Vehicle control method, server, and computer-readable storage medium

Abstract

The application discloses a vehicle control method, a server and a computer readable storage medium. The method comprises: receiving perception information. According to the perception information, a planning function group is determined. According to the planning function group and the perception information, a target instruction group is determined. The vehicle is controlled according to the target instruction group. In this way, by analyzing the perception information, the planning function group is determined actively, and the target instruction group is determined according to the planning function group, so that the control of the vehicle is realized, and the user experience is improved.

Description

Technical Field

[0001] This invention relates to the field of vehicle control technology, and in particular to a vehicle control method, a server, and a computer-readable storage medium. Background Technology

[0002] In related technologies, in-vehicle voice assistants facilitate human-computer interaction with users, making operation more convenient. However, these in-vehicle voice assistants can only passively respond after the user issues a command, lacking the ability to proactively provide services or suggestions, thus affecting the user experience. Summary of the Invention

[0003] This application provides a vehicle control method, a server, and a computer-readable storage medium.

[0004] This application provides a vehicle control method, the method comprising: Receive and sense information; Based on the perceived information, the planning function groups are determined; Based on the planning function group and the perceived information, determine the target instruction group; The vehicle is controlled according to the target instruction set.

[0005] In this way, the server receives the perception information. Next, based on the perception information, the server determines the planned function group. Finally, based on the planned function group and the perception information, the server determines the target command group and controls the vehicle according to the target command group. Thus, by analyzing the perception information, proactively determining the planned function group, and then determining the target command group based on that planned function group, vehicle control is achieved, thereby improving the user experience.

[0006] In some implementations, the perceived information includes user perceived information and environmental perceived information, and determining the planning function group based on the perceived information includes: Based on the user perception information, the user intent is determined, and the user perception information includes user voice requests and / or user behavior; Based on a preset functional knowledge base, a first candidate functional group is determined according to the user intent and the environmental perception information; The planning function group is determined based on the first candidate function group.

[0007] Thus, the server determines the user's intent based on user-perceived information, including voice requests and / or user behavior. Next, based on a pre-defined functional knowledge base, the server determines a first candidate function group according to the user's intent and environmental perception information. Finally, the server determines the planned function group based on the first candidate function group. In this way, by analyzing user-perceived and environmental perception information, understanding the user's intent, and determining the planned function group based on a pre-defined functional knowledge base, vehicle control is achieved, improving the user experience.

[0008] In some implementations, the preset function knowledge base includes multiple preset function points and high-frequency voice request statements associated with each preset function point. The step of determining a first candidate function group based on the preset function knowledge base, according to the user intent and the environmental awareness information, includes: When the user perception information includes the user voice request, the instruction fit value of each preset function point relative to the user intent is determined based on the user intent and the high-frequency voice request expression. Based on the user intent, the environmental awareness information, and the preset function points, determine the effective value of each preset function point relative to the user intent; The ranking score of the preset function points is determined based on the instruction fit value and the function validity value. The first candidate functional group is determined based on the ranking score.

[0009] Thus, when the server receives user-perceived information, including user voice requests, it determines the command fit value of each preset function point relative to the user's intent based on the user's intention and the expression of high-frequency voice requests. Next, the server determines the functional validity value of each preset function point relative to the user's intent based on the user's intent, environmental perception information, and the preset function points. Then, the server determines a ranking score for the preset function points based on the command fit value and functional validity value. Finally, the server determines the first candidate function group based on the ranking score. In this way, by analyzing user voice requests and current environmental information, determining the command fit value and functional validity value of each preset function point relative to the user's intent, and determining the first candidate function group based on the ranking score, vehicle control is achieved, improving the user experience.

[0010] In some implementations, determining the instruction fit value of each preset function point relative to the user intent based on the user intent and the high-frequency voice request statement includes: Calculate the similarity between the user intent and the multiple high-frequency voice request expressions to obtain multiple similarity values, wherein the multiple high-frequency voice request expressions are associated with the same preset function point; The maximum value among the multiple similar values ​​is determined as the instruction fitting value.

[0011] In this way, the server calculates the similarity between the user's intent and multiple high-frequency voice request expressions, obtaining multiple similarity values. Among these, the multiple high-frequency voice request expressions are associated with the same preset function. Next, the server determines the maximum value among the multiple similarity values ​​as the instruction fit value. Thus, by calculating the similarity between the user's intent and multiple high-frequency voice request expressions, and selecting the largest similarity value as the instruction fit value, the server accurately determines the degree of matching between the user's intent and the preset function, improving the intelligent cockpit system's understanding ability and user experience.

[0012] In some implementations, determining the first candidate function group based on a preset functional knowledge base, according to the user intent and the environmental awareness information, includes: When the user-perceived information is the user behavior, the effective value of each preset function point relative to the user intention is determined based on the user intention, the environmental perception information, and the preset function points. Based on the effective values ​​of the functions, determine the ranking score of the preset function points; The first candidate functional group is determined based on the ranking score.

[0013] Thus, when the user's perceived information is user behavior, the server determines the effective value of each preset function point relative to the user's intent based on the user's intention, environmental perception information, and preset function points. Next, the server determines a ranking score for each preset function point based on its effective value. Finally, the server determines the first candidate function group based on the ranking score. In this way, by analyzing user behavior and current environmental information, determining the effective value of each preset function point relative to the user's intent, and determining the first candidate function group based on the ranking score, vehicle control is achieved, improving the user experience.

[0014] In some implementations, determining the planning function group based on the first candidate function group includes: Based on the preset security rules and the trigger location information of the user perception information, a second candidate function group is determined according to the first candidate function group; The planning function group is determined based on the second candidate function group.

[0015] Thus, based on preset safety rules and trigger location information from user perception data, the server determines a second candidate function group according to the first candidate function group. Next, the server determines the planning function group based on the second candidate function group. In this way, by filtering the first candidate function group using preset safety rules and trigger location information to form the second candidate function group, and then determining the final planning function group based on the second candidate function group, driving safety is ensured while meeting user needs.

[0016] In some implementations, determining the second candidate function group based on the trigger location information according to the first candidate function group, based on preset security rules and the user-perceived information, includes: The first candidate function group is filtered according to the preset security rules to determine the first temporary execution function group and the first temporary suggestion function group; Based on the trigger location information, the first temporary execution function group is filtered to determine the second temporary execution function group and the second temporary suggestion function group; The execution sub-function group is determined based on the second temporary execution function group; The suggestion sub-function group is determined based on the first temporary suggestion function group and the second temporary suggestion function group; The second candidate function group is determined based on the execution sub-function group and the suggestion sub-function group.

[0017] Thus, the server filters the first candidate function group according to preset security rules, determining a first temporary execution function group and a first temporary suggestion function group. Next, the server filters the first temporary execution function group according to trigger location information, determining a second temporary execution function group and a second temporary suggestion function group. Then, the server determines the execution sub-function group based on the second temporary execution function group. Subsequently, the server determines the suggestion sub-function group based on the first and second temporary suggestion function groups. Finally, the server determines the second candidate function group based on the execution sub-function group and the suggestion sub-function group. In this way, by using preset security rules and trigger location information, the first candidate function group is subdivided into multiple temporary function groups, and the final execution and suggestion sub-function groups are determined based on these temporary function groups, thereby achieving vehicle control, improving user experience, and ensuring driving safety.

