a certain driving scenario may include: executing car computer functions A, B, and C in sequence. Through user behavior data, we know that after entering the driving scenario, the user has executed car computer functions A and B in sequence. It is predicted that the user will execute car computer function C next, and the universal button is mapped to car computer function C.

[0052] S3. For each driving scenario, based on the user behavior data collected during the same driving scenario, the user operation habits in the corresponding driving scenario are refined and updated to the scenario habit library.

[0053] In this embodiment, as the vehicle is used, it will experience different driving scenarios and collect corresponding user behavior data. New user operating habits can be extracted from a large amount of user behavior data to update the scenario habit library.

[0054] It is understandable that users' operating habits are not static, and the set of user operating habits in the scenario habit library is not necessarily complete or accurate. By continuously refining user operating habits and updating the scenario habit library, it can better fit the user's current operating habits, improve the accuracy of vehicle function prediction, and improve the user's experience of operating the universal button.

[0055] Specifically, for the same driving scenario, if the newly refined user operation habits are already included in the corresponding user operation habit set, no additional operation will be performed; if the newly refined user operation habits conflict with a user operation habit in the corresponding user operation habit set, the user operation habits will be overwritten by the newly refined user operation habits; if the newly refined user operation habits do not conflict with all user operation habits in the corresponding user operation habit set, the newly refined user operation habits will be added to the corresponding user operation habit set.

[0056] In particular, the user behavior data used to predict the vehicle-mounted functions in step S2 is the user behavior data collected after entering the current driving scenario. For example, if the current driving scenario is entered at time t1, the user behavior data from time t1 to date is used to predict the vehicle-mounted functions. The user behavior data used to refine the user's operating habits in step S3 is the user behavior data collected during the same driving scenario. If the same driving scenario is experienced multiple times, the user behavior data collected during the corresponding duration can be used. For example, if the user is in driving scenario A from t2 to t3 and from t4 to t5, the user's operating habits in driving scenario A can be refined based on the user behavior data collected during t2 to t3 and t4 to t5.

[0057] It should be noted that, based on the current level of development of AI models, the three functions of identifying driving scenarios, predicting the user's next vehicle-computer function, and refining user operating habits in this application can be achieved through targeted training based on mainstream AI models. This application does not limit the specific architecture and training mechanism of the AI ​​model.

[0058] For example, the architectural mechanism of the AI ​​model adopts a hybrid deep learning architecture, including a multimodal Transformer backbone network, a reinforcement learning decision module, and a reinforcement learning component.

[0059] Therefore, through this embodiment, the mapping relationship between the universal button and the vehicle computer function is flexibly set in combination with the current driving scene and user operation habits, and the scene habit library is continuously updated through AI learning to improve the prediction accuracy. While meeting the user's functional operation needs, it will not increase the complexity of functional operation, thereby improving the user experience.

[0060] Furthermore, in one embodiment, the step of predicting the vehicle-mounted function that the user will next execute based on the user's operation habit set in the current driving scenario and the user behavior data collected after entering the current driving scenario, and mapping the universal button to the predicted vehicle-mounted function includes:

[0061] Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the multiple car functions that the user will perform next.

[0062] If the predicted multiple car functions match any car function group defined by the user, the universal button is mapped to the corresponding car function group;

[0063] If the predicted multiple car machine functions do not match all car machine function groups defined by the user, the universal button is mapped to the first predicted car machine function.

[0064] In this embodiment, users can define a series of vehicle-mounted functions as a vehicle-mounted function group as needed. After mapping the universal button to the corresponding vehicle-mounted function group, the user only needs to press the universal button once to execute the corresponding series of vehicle-mounted functions, thereby further reducing user operation complexity and improving user experience. When predicting vehicle-mounted functions, it is necessary to predict multiple vehicle-mounted functions and match them with the vehicle-mounted function group to determine whether to map them individually or in groups.

[0065] It should be noted that the matching is not limited to being exactly the same. For example, the predicted multiple vehicle-machine functions include all the vehicle-machine functions of a certain vehicle-machine function group and other functions. The vehicle-machine function group can be mapped in groups first. In subsequent predictions, if there are no other matching vehicle-machine function groups, the other vehicle-machine functions can be mapped separately in turn.

[0066] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0067] The prediction accuracy of the corresponding driving scenario is calculated based on the predicted vehicle computer functions during each driving scenario and the vehicle computer functions actually performed by the user;

[0068] If the prediction accuracy of the current driving scenario is greater than or equal to the accuracy threshold, the universal button is enabled;

[0069] If the prediction accuracy of the current driving scenario is less than the accuracy threshold, the universal button is disabled.

[0070] Specifically, user behavior data includes the vehicle computer functions actually performed by the user.

