Vehicle control method, vehicle and server

Abstract

The invention provides a vehicle control method, a vehicle and a server, the method is applied to the technical field of human-computer interaction of the vehicle, and the method comprises the following steps: sending user perception data collected by the vehicle to the server, so that the server calls a target agent in the server according to the user perception data to generate a first instruction of target atomic power, the first instruction is converted into a second instruction which can be recognized by the vehicle, and the target atomic power is determined from a first atomic power set reported to the server side by the vehicle; receiving a second instruction sent by the server; and executing a function corresponding to the target atomic power according to the second instruction. Based on the scheme, the interaction experience of the user can be effectively improved.

Description

Technical Field

[0001] The present application relates to the field of human-computer interaction technology for vehicles, and more specifically, to a vehicle control method, a vehicle, and a service end in the field of human-computer interaction technology for vehicles. Background Art

[0002] With the development of vehicle human-computer interaction technology, some vehicles support the execution of specific functions based on user intentions. In related technologies, the vertical domain intelligent agent on the server side needs to be bound to the vehicle's execution intelligent agent to form a one-to-one correspondence. Among them, the vertical domain intelligent agent can understand the user's perception data to obtain relevant data, and return the relevant data to the corresponding execution intelligent agent, so that the corresponding execution intelligent agent can perform specific functions according to the relevant data. The strong coupling of the execution intelligent agent and the vertical domain intelligent agent leads to defects such as excessive vehicle system resource occupancy and insufficient scheduling flexibility, which in turn affects the user's interactive experience. Therefore, how to avoid the strong coupling of the execution intelligent agent and the vertical domain intelligent agent to improve the user's interactive experience has become an urgent problem to be solved. Summary of the Invention

[0003] The present application provides a vehicle control method, a vehicle, and a service end, which can effectively improve the user's interactive experience.

[0004] In a first aspect, a vehicle control method is provided, which is applied to a vehicle, and the method includes:

[0005] Sending user perception data collected by the vehicle to a server, so that the server can call a target agent in the server to generate a first instruction of a target atomic capability based on the user perception data, and converting the first instruction into a second instruction recognizable by the vehicle, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server;

[0006] Receive the second instruction sent by the server;

[0007] The function corresponding to the target atomic capability is executed according to the second instruction.

[0008] Through the above solution, by establishing an interaction mechanism in the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed, and the decoupling design of the atomic capability dimension is realized. Without relying on the strong coupling between the execution agent and the vertical domain agent, the vehicle can support the execution of specific functions based on user intentions, effectively improving the user's interactive experience.

[0009] In combination with the first aspect, in some possible implementations, the method also includes: obtaining the atomic capabilities of each application in multiple applications of the vehicle; determining a first atomic capability set based on the atomic capabilities of each application; and reporting the first atomic capability set to the server when preset reporting conditions are met.

[0010] The above solution first obtains the atomic capabilities of each of the vehicle's multiple applications. Secondly, the atomic capabilities of each application are aggregated and standardized to determine a first set of atomic capabilities, forming a complete vehicle capability list. Finally, through a judgment mechanism based on preset reporting conditions, the first set of atomic capabilities is reported to the server at a specific time. This enables dynamic management and precise synchronization of the vehicle's atomic capabilities, enabling the server to make subsequent scheduling decisions based on the vehicle's latest capability status, resolving the scheduling deviation problem caused by lagging capability information in related technologies.

[0011] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, obtaining the atomic capabilities of each application in multiple applications of the vehicle includes: scanning the atomic capabilities of each first application in multiple applications of the vehicle, and the first application is configured to be passively started; responding to the registration instructions of each second application in multiple applications, determining the atomic capabilities of each second application according to the registration instructions of each second application, and the second application is configured to be automatically started.

[0012] This solution allows static scanning to obtain atomic capability information for the first application, avoiding unnecessary application startup overhead. For the second application, dynamic registration enables real-time updates of atomic capabilities. Atomic capability information obtained through both methods is centrally managed in the capability registration center, providing a reliable data foundation for subsequent atomic capability set reporting and server-side scheduling.

[0013] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, the method also includes: if the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, it is determined that the preset reporting conditions are met; if the first atomic capability set is different from the atomic capability set obtained by the vehicle last time, it is determined that the preset reporting conditions are met.