[0018] In some implementations, determining the planning function group based on the second candidate function group includes: Based on the historical database, the planning function group is determined according to the second candidate function group.

[0019] Thus, based on the historical database, the server determines the planning function group according to the second candidate function group. In this way, by using the historical database and based on user feedback on the planning function group, the construction of the planning function group is optimized, realizing personalized decision-making and self-learning, thereby improving the user experience.

[0020] In some implementations, the historical database includes an intent and a historical planning function group associated with the intent, the historical planning function group including a historical execution sub-function group and a historical suggestion sub-function group, and the historical database is updated according to the following steps: If the number of times a preset function point in the historical suggestion sub-function group is adopted reaches a first preset adoption threshold, the preset function point will be adjusted to the historical execution sub-function group to update the historical database. If the number of times the execution function point associated with the vehicle control command to be executed reaches the second preset number of adoptions threshold, the preset function point is adjusted to the historical suggestion sub-function group to update the historical database. The execution function point does not exist in the historical execution sub-function group and does not exist in the historical suggestion sub-function group. If the number of rejections of the preset function point in the execution sub-function group reaches a first preset rejection threshold, the preset function point will be moved to the suggestion sub-function group to update the historical database; and / or If the number of rejections of the preset function point in the suggested sub-function group reaches the second preset rejection threshold, the preset function point will be removed from the second candidate function group to update the historical database.

[0021] Thus, the historical database includes intents and historical planning function groups associated with those intents. The historical planning function groups include historical execution sub-function groups and historical suggestion sub-function groups. The historical database is updated according to the following steps: First, if the number of adoptions of a preset function point in the historical suggestion sub-function group reaches a first preset adoption threshold, the preset function point is moved to the historical execution sub-function group to update the historical database. Second, if the number of adoptions of an execution function point associated with a vehicle control command executed for the intent reaches a second preset adoption threshold, the preset function point is moved to the historical suggestion sub-function group to update the historical database. Execution function points do not exist in either the historical execution sub-function group or the historical suggestion sub-function group. Third, if the number of rejections of a preset function point in the execution sub-function group reaches a first preset rejection threshold, the preset function point is moved to the suggestion sub-function group to update the historical database. Fourth, if the number of rejections of a preset function point in the suggestion sub-function group reaches a second preset rejection threshold, the preset function point is removed from the second candidate function group to update the historical database. In this way, by analyzing the number of times users adopt or reject a product, the historical database can be dynamically adjusted to achieve personalized decision-making and self-learning, thereby improving the user experience.

[0022] In some implementations, the perception information includes current vehicle perception information, and determining the target instruction group based on the planning function group and the perception information includes: Based on the current vehicle perception information, determine the first executability of the preset function points in the execution sub-function group one by one; The first preset function point in the execution sub-function group that is executable is identified as the target execution function point. Based on the target execution function point, determine the target execution instruction; The target instruction group is determined based on the target execution instruction.

[0023] In this way, based on the current vehicle perception information, the server determines the first executable function point in each preset function group. Next, the server identifies the first preset function point in the execution sub-function group that is executable as the target execution function point. Then, based on the target execution function point, the server determines the target execution instruction. Finally, based on the target execution instruction, the server determines the target instruction group. Thus, by analyzing the current vehicle perception information, determining executable function points, and generating the target instruction group, vehicle control is achieved, improving the user experience.

[0024] In some implementations, determining the target instruction group based on the planning function group and the perceived information includes: Based on the current vehicle perception information, determine the second executability of the preset function points in the suggested sub-function groups one by one; The preset function point in the suggested sub-function group that is executable in the second executable state is identified as the target suggested function point; Based on the target suggested function points, determine the target suggested instruction; The target instruction group is determined based on the target suggestion instruction.

[0025] In this way, based on the current vehicle perception information, the server determines the second executability of each preset function point in the suggested sub-function group. Next, the server identifies the preset function points in the suggested sub-function group that are executable in the second executability as target suggested function points. Then, based on the target suggested function points, the server determines the target suggested instructions. Finally, based on the target suggested instructions, the server determines the target instruction group. Thus, by analyzing the current vehicle perception information, determining executable suggested function points, and generating target instruction groups, vehicle control is achieved, improving the user experience.

[0026] In some embodiments, the method further includes: If both the first executable and the second executable are unexecutable, feedback information is generated to guide the user to perform clarification operations based on the feedback information.

[0027] Thus, if both the first and second executability criteria are unavailable, the server generates feedback information to guide the user through clarification. In this way, by providing feedback and guiding the user through clarification, the server helps resolve the issue of the function being unexecutable, improving the user experience.

[0028] This application provides a server, which includes a processor and a memory. The memory stores a computer program, and when the computer program is executed by the processor, it implements the vehicle control method described above.

[0029] This application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the steps of the vehicle control method described above.

[0030] Additional aspects and advantages of embodiments of this application will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of embodiments of this application. Attached Figure Description

[0031] The above and / or additional aspects and advantages of this application will become apparent and readily understood from the description of the embodiments taken in conjunction with the following drawings, wherein: Figure 1 This is one of the schematic flowcharts of a vehicle control method according to certain embodiments of this application; Figure 2 This is a second schematic flowchart of a vehicle control method according to certain embodiments of this application; Figure 3 This is a third schematic flowchart of a vehicle control method according to certain embodiments of this application; Figure 4 This is the fourth flowchart of a vehicle control method according to certain embodiments of this application; Figure 5 This is the fifth of the flowcharts illustrating a vehicle control method according to certain embodiments of this application; Figure 6 This is a schematic flowchart of a vehicle control method according to certain embodiments of this application, number six. Figure 7 This is the seventh flowchart of a vehicle control method according to certain embodiments of this application; Figure 8 This is the eighth flowchart of a vehicle control method according to certain embodiments of this application; Figure 9 This is the ninth flowchart of a vehicle control method according to certain embodiments of this application; Figure 10 This is the tenth schematic flowchart of a vehicle control method according to certain embodiments of this application; Figure 11 This is eleventh of the flowcharts illustrating a vehicle control method according to certain embodiments of this application; Figure 12 This is the twelfth schematic flowchart of a vehicle control method according to certain embodiments of this application. Detailed Implementation

[0032] The embodiments of this application are described in detail below. Examples of the embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the embodiments of this application, and should not be construed as limiting the embodiments of this application.

[0033] In today's automotive industry, in-vehicle voice assistants have become a common technology, greatly enhancing user convenience and driving experience through human-computer interaction. However, traditional in-vehicle voice assistant systems typically only respond passively after the user issues a command, meaning they lack the ability to proactively provide services or suggestions, which significantly limits the improvement of the user experience.

[0034] For example, traditional in-vehicle voice assistants lack proactive interaction; they typically do not actively communicate with users, only responding after the user issues a command. This passive interaction method limits the system's flexibility and prevents it from providing more personalized services.