[0071] In this embodiment, the activation (enabling) and deactivation (disabling) of the universal button depends on the prediction accuracy of the current driving scenario. Assuming that the current set of user operation habits is inaccurate, the prediction accuracy will be less than the accuracy threshold. At this time, the predicted car function is likely to be inconsistent with the user's expectations. If the user presses the universal button, but the executed car function is not the expected car function, it will affect the user experience. Therefore, the universal button is disabled. Correspondingly, if the prediction accuracy is greater than or equal to the accuracy threshold, it is considered that the predicted car function is likely to be consistent with the user's expectations. The universal button is enabled, so that the user can execute the expected car function by pressing the universal button, thereby improving the user experience.

[0072] Optionally, the activation and deactivation of the universal button can be prompted to the user through lights, display screens, voice, etc. After the user is prompted to activate the universal button, he or she can operate the universal button as needed.

[0073] Optionally, after the user operation habit set is updated, the prediction accuracy under the corresponding driving scenario is recalculated to reduce the impact of historical data.

[0074] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0075] Before the vehicle leaves the factory, a corresponding set of user operating habits is preset for one or more specified driving scenarios and stored in a scenario habit library.

[0076] In this embodiment, a set of user operation habits for specified driving scenarios is preset before the vehicle leaves the factory. The preset content is based on big data statistics and represents common operation habits, which makes it easy for the user to quickly start using the universal button after binding the vehicle.

[0077] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0078] After the user binds the vehicle, the user behavior data in one or more specified driving scenarios is collected through online simulated driving;

[0079] Based on the user behavior data collected in each driving scenario through online simulated driving, the user operation habits in the corresponding driving scenario are extracted and updated to the scenario habit library.

[0080] In this embodiment, after the user binds the vehicle, user behavior data in a specified driving scenario is collected through online simulated driving, and the user's operating habits are refined accordingly, so that the user can quickly start using the universal button after binding the vehicle.

[0081] The method for implementing a universal key based on AI learning also includes:

[0082] After the user binds the vehicle, the user's operating habits in one or more specified driving scenarios are collected through questionnaires and updated to the scenario habit library.

[0083] In this embodiment, after the user binds the vehicle, the user's operating habits in a specified driving scenario are collected through a questionnaire, so that the user can quickly start using the universal button after binding the vehicle.

[0084] Specifically, in the above three embodiments, the designated driving scene may be a typical, high-frequency driving scene. In the latter two embodiments, the designated driving scene may also be specified by the user.

[0085] In addition, if the driving scenarios specified in the latter two embodiments have a preset set of user operation habits before leaving the factory, the user's personalized operation habits will be updated based on the common operation habits, thereby improving the user's experience of using the universal button during actual driving.

[0086] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0087] If the user operation habit set for any driving scenario in the scenario habit library is empty, then after experiencing the corresponding driving scenario and parking, collect user behavior data for the corresponding driving scenario through online simulated driving;

[0088] Based on the user behavior data in the corresponding driving scenarios collected through online simulated driving, the user operation habits in the corresponding driving scenarios are refined and updated to the scenario habit library.

[0089] It can be understood that if a vehicle has experienced a certain driving scenario, the possibility of the vehicle experiencing this driving scenario again is high, and the user operation habit set for this driving scenario is empty. If no additional operations are performed, the vehicle needs to collect sufficient user behavior data during the duration of this driving scenario to extract the user operation habits.

[0090] In this embodiment, for driving scenarios that the vehicle has experienced and for which the user operation habit set is empty, user behavior data in the corresponding driving scenarios is collected through online simulated driving, and user operation habits are refined accordingly, so that the user can quickly start using the universal button after the vehicle enters the corresponding driving scenario again, without having to wait until sufficient user behavior data is collected during actual driving before using the universal button.

[0091] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes:

[0092] If the user operation habit set for any driving scenario in the scenario habit library is empty, after experiencing the corresponding driving scenario and parking, the user operation habits for the corresponding driving scenario are collected through a questionnaire and updated to the scenario habit library.

[0093] In this embodiment, for driving scenarios that the vehicle has experienced and for which the user operation habit set is empty, the user operation habits in the specified driving scenario are collected through a questionnaire, so that the user can quickly start using the universal button after the vehicle enters the corresponding driving scenario again, without having to wait until sufficient user behavior data is collected during the actual driving process before using the universal button.

[0094] It is understandable that the user operating habits collected and refined through online simulated driving are more personalized than the user operating habits collected directly through questionnaires, while the selection range of the latter is relatively limited.

[0095] Furthermore, in one embodiment, the scene habit library is associated with the vehicle computer account, and the scene habit libraries of different vehicle computer accounts are independent of each other.