[0014] Through the above scheme, the preset reporting conditions are first set to be met when the first atomic capability set is the first atomic capability set obtained after the vehicle is started, ensuring that the server can obtain complete atomic capability information in a timely manner during the vehicle initialization phase. Secondly, the preset reporting conditions are also set to be met when the first atomic capability set is different from the atomic capability set last obtained by the vehicle, allowing the server to track the dynamic changes of the vehicle's atomic capabilities in real time. This ensures that the server always makes scheduling decisions based on the vehicle's latest atomic capability status.

[0015] In combination with the first aspect and the above-mentioned implementation methods, in some possible implementation methods, executing the function corresponding to the target atomic capability according to the second instruction includes: determining the target application among multiple applications of the vehicle according to the second instruction, and calling the target application to execute the function corresponding to the target atomic capability.

[0016] This approach first identifies the target application from the vehicle's multiple applications based on the second instruction, then calls the target application to execute the function corresponding to the target atomic capability. This precise matching of atomic capability dimensions and a dynamic call mechanism enable on-demand activation of application resources, improving function execution efficiency, system responsiveness, and the user experience.

[0017] In a second aspect, a vehicle control method is provided, which is applied to a server, and the method includes:

[0018] Receive user perception data collected by the vehicle;

[0019] Invoke a first instruction of a target agent in a server to generate a target atomic capability according to user perception data, where the target atomic capability is determined from a first atomic capability set reported by the vehicle to the server;

[0020] converting the first instruction into a second instruction recognizable by the vehicle;

[0021] A second instruction is sent to the vehicle so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

[0022] Through the above solution, by establishing an interaction mechanism in the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed, and the decoupling design of the atomic capability dimension is realized. Without relying on the strong coupling between the execution agent and the vertical domain agent, the vehicle can support the execution of specific functions based on user intentions, effectively improving the user's interactive experience.

[0023] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, calling the target agent in the server side according to the user perception data to generate the first instruction of the target atomic capability includes: determining at least one available agent on the server side, the available agent is an agent configured to be online; determining the available atomic capability of each available agent in the at least one available agent, the available atomic capability is the atomic capability configured to be online; determining at least one candidate agent in the at least one available agent according to the first atomic capability set reported by the vehicle to the server side, the available atomic capability of the candidate agent includes at least one atomic capability in the first atomic capability set; calling the target agent in the at least one candidate agent according to the user perception data to generate the first instruction of the target atomic capability.

[0024] Through the above scheme, through the hierarchical screening mechanism, the offline intelligent agents and atomic capabilities are first excluded, and then the candidate intelligent agents are further screened based on the actual capabilities of the vehicle, and finally accurate scheduling is achieved, which effectively improves the accuracy and efficiency of intelligent agent scheduling and avoids the waste of resources caused by invalid scheduling.

[0025] In combination with the second aspect and the above-mentioned implementation methods, in some possible implementation methods, the method also includes: in response to the first configuration instruction, configuring at least one of the multiple intelligent agents on the server side as an available intelligent agent that is online, and configuring the intelligent agents other than the available intelligent agents among the multiple intelligent agents on the server side as unavailable intelligent agents that are offline; and / or, in response to the second configuration instruction, configuring at least one of the multiple atomic capabilities on the server side as an available atomic capability that is online, and configuring the atomic capabilities other than the available atomic capabilities among the multiple atomic capabilities on the server side as unavailable atomic capabilities that are offline.

[0026] Through the above solution, hierarchical configuration management is utilized to not only manage and control the overall functional modules at the intelligent agent dimension, but also control specific functions at the atomic capability dimension, significantly improving the flexibility and adaptability of system management while ensuring that server-side scheduling decisions are always based on the latest configuration status.

[0027] In a third aspect, a vehicle control device is provided, which is applied to a vehicle and includes:

[0028] a sending unit, configured to send user perception data collected by the vehicle to a server, so that the server can invoke a target agent in the server to generate a first instruction for a target atomic capability based on the user perception data, and convert the first instruction into a second instruction recognizable by the vehicle, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server;

[0029] A receiving unit, configured to receive a second instruction sent by the server;

[0030] The execution unit is configured to execute a function corresponding to the target atomic capability according to the second instruction.

[0031] In a fourth aspect, a vehicle control device is provided, which is applied to a server, and includes:

[0032] A receiving unit, configured to receive user perception data collected by the vehicle;

[0033] a generating unit, configured to call a target agent in a server to generate a first instruction for a target atomic capability based on user perception data, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server;

[0034] a conversion unit, configured to convert the first instruction into a second instruction recognizable by the vehicle;

[0035] The sending unit is used to send a second instruction to the vehicle so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

[0036] A fifth aspect provides a vehicle, characterized in that the vehicle includes:

[0037] a memory for storing executable program code;

[0038] A processor is used to call and run executable program code from the memory, so that the vehicle executes the method in the above-mentioned first aspect or any possible implementation of the first aspect.