[0035] Lacking personalized services, traditional in-vehicle voice assistants often fail to provide personalized services based on users' individual preferences and habits, and the system cannot provide users with a more considerate and convenient driving experience.

[0036] Based on the above issues, please refer to Figure 1 This application provides a vehicle control method, the method comprising: 01: Receiving and sensing information; 02: Determine the planning function groups based on the perceived information; 03: Determine the target instruction group based on the planned functional groups and perceived information; 04: Control the vehicle according to the target instruction group.

[0037] This application also provides a server, including a memory and a processor. The vehicle control method of this application can be implemented by the server of this application. Specifically, the memory stores a computer program, and the processor is used to receive perception information, determine a planning function group based on the perception information, determine a target instruction group based on the planning function group and the perception information, and control the vehicle according to the target instruction group.

[0038] This application also provides a vehicle control device. The vehicle control method of this application can be implemented by the vehicle control device of this application. Specifically, the vehicle control device includes a receiving module and a determining module. The receiving module is used to receive sensing information. The determining module is used to determine a planned function group based on the sensing information. The determining module is also used to determine a target instruction group based on the planned function group and the sensing information, and control the vehicle according to the target instruction group.

[0039] Specifically, perception information refers to data collected by various sensors and systems in a vehicle, which is used to understand and respond to user needs and the environment around the vehicle, including user perception information and environmental perception information.

[0040] User-perceived information refers to data and information collected by a vehicle through its sensors and systems regarding vehicle users (such as drivers and passengers), including user voice requests and user behaviors. User voice requests refer to requests made by users via voice commands, such as "navigate to the nearest gas station" or "play music." User behaviors refer to user actions and behavioral patterns, such as adjusting the seat, switching driving modes, and using the touchscreen. In some implementations, user-perceived information also includes user physiological information, such as heart rate and facial expressions, which can help the system assess the user's mood and comfort.

[0041] Environmental perception information refers to the data and information collected by a vehicle through its onboard sensors, cameras, and other detection devices regarding the vehicle and its surrounding environment. This includes status perception information of onboard equipment functions and information about the current environmental state of the vehicle. Status perception information involves the functional status of various devices within the vehicle, such as whether information systems like navigation, entertainment, and communication systems are operating, and their current settings and status. It can also include whether comfort systems like air conditioning, heated / ventilated seats, and window controls are activated, and their current temperature, fan speed, and other settings. Furthermore, it can indicate whether driver assistance systems such as adaptive cruise control, lane keeping assist, and automatic parking systems are enabled, and their current parameters and status. Environmental state information includes traffic conditions, weather conditions, and road conditions.

[0042] The planned function group refers to a list of functions that, based on the user's current state and needs, as well as environmental perception and vehicle functional status, are considered to best meet the user's intentions or needs through analysis and judgment. These functions can satisfy the user's current needs and optimize vehicle performance and passenger experience. For example, when the user requests "turn on the ventilation," the system will determine the planned function group as "seat ventilation, air conditioning ventilation, and window ventilation" based on the current environmental perception information.

[0043] User requirements refer to the expectations and instructions from vehicle users regarding vehicle functions, performance, or services in a specific context, as obtained by the server.

[0044] A target instruction group refers to a series of specific instructions determined by the system based on the planned function group and the vehicle's current perception information. These instructions can control various systems and devices in the vehicle to achieve the functional points defined in the planned function group. For example, when the user requests "turn on ventilation", the system will determine the planned function group as "seat ventilation, air conditioning ventilation, and window ventilation" based on the current environmental perception information. Then, based on the current status perception information of the vehicle's devices, it will determine that the target instruction group may be "Execute instruction: turn on driver's seat ventilation; Suggested instruction: turn on passenger seat ventilation".

[0045] The server receives perception information, which is user perception information and environmental perception information collected by the server through various sensors in the vehicle.

[0046] Next, based on the received perception information, the server analyzes and determines the planned function group that needs to be executed in the current situation. For example, when the user requests "turn on ventilation", the server will analyze this information and determine the planned function group as "seat ventilation, air conditioning ventilation and window ventilation".

[0047] Then, based on the planned function groups and perception information, the system further determines specific target instruction groups to control relevant vehicle systems, such as air conditioning, seat heating, and navigation. Continuing the example above, after the server determines that the planned function groups are "seat ventilation, air conditioning ventilation, and window ventilation," it then determines the target instruction groups based on the current vehicle equipment status perception information, which may be "Execute instruction: Turn on driver's seat ventilation; Suggested instruction: Turn on passenger seat ventilation."

[0048] Finally, based on the target instruction set, the server sends instructions to the various control systems of the vehicle to achieve control of the vehicle.

[0049] In summary, in the vehicle control method and server provided in this application, the server receives perception information. Next, the server determines a planning function group based on the perception information. Finally, the server determines a target instruction group based on the planning function group and the perception information, and controls the vehicle according to the target instruction group. Thus, by analyzing the perception information, actively determining the planning function group, and determining the target instruction group based on the planning function group, vehicle control is achieved, thereby improving the user experience.

[0050] Please see Figure 2 In some implementations, the perceived information includes user perceived information and environmental perceived information. Step 02 (determining the planning function group based on the perceived information) includes: 021: Determine user intent based on user-perceived information; 022: Based on a pre-set functional knowledge base, determine the first candidate functional group according to user intent and environmental awareness information; 023: Based on the first candidate functional group, determine the planning functional group.

[0051] In some implementations, the determining module is further configured to determine user intent based on user perception information, and to determine a first candidate function group based on a preset functional knowledge base, user intent, and environmental perception information, and to determine a planning function group based on the first candidate function group.

[0052] In some implementations, the processor is further configured to determine user intent based on user-perceived information; and to determine a first candidate function group based on a preset functional knowledge base, user intent, and environmental perception information; and to determine a planning function group based on the first candidate function group.

[0053] Specifically, the preset functional knowledge base refers to a database that includes various functions and their associated high-frequency voice request expressions. It provides the information foundation needed for the support system's decision-making and operations, namely, for processing user needs and determining the first candidate functional group. By analyzing user-perceived information and environmental-perceived information, the server can determine the most suitable functional point from the preset knowledge base and generate corresponding operation instructions.

[0054] User voice requests refer to specific requests or commands that users convey to the vehicle via voice commands. These can be simple operation instructions, such as "turn on the air conditioning" or "play music," or more complex queries or requests, such as "where is the nearest gas station?" or "find me a good restaurant."

[0055] User behavior refers to the various operations and activities performed by users inside the vehicle, including touching the screen, pressing buttons, adjusting the seat, etc.

[0056] The first candidate function group refers to a set of function points initially selected from a preset function knowledge base based on the user's intent and environmental awareness information. These function points are considered to be the most likely options to meet the user's current needs.

[0057] The first step for an intelligent cockpit system is to understand the user's voice requests and actions to determine the user's intent. Continuing the example above, based on the user's voice request "turn on the ventilation," the system determines the user's intent to be "turn on the ventilation."