[0096] In this embodiment, when different users use the same vehicle, they can quickly switch the scene habit library by switching the vehicle computer account. When the same user uses different vehicles, he can quickly configure the scene habit library by logging into his own vehicle computer account, thereby better meeting the user's personalized needs and improving the user experience.

[0097] On the second aspect, the embodiment of the present application also provides a universal key implementation device based on AI learning.

[0098] Figure 3 A schematic diagram of the functional modules of a universal key implementation device based on AI learning in one embodiment of the present application is shown.

[0099] Reference Figure 3 In one embodiment, a universal key implementation device based on AI learning includes:

[0100] The habit query module 10 is used to identify the current driving scene based on the collected scene data and obtain the user operation habit set in the current driving scene from the scene habit library;

[0101] The operation mapping module 20 is used to predict the vehicle function that the user will next perform based on the user's operation habits in the current driving scene and the user behavior data collected after entering the current driving scene, and map the universal button to the predicted vehicle function;

[0102] The habit updating module 30 is used to extract the user operation habits in each driving scenario based on the user behavior data collected during the same driving scenario, and update the user operation habits in the scenario habit library.

[0103] Furthermore, in one embodiment, the operation mapping module 20 is configured to:

[0104] Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the multiple car functions that the user will perform next.

[0105] If the predicted multiple car functions match any car function group defined by the user, the universal button is mapped to the corresponding car function group;

[0106] If the predicted multiple car machine functions do not match all car machine function groups defined by the user, the universal button is mapped to the first predicted car machine function.

[0107] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes a key start / stop module for:

[0108] The prediction accuracy of the corresponding driving scenario is calculated based on the predicted vehicle computer functions during each driving scenario and the vehicle computer functions actually performed by the user;

[0109] If the prediction accuracy of the current driving scenario is greater than or equal to the accuracy threshold, the universal button is enabled;

[0110] If the prediction accuracy of the current driving scenario is less than the accuracy threshold, the universal button is disabled.

[0111] Furthermore, in one embodiment, the method for implementing a universal key based on AI learning further includes a habit preset module for:

[0112] Before the vehicle leaves the factory, a corresponding set of user operating habits is preset for one or more specified driving scenarios and stored in a scenario habit library.

[0113] Furthermore, in one embodiment, the habit updating module 30 is further configured to:

[0114] After the user binds the vehicle, the user behavior data in one or more specified driving scenarios is collected through online simulated driving;

[0115] Based on the user behavior data collected in each driving scenario through online simulated driving, the user operation habits in the corresponding driving scenario are extracted and updated to the scenario habit library.

[0116] Furthermore, in one embodiment, the habit updating module 30 is further configured to:

[0117] After the user binds the vehicle, the user's operating habits in one or more specified driving scenarios are collected through questionnaires and updated to the scenario habit library.

[0118] Furthermore, in one embodiment, the habit updating module 30 is further configured to:

[0119] If the user operation habit set for any driving scenario in the scenario habit library is empty, then after experiencing the corresponding driving scenario and parking, collect user behavior data for the corresponding driving scenario through online simulated driving;

[0120] Based on the user behavior data in the corresponding driving scenarios collected through online simulated driving, the user operation habits in the corresponding driving scenarios are refined and updated to the scenario habit library.

[0121] Furthermore, in one embodiment, the habit updating module 30 is further configured to:

[0122] If the user operation habit set for any driving scenario in the scenario habit library is empty, after experiencing the corresponding driving scenario and parking, the user operation habits for the corresponding driving scenario are collected through a questionnaire and updated to the scenario habit library.

[0123] Furthermore, in one embodiment, the scene habit library is associated with the vehicle computer account, and the scene habit libraries of different vehicle computer accounts are independent of each other.

[0124] Among them, the functional implementation of each module in the above-mentioned AI learning-based universal key implementation device corresponds to the steps in the above-mentioned AI learning-based universal key implementation method embodiment, and its functions and implementation processes will not be repeated here one by one.

[0125] It should be noted that the serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.

[0126] The terms "including" and "having" and any variations thereof in the specification and claims of this application and the above-mentioned drawings are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but optionally includes steps or units that are not listed, or optionally includes other steps or units inherent to these processes, methods, products or devices. The terms "first", "second" and "third" are used to distinguish different objects, etc., and do not represent a sequence, nor do they limit the "first", "second" and "third" to different types.

[0127] In the description of the embodiments of this application, the words "exemplary," "for example," or "for example" are used to indicate examples, illustrations, or descriptions. Any embodiment or design described as "exemplary," "for example," or "for example" in the embodiments of this application should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of words such as "exemplary," "for example," or "for example" is intended to present the relevant concepts in a concrete manner.