[0039] In a sixth aspect, a computer program product is provided, which includes: a computer program code, which, when executed on a computer, enables the computer to execute the method in the first aspect or any possible implementation of the first aspect.

[0040] In the seventh aspect, a computer-readable storage medium is provided, which stores a computer program code. When the computer program code runs on a computer, the computer executes the method in the above-mentioned first aspect or any possible implementation of the first aspect. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 This is an example schematic diagram of an embodiment of the present application providing an example of implementing function control based on a vertical domain agent and an execution agent;

[0042] Figure 2 This is a flow chart of a vehicle control method provided in an embodiment of the present application;

[0043] Figure 3 This is a schematic diagram of a process for reporting a first atomic capability set provided in an embodiment of the present application;

[0044] Figure 4 This is a schematic diagram of a static scanning and dynamic registration process provided by an embodiment of the present application;

[0045] Figure 5 This is a flow chart of a vehicle control method provided in an embodiment of the present application;

[0046] Figure 6 This is a schematic diagram of a process for generating a first instruction provided by an embodiment of the present application;

[0047] Figure 7 This is an example schematic diagram of implementing function control based on atomic capabilities provided in an embodiment of the present application;

[0048] Figure 8 This is a schematic structural diagram of a vehicle control device provided in an embodiment of the present application;

[0049] Figure 9 This is a schematic structural diagram of a vehicle control device provided in an embodiment of the present application;

[0050] Figure 10 This is a schematic structural diagram of a vehicle provided in an embodiment of the present application;

[0051] Figure 11 This is a structural diagram of a server provided in an embodiment of the present application. DETAILED DESCRIPTION

[0052] The following will clearly and thoroughly describe the technical solutions in this application in conjunction with the accompanying drawings. In the description of the embodiments of this application, unless otherwise specified, " / " means or, for example, A / B can mean A or B: "and / or" in the text is only a description of the association relationship of associated objects, indicating that there can be three relationships, 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 this application, "multiple" means two or more than two.

[0053] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of technical features indicated. Therefore, features defined as "first" or "second" may explicitly or implicitly include one or more of such features.

[0054] With the development of vehicle human-computer interaction technology, some vehicles support the execution of specific functions based on user intentions. In related technologies, the vertical domain intelligent agent on the server side needs to be bound to the execution intelligent agent of the vehicle to form a one-to-one correspondence. Among them, the vertical domain intelligent agent can understand the intention of the user perception data to obtain relevant data, and return the relevant data to the corresponding execution intelligent agent, so that the corresponding execution intelligent agent can perform specific functions according to the relevant data. It should be noted that the intelligent agent involved in this application can also be called Agent, which refers to an intelligent entity that can autonomously perceive the environment, make decisions and perform actions, including but not limited to software programs, hardware modules or a combination thereof.

[0055] For example, see Figure 1 , Figure 1 This is an example diagram of an embodiment of the present application providing an example of a function control based on a vertical domain agent and an execution agent. The vehicle includes a user perception unit and a set of execution agents.

[0056] The user perception unit is used to collect the user's multimodal interaction information to achieve multimodal perception. For example, the user perception unit can perceive the user's interactive behavior, the user's interactive instructions, and other interactive data of the user to obtain user perception data. The user's interactive behavior refers to the action characteristics of the user when interacting with the vehicle; the user's interactive instructions refer to the clear operation instructions input by the user through voice or touch; the user's other interactive data refers to the environmental data or user status data collected by the vehicle through sensors. It can be understood that the user perception data includes at least one of the above-mentioned interactive behaviors, interactive instructions and other interactive data. In addition, the execution agent set refers to the set of agents deployed by the vehicle to perform specific functional operations. For example, the vehicle's execution agent set includes execution agent A, execution agent B, and execution agent C.

[0057] The server includes an intelligent agent scheduling system, which is used to receive user perception data and schedule the corresponding vertical domain intelligent agent for processing. Among them, the intelligent agent scheduling system includes an intelligent agent hub, and vertical domain intelligent agent A, vertical domain intelligent agent B, and vertical domain intelligent agent C that can be scheduled by the intelligent agent hub. Vertical domain refers to the vertical domain. For example, in the field of human-computer interaction technology of vehicles, the vertical domain can be the navigation control field, entertainment system control field or vehicle equipment control field, etc., and there is no restriction on this. The above-mentioned vertical domain intelligent agent A is bound to the execution intelligent agent A, the vertical domain intelligent agent B is bound to the execution intelligent agent B, and the vertical domain intelligent agent C is bound to the execution intelligent agent C.