[0058] Next, based on a preset functional knowledge base, the server filters possible combinations of functions from the knowledge base according to user intent and environmental awareness information, forming the first candidate functional group. Continuing the above example, based on the user's request to "turn on ventilation" and the preset functional knowledge base, the first candidate functional group is determined to be "seat ventilation, air conditioning ventilation, window ventilation, sunroof ventilation, and air purifier ventilation".

[0059] Finally, the server further determines the final planned function group based on the first candidate function group. Continuing the example above, the server determines the planned function group as "seat ventilation, air conditioning ventilation, window ventilation, sunroof ventilation, and air purifier ventilation" based on the first candidate function group "seat ventilation, air conditioning ventilation, and window ventilation".

[0060] In this way, by analyzing user perception information and environmental perception information, the system can understand user intentions and determine the planned function groups based on a pre-set functional knowledge base, thereby achieving vehicle control and improving the user experience.

[0061] Please see Figure 3In some implementations, the preset function knowledge base includes multiple preset function points and high-frequency voice request statements associated with each preset function point. Step 022 (based on the preset function knowledge base, determining the first candidate function group according to user intent and environmental awareness information) includes: 0221: When the user's perceived information includes the user's voice request, determine the instruction fit value of each preset function point relative to the user's intent based on the user's intent and the expression of high-frequency voice requests; 0222: Based on user intent, environmental awareness information, and preset function points, determine the effective value of each preset function point relative to the user intent; 0223: Determine the ranking and scoring of preset function points based on instruction fit value and function validity value; 0224: Determine the first candidate functional group based on the ranking score.

[0062] In some implementations, the determining module is further configured to, when the user-perceived information includes a user voice request, determine the instruction fit value of each preset function point relative to the user's intent based on the user's intent and the high-frequency voice request expression; and determine the functional validity value of each preset function point relative to the user's intent based on the user's intent, environmental perception information, and the preset function points. The determining module is also configured to determine a ranking score for the preset function points based on the instruction fit value and the functional validity value; and determine a first candidate function group based on the ranking score.

[0063] In some implementations, the processor is further configured to, when user-perceived information includes a user voice request, determine an instruction fit value for each preset function point relative to the user's intent based on the user's intent and the high-frequency voice request expression; and determine a functional validity value for each preset function point relative to the user's intent based on the user's intent, environmental perception information, and the preset function points; and determine a ranking score for the preset function points based on the instruction fit value and the functional validity value, and determine a first candidate function group based on the ranking score.

[0064] Specifically, high-frequency voice request expressions refer to specific words or phrases that users frequently use during voice interaction to express operational instructions or requests for vehicle system functions. For example, high-frequency voice request expressions associated with the "turn on the air conditioning" function might include "cooling," "blowing," or "turn on the air conditioning."

[0065] The command fit value measures the degree of matching between a user's command and a pre-defined function. When the server receives a user's voice command, it compares the command with pre-defined function points in the pre-defined function knowledge base to determine which function best matches the user's intent. Specifically, it refers to the "text similarity" between the user's command and the high-frequency voice request statements of the vehicle's functions, measuring the degree of matching between the user's command and the functions recognized by the system. The value ranges from 0 to 1, with higher values ​​indicating a higher degree of matching between the command and the function. The command fit value is usually calculated based on the semantic content and contextual information of the user's voice request. A high command fit value means that the user's voice request matches a certain function very well, and therefore that function is likely the operation the user wants to perform.

[0066] The functional effectiveness value measures the degree to which a preset function meets user needs, and can be used to evaluate the applicability and effectiveness of the preset function in the current vehicle state and environment. Specifically, it refers to the "effectiveness" of achieving user intent or user needs, and is used to measure how well a function meets user needs. The value range is 0.5-1, with higher values ​​indicating more effective function points and better ability to meet user needs. The server will determine whether a function point can be effectively executed and achieve the expected effect based on the vehicle's current perception information (such as vehicle speed, temperature, weather, etc.). For example, if the vehicle is traveling at high speed, the functional effectiveness value of the "open the window for ventilation" function point may be low, because opening the window at high speed may generate greater wind noise and safety hazards.

[0067] The ranking score refers to a score obtained by comprehensively evaluating candidate functions based on multiple dimensions during the construction of the first candidate function group. It is used to determine the ranking of the selected preset function points within the first candidate function group. In some implementations, the ranking score is calculated as follows: Where T represents the command fit value, The linear weights representing the command fit values, The power weight of the instruction fit value is represented by N, where N represents the effective value of the function. Linear weights representing the effective value of the function. This represents the power-law weight of the effective value of the function. When the weight of a certain dimension is 0, the corresponding value is 1 (no gain).

[0068] It should be noted that in some implementations, the ranking of the first candidate functional group can also be obtained directly through order fusion, that is, focusing directly on the ranking without caring about the specific numerical value, and sorting and fusion are performed through the ranking dimension. This method is suitable for handling scenarios where only ranking is needed and precise scoring is not required. The advantage of this method is that it can get rid of the influence of scale differences between sub-modules, but the disadvantage is that it loses a lot of information.

[0069] When the user's perceived information includes voice requests, the system will calculate the instruction fit value of each preset function point relative to the user's intent based on the user's intent and the expression of high-frequency voice requests. This instruction fit value reflects the degree of matching between the user's request and the function point.

[0070] Next, the server calculates the effective value of each preset function point relative to the user's intent based on the user's intent, environmental awareness information, and preset function points. This effective value reflects the effectiveness of each function point in meeting the user's needs in the current environment.

[0071] Then, the server combines the instruction fit value and the function validity value to calculate a ranking score for each preset function point.

[0072] Finally, based on the ranking scores, the server selects the top-scoring preset function points from the preset function points to form the first candidate function group.

[0073] In this way, by analyzing the user's voice request and the current environment information, the command fit value and function validity value of each preset function point relative to the user's intention are determined, and the first candidate function group is determined according to the ranking score, thereby realizing the control of the vehicle and improving the user experience.

[0074] Please see Figure 4 In some implementations, step 0221 (determining the instruction fit value of each preset function point relative to the user's intent based on the user's intent and the high-frequency voice request expression) includes: 02211: Calculate the similarity between the user's intent and multiple high-frequency voice request expressions to obtain multiple similarity values; 02212: Determine the maximum value among multiple similar values ​​as the instruction fitting value.

[0075] In some implementations, the determining module is further configured to calculate the similarity between the user's intent and multiple high-frequency voice request expressions, obtaining multiple similarity values, and determining the maximum value among the multiple similarity values ​​as the instruction fit value.

[0076] In some implementations, the processor is also configured to calculate the similarity between the user's intent and multiple high-frequency voice request expressions, obtaining multiple similarity values, and determining the maximum value among the multiple similarity values ​​as the instruction fit value.

[0077] Specifically, associating multiple high-frequency voice request expressions with the same preset function point means that each preset function point can be associated with multiple high-frequency voice request expressions. For example, the "turn on the air conditioner" function can be associated with multiple high-frequency voice request expressions such as "cooling", "blowing", and "turn on the air conditioning".

[0078] For each preset function, the server uses natural language processing techniques to calculate the semantic similarity between all associated high-frequency voice request statements and the user's intent. For example, algorithms such as the bag-of-words model, TF-IDF, and neural networks are used to analyze the text content of the user's voice request and the text content of high-frequency voice request statements.