[0128] In the description of the embodiments of the present application, unless otherwise specified, “ / ” means or, for example, A / B can mean A or B; “and / or” in the text is merely a description of the association relationship of associated objects, indicating that three relationships may exist, for example, A and / or B can mean: A exists alone, A and B exist at the same time, and B exists alone. In addition, in the description of the embodiments of the present application, “multiple” refers to two or more than two.

[0129] In some processes described in the embodiments of the present application, multiple operations or steps are included that appear in a specific order. However, it should be understood that these operations or steps may not be performed in the order in which they appear in the embodiments of the present application or may be performed in parallel. The sequence numbers of the operations are only used to distinguish between different operations, and the sequence numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations or steps may be performed in sequence or in parallel, and these operations or steps may be combined.

[0130] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus the necessary general hardware platform, of course, it can also be implemented by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for enabling a terminal device to execute the methods described in each embodiment of the present application.

[0131] The above are only preferred embodiments of the present application and do not limit the patent scope of the present application. Any equivalent structure or equivalent process transformation made using the contents of the present application specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A method for realizing a universal key based on AI learning, characterized in that: The method for implementing the universal key based on AI learning includes: Identify the current driving scenario based on the collected scenario data and obtain the user operation habit set under the current driving scenario from the scenario habit library; Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the car function the user will perform next and maps the universal button to the predicted car function. For each driving scenario, based on the user behavior data collected during the same driving scenario, the user operation habits in the corresponding driving scenario are refined and updated to the scenario habit library.

2. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The step of predicting the vehicle computer function that the user will next execute based on the user operation habit set in the current driving scene and the user behavior data collected after entering the current driving scene, and mapping the universal button to the predicted vehicle computer function includes: Based on the user's operating habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, the system predicts the multiple car functions that the user will perform next. If the predicted multiple car functions match any car function group defined by the user, the universal button is mapped to the corresponding car function group; If the predicted multiple car machine functions do not match all car machine function groups defined by the user, the universal button is mapped to the first predicted car machine function.

3. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The method for implementing the universal key based on AI learning also includes: The prediction accuracy of the corresponding driving scenario is calculated based on the predicted vehicle computer functions during each driving scenario and the vehicle computer functions actually performed by the user; If the prediction accuracy of the current driving scenario is greater than or equal to the accuracy threshold, the universal button is enabled; If the prediction accuracy of the current driving scenario is less than the accuracy threshold, the universal button is disabled.

4. The method for realizing a universal key based on AI learning according to claim 1, wherein: The method for implementing the universal key based on AI learning also includes: Before the vehicle leaves the factory, a corresponding set of user operating habits is preset for one or more specified driving scenarios and stored in a scenario habit library.

5. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The method for implementing the universal key based on AI learning also includes: After the user binds the vehicle, the user behavior data in one or more specified driving scenarios is collected through online simulated driving; Based on the user behavior data collected in each driving scenario through online simulated driving, the user operation habits in the corresponding driving scenario are extracted and updated to the scenario habit library.

6. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The method for implementing the universal key based on AI learning also includes: After the user binds the vehicle, the user's operating habits in one or more specified driving scenarios are collected through questionnaires and updated to the scenario habit library.

7. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The method for implementing the universal key based on AI learning also includes: If the user operation habit set for any driving scenario in the scenario habit library is empty, then after experiencing the corresponding driving scenario and parking, collect user behavior data for the corresponding driving scenario through online simulated driving; Based on the user behavior data in the corresponding driving scenarios collected through online simulated driving, the user operation habits in the corresponding driving scenarios are refined and updated to the scenario habit library.

8. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The method for implementing the universal key based on AI learning also includes: If the user operation habit set for any driving scenario in the scenario habit library is empty, after experiencing the corresponding driving scenario and parking, the user operation habits for the corresponding driving scenario are collected through a questionnaire and updated to the scenario habit library.

9. The method for realizing a universal key based on AI learning according to claim 1, characterized in that: The scene habit library is associated with the car computer account, and the scene habit libraries of different car computer accounts are independent of each other.

10. A universal key implementation device based on AI learning, characterized in that: The AI ​​learning-based universal key implementation device includes: The habit query module is used to identify the current driving scenario based on the collected scenario data and obtain the user operation habit set in the current driving scenario from the scenario habit library; The operation mapping module is used to predict the car function that the user will perform next based on the user's operation habits in the current driving scenario and the user behavior data collected after entering the current driving scenario, and map the universal button to the predicted car function; The habit update module is used to extract the user operation habits in each driving scenario based on the user behavior data collected during the same driving scenario, and update them to the scenario habit library.

Images