[0058] In practice, the vehicle's user perception unit provides user perception data to the server-side agent scheduling system. The agent hub in the agent scheduling system determines the corresponding vertical agent based on the characteristics of the user perception data and forwards the user perception data to the vertical agent for processing, generating a function execution instruction that matches the user's intent. The agent scheduling system then returns this function execution instruction to the vehicle, and the corresponding execution agent in the vehicle executes the predetermined function according to the function execution instruction.

[0059] However, the strong coupling between the executive agent and the vertical domain agent leads to defects such as excessive vehicle system resource usage and insufficient scheduling flexibility, which are manifested in the following aspects:

[0060] First, the binding relationship between the server-side vertical agent and the vehicle's execution agent results in excessive system coupling. Because each vertical agent must correspond to a specific execution agent, adjustments to the server-side agent's functionality must also be made to the corresponding vehicle execution agent. This architecture necessitates bidirectional adaptation for system updates and maintenance, significantly increasing development costs and the complexity of version management.

[0061] Second, the vehicle's execution agent relies on the application to which it belongs to automatically start up when the system is turned on, resulting in a waste of system resources. To ensure that the execution agent is responsive to server calls at any time, the relevant application must remain resident in memory. When a vehicle has multiple applications with execution agents installed, these applications automatically load at system startup, occupying a large amount of processor and memory resources. In actual operation, many execution agents may remain idle for long periods of time, while their host applications continue to consume system resources, resulting in a decline in overall vehicle performance and, in severe cases, even causing system lag.

[0062] Third, functional overlap exists between different execution agents, leading to duplicate development. Because execution agents utilize a holistic design, even if multiple agents contain the same functional modules (such as pop-up window control and window operation), functional reuse is impossible. Developers must implement the same functionality separately for each execution agent, resulting in a waste of development resources. Furthermore, this duplication increases the maintenance cost of subsequent feature updates, as modifications to the same functionality must be implemented simultaneously across multiple execution agents.

[0063] Fourth, the server-side vertical agent cannot accurately match the vehicle's actual capabilities when scheduling tasks. Because scheduling decisions are based solely on user perception data and lack real-time awareness of the vehicle's currently available functions, this can lead to ineffective scheduling. Typical scenarios include assigning tasks to vehicles without specific applications installed, or erroneously triggering unsupported hardware functions. Such scheduling discrepancies not only reduce system reliability but also cause user interruptions, impacting the user experience.

[0064] Based on the above reasons, it can be seen that the strong coupling between the execution agent and the vertical domain agent will affect the user's interactive experience. Therefore, how to avoid the strong coupling between the execution agent and the vertical domain agent to improve the user's interactive experience has become an urgent problem to be solved.

[0065] In response to the above problems, the solution provided by the embodiment of the present application mainly includes: first, the vehicle sends the collected user perception data to the server, and the user perception data is used to trigger the vertical domain intelligent agent of the server to generate a first instruction corresponding to the target atomic capability, and convert the first instruction into a second instruction recognizable by the vehicle, wherein the target atomic capability is selected from the first atomic capability set actively reported by the vehicle; secondly, the vehicle receives the second instruction fed back by the server; finally, the vehicle executes the specific function corresponding to the target atomic capability according to the second instruction. In this way, by establishing an interaction mechanism in the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed.

[0066] The embodiments of the present application specifically address several deficiencies of the related art:

[0067] First, to address the system coupling issue caused by the strong binding between the server-side vertical domain agent and the vehicle execution agent, the embodiment of the present application enables the server to perform functional matching in the atomic capability dimension by having the vehicle actively report the first atomic capability set, thus eliminating the one-to-one correspondence between the vertical domain agent and the execution agent. The server only needs to pay attention to the availability of the target atomic capability and does not need to maintain the binding relationship between the vertical domain agent and the execution agent, thus reducing the coupling degree of the system architecture.

[0068] Second, to address the resource waste caused by the automatic startup of applications that rely on executing intelligent agents, this embodiment of the application uses a dynamic reporting mechanism for the first set of atomic capabilities, eliminating the need for vehicles to keep all applications corresponding to these capabilities resident in memory. The server schedules based on the real-time reported atomic capability status, and the vehicle can activate the corresponding application execution target atomic capability on demand, significantly reducing the processor and memory resource usage of resident applications.