[0079] Next, the server selects the largest similarity value from these values ​​as the instruction fit value for the preset function. This instruction fit value represents the similarity between the user's intent and the most frequently used voice request expression that best matches the preset function.

[0080] In this way, by calculating the similarity between user intent and multiple high-frequency voice request expressions, and selecting the largest similarity value as the instruction fit value, the degree of matching between user intent and preset function points can be accurately determined, thereby improving the understanding ability and user experience of the intelligent cockpit system.

[0081] Please see Figure 5 In some implementations, step 022 (determining the first candidate function group based on a preset function knowledge base, user intent, and environmental awareness information) includes: 0224: When the user's perceived information is user behavior, determine the effective value of each preset function point relative to the user's intent based on the user's intent, environmental perception information, and preset function points. 0225: Determine the ranking and scoring of preset function points based on the effective function values; 0226: Determine the first candidate functional group based on the ranking score.

[0082] In some implementations, the determining module is used to determine the effective value of each preset function point relative to the user's intent, based on the user's intent, environmental awareness information, and preset function points, when the user's perceived information is user behavior. It also determines a ranking score for the preset function points based on the effective function values, and determines a first candidate function group based on the ranking score.

[0083] In some implementations, the processor is further configured to, when the user-perceived information is a user behavior, determine the effective value of each preset function point relative to the user's intent based on the user's intent, environmental awareness information, and preset function points; determine a ranking score for the preset function points based on the effective function values; and determine a first candidate function group based on the ranking score.

[0084] Specifically, when the user-perceived information is user behavior, i.e., excluding user voice requests, the server determines the effective value of each preset function point relative to the user's intent based on the user's intent, environmental awareness information, and preset function points. Next, the server directly determines the ranking score of the preset function points based on the effective function values. Finally, the server determines the first candidate function group based on the ranking score.

[0085] In this way, by analyzing user behavior and current environmental information, the effective value of each preset function point relative to the user's intent is determined, and the first candidate function group is determined based on the ranking score, thereby realizing vehicle control and improving user experience.

[0086] Please see Figure 6 In some implementations, step 023 (determining the planning function group based on the first candidate function group) includes: 0231: Based on the trigger location information of preset security rules and user perception information, determine the second candidate function group according to the first candidate function group; 0232: Determine the planning function group based on the second candidate function group.

[0087] In some implementations, the determining module is further configured to determine a second candidate function group based on the trigger location information of preset security rules and user-perceived information, according to the first candidate function group, and to determine a planning function group based on the second candidate function group.

[0088] In some implementations, the processor is further configured to determine a second candidate function group based on the trigger location information of preset security rules and user-perceived information, according to the first candidate function group, and to determine a planning function group based on the second candidate function group.

[0089] Specifically, preset safety rules refer to built-in safety standards and restrictions used to ensure that vehicle functions comply with safety standards and avoid potential dangerous situations. That is, when constructing and planning function groups, it is necessary to consider which preset functions might pose a risk to user driving safety or startle the user. By classifying functions with safety risks into a suggestion set, the likelihood of users being disturbed or startled while driving can be effectively reduced, ensuring driving safety. For example, suddenly activating the seat massage function while driving at high speed, or prohibiting functions such as making phone calls and watching videos while the vehicle is in motion. These functions will be directly classified into the suggestion set, and then a decision on whether to provide a suggestion will be made based on other weighting or other data information.

[0090] User-perceived information, specifically trigger location information, refers to how features are prioritized for users in the specific trigger location; users in other locations are more likely to receive suggestions. Based on this trigger location information, more personalized feature recommendations can be provided. For example, when a user is sitting in the front passenger seat, they are more likely to activate the front passenger seat massage function, thus improving the user experience.

[0091] The server first filters the first candidate function group according to preset safety rules. Continuing the example above, when the vehicle is currently traveling at high speed, opening the sunroof may generate significant wind noise and safety hazards, so the preset function of sunroof ventilation is excluded.

[0092] Next, consider the trigger location information for user perception information, i.e., the user's position in the vehicle (such as the driver's seat, the front passenger seat, etc.). Continuing the example above, if the server detects that the driver or passenger's position is far from the sunroof, it may prioritize other functions that are closer to the user, such as seat ventilation or air conditioning ventilation, instead of considering sunroof ventilation.

[0093] In this way, by pre-setting safety rules and triggering location information, the first candidate function group is filtered to form the second candidate function group, and the final planning function group is determined based on the second candidate function group, thereby ensuring driving safety and meeting user needs.

[0094] Please see Figure 7 In some implementations, step 0231 (determining a second candidate function group based on the trigger location information of preset security rules and user-perceived information, according to the first candidate function group) includes: 02311: Based on preset security rules, the first candidate function group is screened to determine the first temporary execution function group and the first temporary suggestion function group; 02312: Based on the trigger location information, the first temporary execution function group is filtered to determine the second temporary execution function group and the second temporary suggestion function group; 02313: Based on the second temporary execution function group, determine the execution sub-function group; 02314: Determine the suggestion sub-function groups based on the first temporary suggestion function group and the second temporary suggestion function group; 02315: Determine the second candidate function group based on the execution sub-function group and the suggestion sub-function group.

[0095] In some embodiments, the determining module is further configured to filter the first candidate function group according to preset security rules to determine a first temporary execution function group and a first temporary suggestion function group; and to filter the first temporary execution function group according to trigger location information to determine a second temporary execution function group and a second temporary suggestion function group. The determining module is further configured to determine an execution sub-function group based on the second temporary execution function group; and to determine a suggestion sub-function group based on the first and second temporary suggestion function groups; and to determine a second candidate function group based on the execution sub-function group and the suggestion sub-function group.

[0096] In some embodiments, the processor is further configured to filter the first candidate function group according to preset security rules to determine a first temporary execution function group and a first temporary suggested function group; and to filter the first temporary execution function group according to trigger location information to determine a second temporary execution function group and a second temporary suggested function group. The processor is further configured to determine an execution sub-function group based on the second temporary execution function group; and to determine a suggested sub-function group based on the first and second temporary suggested function groups; and to determine a second candidate function group based on the execution sub-function group and the suggested sub-function group.

[0097] Specifically, the server filters the first candidate function group according to preset security rules to determine the first temporary execution function group and the first temporary suggested function group. Continuing the example above, the first candidate function group "seat ventilation, air conditioning ventilation, window ventilation, sunroof ventilation, and air purifier ventilation" is filtered according to preset security rules. Since the vehicle is traveling at high speed, window ventilation is considered to require further user confirmation, and sunroof ventilation is considered unsafe. Therefore, the first temporary execution function group "seat ventilation, air conditioning ventilation, and air purifier ventilation" and the first temporary suggested function group "window ventilation" are determined.

[0098] Next, the server filters the first temporary execution function group based on the trigger location information to determine the second temporary execution function group and the second temporary suggested function group. Continuing the example above, seat ventilation may be directly executed because the user is sitting directly in the seat, while air conditioning ventilation and air purifier ventilation may be suggested for execution because they affect the overall interior environment of the vehicle. Therefore, the second temporary execution function group is determined to be "seat ventilation" and the second temporary suggested function group is "air conditioning ventilation and air purifier ventilation".

[0099] Then, the server determines the execution sub-function group based on the second temporary execution function group. Continuing the example above, the second temporary execution function group "seat ventilation" is determined as the execution sub-function group "seat ventilation".

[0100] Subsequently, the server determines the suggested sub-function groups based on the first temporary suggested function group and the second temporary suggested function group. Continuing the example above, based on the first temporary suggested function group "window ventilation" and the second temporary suggested function group "air conditioning ventilation and air purifier ventilation", the suggested sub-function group is determined to be "air conditioning ventilation, window ventilation and air purifier ventilation".

[0101] Finally, the server determines the second candidate function group based on the executed sub-function group and the suggested sub-function group. Continuing the example above, based on the executed sub-function group "seat ventilation" and the suggested sub-function group "air conditioning ventilation, window ventilation and air purifier ventilation", the second candidate function group is determined to be "seat ventilation (executed), air conditioning ventilation (suggested), window ventilation (suggested) and air purifier ventilation (suggested)".

[0102] In this way, by pre-setting safety rules and trigger location information, the first candidate function group is subdivided into multiple temporary function groups, and the final execution sub-function group and suggested sub-function group are determined based on the temporary function groups, thereby achieving vehicle control, improving user experience, and ensuring driving safety.

[0103] Please see Figure 8 In some implementations, step 0232 (determining the planning function group based on the second candidate function group) includes: 02321: Based on the historical database, determine the planning function group according to the second candidate function group.

[0104] In some implementations, the determining module is also used to determine the planning function group based on a historical database and according to the second candidate function group.

[0105] In some implementations, the processor is also configured to determine the planning function group based on a historical database and according to a second candidate function group.

[0106] Specifically, a historical database refers to a database that stores users' past usage habits, preferences, and behaviors. This database includes the user's intentions and the historical planned function groups associated with these intentions, that is, the combination of functions that the user has used in different situations, recording the frequency, preference, and time of user function usage.

[0107] The server first accesses the historical database, which contains information such as the user's past interaction records, feature usage preferences, and common operation patterns.

[0108] Next, the server analyzes and matches each function in the second candidate function group with data in the historical database.

[0109] Finally, based on the analysis results of historical data, the server selects the functions that best match the user's historical preferences and behavioral patterns from the second candidate function group to form the planned function group.

[0110] Continuing with the example above, if the database shows that the user always refuses the server's suggested air purifier ventilation function whenever they need to turn on the ventilation, then the preset function "air purifier ventilation" will be removed from the second candidate function group. Therefore, the final planned function group is "seat ventilation (executed), air conditioning ventilation (recommended), and window ventilation (recommended)".

[0111] In this way, by leveraging historical databases and analyzing user feedback on planning function groups, the construction of these groups can be optimized, enabling personalized decision-making and self-learning, thereby enhancing the user experience.

[0112] Please see Figure 9 In some implementations, the historical database includes intents and historical planning function groups associated with those intents. The historical planning function groups include historical execution sub-function groups and historical suggestion sub-function groups. The historical database is updated according to the following steps: 023211: If the number of times a preset function point in the historical suggestion sub-function group is adopted reaches the first preset adoption threshold, the preset function point will be moved to the historical execution sub-function group to update the historical database. 023212: If the number of times the execution function point associated with the intended vehicle control command is adopted reaches the second preset adoption threshold, the preset function point will be adjusted to the historical suggestion sub-function group to update the historical database. 023213: If the number of rejections of a preset function point in a sub-function group reaches a first preset rejection threshold, the preset function point will be moved to a suggested sub-function group to update the historical database; and / or 023214: If the number of rejections of a preset function point in the suggested sub-function group reaches the second preset rejection threshold, the preset function point will be removed from the second candidate function group to update the historical database.

[0113] In some embodiments, the vehicle control device further includes a database update module. This module is configured to update the historical database by adjusting a preset function point to a historical execution sub-function group when the number of times a preset function point in a historical suggested sub-function group has been adopted reaches a first preset adoption threshold. It also updates the historical database by adjusting a preset function point to a historical suggested sub-function group when the number of times an execution function point associated with an intended vehicle control command has been adopted reaches a second preset adoption threshold. The database update module is further configured to update the historical database by adjusting a preset function point to a suggested sub-function group when the number of times a preset function point in an execution sub-function group has been rejected reaches a first preset rejection threshold. Finally, it removes a preset function point from a second candidate function group when the number of times a preset function point in a suggested sub-function group has been rejected reaches a second preset rejection threshold.

[0114] In some embodiments, the processor is further configured to, when the number of times a preset function point in the historical suggested sub-function group has been adopted reaches a first preset adoption threshold, adjust the preset function point to the historical execution sub-function group to update the historical database. And when the number of times an execution function point associated with an intended vehicle control command has been adopted reaches a second preset adoption threshold, the processor is further configured to, when the number of times a preset function point in the execution sub-function group has been rejected reaches a first preset rejection threshold, adjust the preset function point to the suggested sub-function group to update the historical database. And when the number of times a preset function point in the suggested sub-function group has been rejected reaches a second preset rejection threshold, the processor is further configured to, when the number of times a preset function point in the suggested sub-function group has been rejected reaches a second preset rejection threshold, remove the preset function point from the second candidate function group to update the historical database.

[0115] Specifically, firstly, if the number of times a preset function point in the historical suggestion sub-function group is adopted reaches a first preset adoption threshold, the server will consider the function important to the user and will move the function point to the historical execution sub-function group. That is, when a similar scenario occurs again, the system will directly execute the function without requiring user confirmation. For example, if a user frequently says "I'm hot" when entering the vehicle, the system, based on semantic understanding, will recommend the "turn on the air conditioning" function to the user. Historical data: The historical database records the user's multiple adoptions of the "turn on the air conditioning" suggestion. Update: "Turn on the air conditioning" is moved from the "suggestion" group to the "execution" group. The next time the user enters the vehicle and says "I'm hot," the system will directly turn on the air conditioning without requiring user confirmation.

[0116] Second, if the number of times an execution function point associated with a vehicle control command executed for a specific intent is adopted reaches a second preset adoption threshold, then that function point is moved to the historical suggestion sub-function group. For example, if a user says "I'm tired" after a long drive, the system, based on semantic understanding, recommends the "seat massage" function to the user. Historical data: The historical database records the user's behavior of selecting the "turn on music" function after receiving the "seat massage" suggestion. Update: "Turn on music" is moved from the "candidate function group" to the "suggestion" group. The next time a user enters the vehicle and says "I'm tired," the system will recommend the "turn on music" function.

[0117] Third, if the number of rejections of a preset function point in a sub-function group reaches the first preset rejection threshold, the function point will be moved to the suggestion sub-function group. For example, if a user says "close the window" when it's raining, the system will directly execute the "close the window" function based on semantic understanding. Historical data: The historical database records multiple instances where users closed and then reopened the windows. Update: "Close the window" has been moved from the "Execute" group to the "Suggest" group. The next time a user enters the vehicle and says "close the window," the system will ask the user if they really need to close the window.