[0069] Third, to address the issue of duplicate development caused by overlapping functions across different execution agents, this embodiment of the application breaks down functions into atomic capability dimensions for management. The same function only needs to be developed once for the atomic capability implementation, and different vertical domain agents can share and call the same atomic capability, avoiding duplicate implementation of the same function across different execution agents and reducing development and maintenance costs.

[0070] Fourth, to address the issue of vertical agent scheduling failing to match the actual vehicle capabilities, this embodiment of the application enables the vehicle to periodically report a first atomic capability set, ensuring that the server always has current available capabilities. Before generating a first instruction, the server's vertical agent can verify whether the target atomic capability exists in the first atomic capability set, ensuring that the scheduling instruction matches the vehicle's actual capabilities.

[0071] In summary, the embodiments of the present application, through the decoupling design of the atomic capability dimension, can enable the vehicle to support the execution of specific functions based on user intentions without relying on the strong coupling of the execution agent and the vertical domain agent, thereby effectively improving the user's interactive experience.

[0072] The following will be combined Figure 2 - Figure 7 , the vehicle control method provided in the embodiment of the present application is introduced in detail.

[0073] See Figure 2 , Figure 2 This is a flow chart of a vehicle control method provided in an embodiment of the present application. Figure 2 As shown, the method of the embodiment of the present application is applied to a vehicle and may include the following steps S101-S103.

[0074] S101, sending user perception data collected by the vehicle to the server, so that the server can call the target intelligent agent in the server according to the user perception data to generate a first instruction of the target atomic capability, and convert the first instruction into a second instruction recognizable by the vehicle. The target atomic capability is determined from the first atomic capability set reported by the vehicle to the server.

[0075] Specifically, the atomic capability involved in this embodiment refers to the smallest functional unit that can be executed by the vehicle's application, such as window control, air conditioning adjustment, multimedia playback, etc. It can be understood that based on the design of atomic capabilities, you can choose not to deploy the execution agent in the vehicle. The server involved in this embodiment refers to a remote server or server cluster where the agent scheduling system is deployed. The vehicle establishes a communication connection relationship with the server, and illustratively, the communication connection relationship can be a long link relationship. A long link relationship refers to a network connection established between the vehicle and the server that is kept open continuously, and both parties can perform two-way data transmission on this connection at any time.

[0076] One of the basic conditions for implementing this embodiment is that the vehicle reports a first atomic capability set to the server. The first atomic capability set refers to the summary information of all atomic capabilities currently available to the vehicle, representing the vehicle's current functional execution capabilities.

[0077] User perception data collected by the vehicle refers to the user's multimodal interaction information collected by the user perception unit. For example, the user perception unit can perceive the user's interactive behavior, user interaction commands, and other user interaction data to obtain user perception data. User interaction behavior refers to the characteristics of the user's movements when interacting with the vehicle; user interaction commands refer to the user's explicit operational instructions input through voice or touch; and other user interaction data refers to environmental data or user status data collected by the vehicle's sensors.

[0078] Based on preset trigger conditions or cycles, the vehicle can send user perception data collected by the vehicle to the server. Specifically, data transmission is triggered when user interaction behavior, user interaction instructions, or when timed reporting conditions are met.

[0079] Correspondingly, the server receives the user perception data sent by the vehicle, and calls the target agent in the server to generate the first instruction of the target atomic capability based on the user perception data. Among them, the target agent is selected from multiple agents on the server, and the multiple agents on the server can be vertical domain agents. It is understandable that the target agent has the ability to analyze user intentions (or the user intention analysis ability of the vertical domain), and can process the user perception data to generate the first instruction of the target atomic capability. Among them, the target atomic capability refers to the atomic capability corresponding to the function to be executed that matches the user intention. It should be noted that the number of target atomic capabilities can be one or more, and there is no limitation on this. The first instruction of the target atomic capability refers to the function execution instruction described in the standard format of the server. It is understandable that since the server and the vehicle use different instruction format specifications, the first instruction cannot be directly executed by the vehicle, so the server converts the first instruction into a second instruction that can be recognized by the vehicle. Among them, recognizable refers to the instruction parsing specification that meets the preset vehicle settings.

[0080] S102: Receive a second instruction sent by the server.

[0081] Specifically, after converting the first instruction into a second instruction recognizable by the vehicle, the server sends the second instruction to the vehicle. Correspondingly, the vehicle receives the second instruction sent by the server.

[0082] S103: Execute the function corresponding to the target atomic capability according to the second instruction.