[0118] Fourth, if the number of rejections of a preset feature in a suggested sub-function group reaches the second preset rejection threshold—that is, if the user repeatedly ignores the system's recommended features—the system will consider the feature unimportant to the user and remove it from the second candidate feature group. For example, if a user frequently says "I'm cold" in cold weather, the system, based on semantic understanding, will recommend the "seat heating" feature to the user. Historical Data: The historical database records the user's repeated ignoring of the "seat heating" suggestion. Update: Remove "seat heating" from the "suggestions" group. The next time the user enters the vehicle and says "I'm cold," the system will no longer recommend the "seat heating" feature.

[0119] In this way, by analyzing the number of times users adopt or reject a product, the historical database can be dynamically adjusted to achieve personalized decision-making and self-learning, thereby improving the user experience.

[0120] Please see Figure 10 In some implementations, the perception information includes current vehicle perception information. Step 03 (determining the target instruction group based on the planning function group and the perception information) includes: 031: Based on the current vehicle perception information, determine the first executable capability of the preset function points in the execution sub-function group one by one; 032: The first preset function point in the execution sub-function group that is executable is identified as the target execution function point; 033: Determine the target execution instructions based on the target execution function points; 034: Determine the target instruction group based on the target execution instructions.

[0121] In some implementations, the determining module is further configured to determine the first executability of preset function points in the execution sub-function group one by one based on the current vehicle perception information, and to determine the first preset function point in the execution sub-function group whose first executability is executable as the target execution function point. The determining module is further configured to determine the target execution instruction based on the target execution function point, and to determine the target instruction group based on the target execution instruction.

[0122] In some implementations, the processor is further configured to determine, based on current vehicle perception information, the first executableness of preset function points in each execution sub-function group, and to determine the first preset function point in the execution sub-function group whose first executableness is determined as the target execution function point. The processor is further configured to determine a target execution instruction based on the target execution function point, and to determine a target instruction group based on the target execution instruction.

[0123] Specifically, current vehicle perception information refers to the status information of various devices within the vehicle that it obtains through sensors and other equipment.

[0124] Target execution instructions refer to specific instructions determined and generated based on the user's intent, the vehicle's current state, and environmental perception information, used to control the vehicle's relevant systems or equipment to perform specific operations.

[0125] Based on the current vehicle perception information, determine whether each preset function point in the execution sub-function group is executable. Execution depends on the current vehicle state and environment; for example, if the interior temperature is already very low, the "turn on the air conditioning" function cannot be executed. Find the first executable preset function point in the execution sub-function group and designate it as the target execution function point.

[0126] Next, based on the target function point, determine the instructions required to execute that function. Finally, add the target execution instructions to the target instruction group to form the final target instruction group.

[0127] Continuing the example above, the vehicle obtains the following current vehicle perception information from the sensors: "Seat occupancy information: Driver and front passenger are occupied; Wake-up voice zone: Driver's seat; Air conditioning status: Off; Window status: Open; Seat ventilation status: Off." Since "Seat ventilation" is the first item in the planned function group and is currently off, the target execution function point is determined to be "Seat ventilation." Subsequently, based on the target execution function point "Seat ventilation," the target execution command is determined to be "Turn on driver's seat ventilation."

[0128] In this way, by analyzing the current vehicle perception information, executable function points are determined and target instruction sets are generated, thereby achieving vehicle control and improving the user experience.

[0129] Please see Figure 11 In some implementations, step 03 (determining the target instruction group based on the planned function group and the perceived information) includes: 035: Based on the current vehicle perception information, determine the second executability of the preset function points in the suggested sub-function groups one by one; 036: The preset function point with the second executable status in the suggested sub-function group is identified as the target suggested function point; 037: Based on the target suggested function points, determine the target suggested instructions; 038: Determine the target instruction group based on the target suggestion instructions.

[0130] In some implementations, the determining module is further configured to determine the second executability of preset function points in the suggested sub-function groups one by one based on the current vehicle perception information, and to determine the preset function points in the suggested sub-function groups that have the second executability as executable as target suggested function points. The determining module is further configured to determine target suggested instructions based on the target suggested function points, and to determine target instruction groups based on the target suggested instructions.

[0131] In some implementations, the processor is further configured to determine, based on current vehicle perception information, the second executableness of preset function points in the suggested sub-function groups one by one, and to determine the preset function points in the suggested sub-function groups that have the second executableness as target suggested function points. The processor is further configured to determine target suggested instructions based on the target suggested function points, and to determine target instruction groups based on the target suggested instructions.

[0132] Specifically, target suggestion instructions refer to specific instructions determined and generated based on the user's intent, the vehicle's current state, and environmental perception information, used to suggest that the relevant systems or devices controlling the vehicle perform specific operations.

[0133] Based on the current vehicle perception information, determine whether each preset function point in the suggested sub-function group is executable. Execution depends on the current vehicle status and environment; for example, if the interior temperature is already very low, the "turn on the air conditioning" function cannot be executed. Find the first executable preset function point in the suggested sub-function group and designate it as the target suggested function point.

[0134] Next, based on the target suggested function points, the instructions required to execute the function are determined, and target suggested instructions are generated. Finally, the target suggested instructions are added to the target instruction group to form the final target instruction group.

[0135] Continuing the example above, the vehicle obtains the following current vehicle perception information from the sensors: "Seat occupancy information: Driver and front passenger are occupied; Wake-up voice zone: Driver's seat; Air conditioning on status: Off; Window open status: Open; Seat ventilation status: Off." Since the first suggested sub-function is "Air conditioning on" and it is currently off, the target suggested function is determined to be "Air conditioning ventilation." Subsequently, based on the target suggested function "Air conditioning ventilation," the target suggested instruction is determined to be "Turn on air conditioning recirculation."

[0136] Ultimately, the target instruction set was determined to be "Turn on driver's seat ventilation (execute); turn on air conditioning recirculation (recommendation)".

[0137] In this way, by analyzing the current vehicle perception information, the executable suggested function points are determined and the target instruction group is generated, thereby realizing the control of the vehicle and improving the user experience.

[0138] Please see Figure 12 In some implementations, the method further includes: 039: If both the first and second executability conditions are not executable, generate feedback information to guide the user to perform clarification operations based on the feedback information.

[0139] In some embodiments, the vehicle control device further includes a generation module for generating feedback information to guide the user to perform clarification operations based on the feedback information if both the first and second executability are not executable.

[0140] In some implementations, the processor is also configured to generate feedback information if both the first and second executability are not executable, in order to guide the user to perform clarification operations based on the feedback information.

[0141] Specifically, feedback information refers to the information provided by the system to the user, which informs the user of the system's understanding and response to their request or operation. It can guide the user to clarify or rephrase their request if the user's request is not correctly understood.

[0142] When both the first and second executability criteria are not executable, it indicates that the system cannot determine the user's needs based on the current information, and further clarification from the user is required. The system can provide various ways to guide the user in this clarification, such as: voice interaction: asking the user what operation they need to perform; text prompts: prompting the user to input commands or select functions; or visual displays: showing a list of available functions for the user to choose from.