[0083] Specifically, in some cases, the vehicle has only one application that supports the target atomic capability, so the application can be directly called to execute the function corresponding to the target atomic capability. For example, the vehicle queries the corresponding target application in the pre-registered atomic capability and application mapping relationship based on the target atomic capability identifier contained in the second instruction; if the query result is a unique match, the target application is started or woken up, and the execution parameters contained in the second instruction are passed to the target application; the target application completes the functional operation corresponding to the target atomic capability based on the received execution parameters. For example, when the target atomic capability is "open the sunroof", the vehicle determines the only sunroof control application that supports this capability, and calls this application to execute the sunroof opening function.

[0084] In some cases, if the vehicle has multiple applications that support the target atomic capability, the optimal application can be selected to execute the function based on preset priority rules or user configuration. For example, the vehicle queries all candidate applications that support the atomic capability in the pre-registered atomic capability and application mapping relationship based on the target atomic capability identifier contained in the second instruction; then the target application for final execution can be determined based on at least one of the following strategies: sort by application priority and select the application with the highest priority; select the application that is already active based on the current running status of the application to reduce resource consumption; determine the target application with reference to the user's historical selection records or preference settings; select the optimal application based on the completeness of the application function or performance indicators. After determining the target application, the vehicle forwards the second instruction to the application to execute the corresponding function. For example, when multiple media applications support the "play music" atomic capability, the system can give priority to the default player application that the user has recently used to execute the function.

[0085] In this embodiment, first, the vehicle sends the collected user perception data to the server. The user perception data is used to trigger the vertical domain intelligent agent of the server to generate a first instruction corresponding to the target atomic capability, and convert the first instruction into a second instruction recognizable by the vehicle, wherein the target atomic capability is selected from the first atomic capability set actively reported by the vehicle; secondly, the vehicle receives the second instruction fed back by the server; finally, the vehicle executes the specific function corresponding to the target atomic capability according to the second instruction. In this way, by establishing an interaction mechanism of the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed, and the decoupling design of the atomic capability dimension is realized. Without relying on the strong coupling of the execution intelligent agent and the vertical domain intelligent agent, the vehicle can support the execution of specific functions based on user intentions, effectively improving the user's interactive experience.

[0086] See Figure 3 , provides a flowchart of reporting the first atomic capability set for the embodiment of the present application, such as Figure 3 As shown, the method of the embodiment of the present application may further include the following steps S201-S203.

[0087] S201, obtaining atomic capabilities of each of multiple applications in a vehicle;

[0088] S202, determining a first atomic capability set according to the atomic capabilities of each application;

[0089] S203: When a preset reporting condition is met, report the first atomic capability set to the server.

[0090] Specifically, the vehicle involved in this embodiment is configured with multiple applications, and the atomic capabilities of different applications may be different. Therefore, in order to accurately reflect the set of functions that the vehicle can currently execute, it is first necessary to obtain the atomic capabilities of each application in the vehicle's multiple applications. Regarding this step, in some possible implementations, atomic capabilities can be obtained through static scanning or dynamic registration. Static scanning refers to the vehicle's capability management module actively determining the atomic capabilities of certain applications in a scanning manner; dynamic registration refers to the application actively registering the atomic capabilities it supports to the capability management module through relevant interfaces during the operation of certain applications.

[0091] Furthermore, based on the atomic capabilities of each application, a first atomic capability set is determined. Specifically, the atomic capabilities obtained by static scanning and dynamic registration are deduplicated and standardized, a unified atomic capability description format is generated, and the mapping relationship between each atomic capability and the application to which it belongs is recorded. Among them, the standardization process can be to unify the naming rules, parameter formats and functional description methods of the atomic capabilities. The storage form of the first atomic capability set can be a list, a hash table or a database record, and its content can be an atomic capability identifier, a functional description, parameter requirements and application information that supports the capability.

[0092] Taking into account the network transmission load and server processing efficiency, after obtaining the first atomic capability set, the vehicle does not immediately report the first atomic capability set to the server, but needs to determine whether the preset reporting conditions are met. Among them, the preset reporting conditions refer to the set of rules that trigger the vehicle to synchronize the atomic capability set to the server, and its setting must take into account both data real-time and system performance. Exemplarily, the preset reporting conditions can be set based on at least one of the following situations: when the first startup is completed, when the atomic capability set changes, when the periodic time interval arrives, or when an active query request from the server is received.

[0093] If the preset reporting conditions are not met, the vehicle caches the first atomic capability set locally and continuously monitors for changes in atomic capabilities or reporting condition triggering events. If the preset reporting conditions are met, the vehicle encapsulates the first atomic capability set in a preset data format and sends it to the server through the established communication relationship. The server then receives the first atomic capability set reported by the vehicle and uses it for subsequent agent scheduling decisions.

[0094] In this embodiment, the atomic capabilities of each of the vehicle's multiple applications are first acquired. Next, the atomic capabilities of each application are aggregated and standardized to determine a first set of atomic capabilities, forming a complete vehicle capability list. Finally, through a judgment mechanism based on pre-set reporting conditions, the first set of atomic capabilities is reported to the server at specific times. This allows for dynamic management and precise synchronization of the vehicle's atomic capabilities, enabling the server to make subsequent scheduling decisions based on the vehicle's latest capability status, resolving the scheduling bias issue in related technologies caused by lagging capability information.

[0095] See Figure 4 , provides a schematic diagram of a static scanning and dynamic registration process for the embodiment of the present application, such as Figure 4 As shown, the method of the embodiment of the present application may include the following steps S301-S302, and steps S301-S302 may be used as Figure 3 The detailed steps of step S201 of the embodiment are shown.

[0096] S301, scanning atomic capabilities of each first application in a plurality of applications of a vehicle, where the first application is configured to be passively started;

[0097] S302 : In response to a registration instruction of each second application among a plurality of applications, the atomic capability of each second application is determined according to the registration instruction of each second application, and the second application is configured to be automatically started.

[0098] Specifically, the vehicle involved in this embodiment includes multiple applications, including a first application and a second application. The first application is an application that does not automatically load and run when the vehicle is started, but is activated and started only when a clear function call request or other instruction is received; the second application is an application that automatically loads and runs when the vehicle is started, and remains running without waiting for external calls.

[0099] It is understandable that after the vehicle's on-board system is turned on, the first application will not automatically start running. Therefore, it is necessary to scan the atomic capabilities of each first application in the vehicle's multiple applications based on a static scanning method. In some possible implementations, the vehicle's capability management module traverses all first applications installed in the storage device according to a preset scanning path when the system starts; for each first application, its preset manifest file or configuration file is read, which contains description information of all atomic capabilities supported by the first application; the capability management module parses the description information and extracts metadata such as atomic capability identifiers, function descriptions, input parameter formats, and return value types; a mapping relationship is established between the parsed atomic capability information and the corresponding first application identifier, and the information is stored in the capability registration center of the capability management module. It is understandable that the scanning process does not require the actual startup of the first application to obtain its atomic capability information.

[0100] Accordingly, after the vehicle's on-board system is turned on, the second application will automatically start running. Therefore, based on the dynamic registration method, in response to the registration instructions of each second application in multiple applications, the atomic capabilities of each second application can be determined according to the registration instructions of each second application. In some possible implementations, after completing the startup initialization, each second application actively sends a registration request to the capability registration center of the capability management module, and the request contains a detailed description of all atomic capabilities supported by the second application; the capability registration center of the capability management module provides a standardized registration interface to receive and verify the registration instructions sent by the second application; the verification content includes but is not limited to application identity authentication, atomic capability format verification and permission check; for registration instructions that pass the verification, the capability management module extracts the atomic capability information therein, establishes a mapping relationship with the corresponding second application identifier, and updates it to the capability registration center of the capability management module; at the same time, the provider information and registration timestamp of each atomic capability are recorded. It can be understood that the dynamic registration mechanism allows the second application to update its atomic capability information during operation, including adding, modifying or deleting atomic capabilities.

[0101] In this embodiment, the combined implementation of static scanning and dynamic registration ensures that the vehicle can fully and accurately understand the atomic capability status of all applications. For the first application, obtaining its atomic capability information through static scanning avoids unnecessary application startup overhead; for the second application, dynamic registration enables real-time updates of atomic capabilities. Atomic capability information obtained through both methods is centrally managed in the capability registration center, providing a reliable data foundation for subsequent atomic capability set reporting and server-side scheduling.

[0102] In one embodiment, combining Figure 3 In the embodiment shown, the method of the embodiment of the present application may further include the following steps:

[0103] If the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, it is determined that the preset reporting condition is met;

[0104] If the first atomic capability set is different from the atomic capability set acquired last time by the vehicle, it is determined that the preset reporting condition is met.

[0105] Specifically, this embodiment proposes determining whether the preset reporting condition is satisfied under two circumstances.