[0143] For example, if a user is drowsy in the car, the intelligent cockpit system will determine the functions of "turn on the ventilation system" and "play refreshing music" based on the user's behavior and the current environmental information, forming the first candidate function group.

[0144] Based on preset safety rules and trigger location information, a second candidate function group is formed, including the functions of "turn on the ventilation system" and "play refreshing music". Based on the historical database, "play refreshing music" is elevated to the execution sub-function group. Based on current vehicle perception information, it is determined that neither "turn on the ventilation system" nor "play refreshing music" can be executed, for example, if the vehicle is in motion and the interior temperature is already very low. The system generates feedback information, prompting the user, "Ventilation or music playback functions cannot be executed at this time. What would you like to do?" The user can reply via voice or text, "I want to adjust the seat position," and the system will execute the corresponding operation according to the user's command.

[0145] In this way, by providing feedback and guiding users to clarify their actions, we can help them resolve issues where functions cannot be executed, thereby improving the user experience.

[0146] This application also provides a computer-readable storage medium having a computer program stored thereon. When the computer program is executed by a processor, it implements the steps of the vehicle control method described above.

[0147] It is understood that a computer program includes computer program code. Computer program code can be in the form of source code, object code, executable files, or some intermediate form. Computer-readable storage media can include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), and software distribution media, etc.

[0148] In this specification, the terms "specifically," "furthermore," "particularly," "understandably," etc., refer to specific features, structures, materials, or characteristics described in connection with embodiments or examples that are included in at least one embodiment or example of this application. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0149] Any process or method description in the flowchart or otherwise herein can be understood as representing a module, segment, or portion of executable request code comprising one or more steps for implementing a particular logical function or process, and the scope of the preferred embodiments of this application includes additional implementations in which functions may be performed not in the order shown or discussed, including substantially simultaneously or in reverse order according to the functions involved, as should be understood by those skilled in the art to which embodiments of this application pertain.

[0150] Although embodiments of this application have been shown and described above, it is understood that the above embodiments are exemplary and should not be construed as limiting this application. Those skilled in the art can make changes, modifications, substitutions and variations to the above embodiments within the scope of this application.

Claims

1. A vehicle control method, characterized in that, The method includes: Receive sensing information, including user sensing information, environmental sensing information, and current vehicle sensing information; Based on the user's perceived information, determine the user's intent; Based on the user intent and the environmental awareness information, a first candidate function group is selected from the preset function knowledge base. The preset function knowledge base includes multiple preset function points and high-frequency voice request expressions associated with each preset function point. Based on preset security rules and the trigger location information of the user perception information, the first candidate function group is filtered to determine the second candidate function group, which includes an execution sub-function group and a suggestion sub-function group. Based on the second candidate function group, determine the planning function group; Based on the current vehicle perception information, determine the first executability of each preset function point in the execution sub-function group; and determine the first preset function point in the execution sub-function group whose first executability is executable as the target execution function point; and determine the target execution instruction based on the target execution function point; and / or Based on the current vehicle perception information, determine the second executability of each preset function point in the suggested sub-function group; and determine the preset function points in the suggested sub-function group whose second executability is executable as target suggested function points; and determine the target suggested instruction based on the target suggested function points. Determine the target instruction group based on the target execution instruction and / or the target suggestion instruction; The vehicle is controlled according to the target instruction set.

2. The method according to claim 1, characterized in that, The user-perceived information includes user voice requests and / or user non-voice operation information; the preset function knowledge base includes high-frequency voice request statements associated with each preset function point; and the step of filtering and determining a first candidate function group from the preset function knowledge base based on the user intent and the environmental perception information includes: When the user perception information includes the user voice request, based on the user intent and the high-frequency voice request expression, a command fit value for each preset function point relative to the user intent is determined, and the command fit value is used to indicate the degree of matching between the user intent and the preset function point; Based on the user intent, the environmental awareness information, and the preset function points, a functional validity value for each preset function point relative to the user intent is determined. The functional validity value is used to characterize the applicability of the preset function point under the environmental awareness information. The ranking score of the preset function points is determined based on the instruction fit value and the function validity value. The first candidate functional group is determined based on the ranking score.

3. The method according to claim 2, characterized in that, The step of determining the instruction fit value of each preset function point relative to the user intent based on the user intent and the high-frequency voice request expression includes: Calculate the similarity between the user intent and the multiple high-frequency voice request expressions to obtain multiple similarity values, wherein the multiple high-frequency voice request expressions are associated with the same preset function point; The maximum value among the multiple similar values ​​is determined as the instruction fitting value.

4. The method according to claim 2, characterized in that, The step of selecting and determining a first candidate function group from a preset function knowledge base based on the user intent and the environmental awareness information includes: When the user-perceived information is the user's non-voice operation information, the effective value of each preset function point relative to the user's intention is determined based on the user's intention, the environmental perception information, and the preset function points. Based on the effective values ​​of the functions, determine the ranking score of the preset function points; The first candidate functional group is determined based on the ranking score.

5. The method according to claim 1, characterized in that, The process of filtering the first candidate function group and determining the second candidate function group based on the trigger location information of the preset security rules and the user perception information includes: The first candidate function group is filtered according to the preset security rules to determine the first temporary execution function group and the first temporary suggestion function group; Based on the trigger location information, the first temporary execution function group is filtered to determine the second temporary execution function group and the second temporary suggestion function group; The execution sub-function group is determined based on the second temporary execution function group; The suggestion sub-function group is determined based on the first temporary suggestion function group and the second temporary suggestion function group; The second candidate function group is determined based on the execution sub-function group and the suggestion sub-function group.

6. The method according to claim 5, characterized in that, The step of determining the planning function group based on the second candidate function group includes: Based on the historical database, the planning function group is determined according to the second candidate function group.

7. The method according to claim 6, characterized in that, The historical database includes intents and historical planning function groups associated with those intents. The historical planning function groups include historical execution sub-function groups and historical suggestion sub-function groups. The historical database is updated according to the following steps: If the number of times a preset function point in the historical suggestion sub-function group is adopted reaches a first preset adoption threshold, the preset function point will be adjusted to the historical execution sub-function group to update the historical database. If the number of times the execution function point associated with the vehicle control command to be executed reaches the second preset number of adoptions threshold, the preset function point is adjusted to the historical suggestion sub-function group to update the historical database. The execution function point does not exist in the historical execution sub-function group and does not exist in the historical suggestion sub-function group. If the number of rejections of the preset function point in the execution sub-function group reaches a first preset rejection threshold, the preset function point will be moved to the suggestion sub-function group to update the historical database; and / or If the number of rejections of the preset function point in the suggested sub-function group reaches the second preset rejection threshold, the preset function point will be removed from the second candidate function group to update the historical database.

8. The method according to claim 1, characterized in that, The method further includes: If both the first executable and the second executable are not executable, feedback information is generated to guide the user to perform a clarification operation based on the feedback information.

9. A server, characterized in that, The server includes a processor and a memory, the memory storing a computer program that, when executed by the processor, implements the method according to any one of claims 1-8.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the program is executed by a processor, it implements the steps of the method as described in any one of claims 1-8.

Images