[0106] In one case, if the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, then it can be determined that the preset reporting conditions are met. Among them, the atomic capability set obtained for the first time after the vehicle is started refers to the summary information of all atomic capabilities fully collected for the first time by the capability registration center of the capability management module after the vehicle completes the system startup initialization. Regarding this process, in some possible implementations, the capability registration center of the capability management module establishes an initialization detection mechanism when the system starts. When it is detected that the atomic capability information of all applications has been collected, the first atomic capability set is generated and marked as the set obtained for the first time.

[0107] In another case, if the first atomic capability set is different from the atomic capability set acquired by the vehicle last time, then it can be determined that the preset reporting conditions are met. Among them, the atomic capability set acquired by the vehicle last time refers to the atomic capability summary information that was completely collected and stored by the capability registration center of the capability management module in the previous time. Regarding this process, in some possible implementations, the capability registration center of the capability management module regularly or in real time compares the currently acquired first atomic capability set with the atomic capability set acquired last time; when the comparison result shows that there is a difference in the number or type of atomic capabilities, it is determined that there is a difference. It can be understood that this mechanism ensures that the service end can promptly know the changes in the atomic capabilities of the vehicle end.

[0108] In this embodiment, the preset reporting condition is first set to be satisfied when the first atomic capability set is the first atomic capability set obtained after the vehicle is started, ensuring that the server can obtain complete atomic capability information in a timely manner during the vehicle initialization phase; secondly, the preset reporting condition is also set to be satisfied when the first atomic capability set is different from the atomic capability set obtained by the vehicle last time, so that the server can track the dynamic changes of the atomic capabilities of the vehicle in real time. In this way, on the one hand, it is ensured that the server can immediately understand the atomic capabilities supported by the vehicle when the vehicle is started, and on the other hand, it achieves timely synchronization of atomic capability changes, thereby ensuring that the server always makes scheduling decisions based on the latest atomic capability status of the vehicle.

[0109] In one embodiment, Figure 2 Step S103 of the embodiment shown is further refined and may include the following steps:

[0110] A target application is determined from a plurality of applications of the vehicle according to the second instruction, and the target application is called to execute a function corresponding to the target atomic capability.

[0111] Specifically, the target application in this embodiment refers to a specific application instance in the current vehicle that can execute the target atomic capability. It is understood that due to the vehicle's atomic capability-based functional management mechanism, the same target atomic capability may be supported by multiple different applications. Therefore, it is necessary to determine the appropriate target application for execution based on the relevant information in the second instruction.

[0112] Regarding the process of determining the target application, in some possible implementations, the capability scheduling center of the capability management module first parses the target atomic capability identifier contained in the second instruction, which corresponds to the target atomic capability identifier in the first instruction issued by the server; then, the capability scheduling center of the capability management module queries the capability registration center for the mapping relationship between the locally stored atomic capability and the application; based on the query results, the capability scheduling center of the capability management module screens out a set of candidate applications that support the target atomic capability; finally, the final target application is determined from the set of candidate applications according to a preset selection strategy. Among them, the preset selection strategy may include but is not limited to the following rules: give priority to applications that are already in running state to reduce resource consumption; select the application with the highest priority according to the application priority sorting; select the application that is most likely to be accepted by the user with reference to the user's historical usage records or preference settings; select the optimal application according to the application function completeness or performance indicators.

[0113] Regarding the process of calling the target application to execute a function, in some possible implementations, the capability scheduling center of the capability management module first verifies the current status of the target application: if the target application is already in a running state, a function call request containing execution parameters is directly sent to it; if the target application is not running, the target application is first started and waits for its initialization to be completed before sending a function call request. The function call request contains execution parameters extracted from the second instruction. These execution parameters have been formatted and conform to the interface specifications of the target application. After receiving the function call request, the target application executes the specific operations corresponding to the target atomic capability according to the parameters, such as controlling vehicle hardware equipment, displaying an interactive interface, or processing media data. After the execution is completed, the target application can return the execution results to the capability scheduling center of the capability management module. The capability scheduling center of the capability management module can choose to feed back the results to the user or record the execution log.

[0114] In this embodiment, all candidate applications supporting the target atomic capability are first identified based on the target atomic capability identifier in the second instruction. The optimal target application is then selected through a multi-dimensional evaluation. Differentiated calling strategies are then implemented based on the target application's operating status to ensure the reliability and efficiency of function execution. Finally, a standardized interface protocol is used to implement command transmission between the capability dispatch center of the capability management module and the target application. This ensures the flexibility of atomic capability execution and optimizes system resource utilization. Furthermore, a standardized calling mechanism reduces the coupling between applications, providing fundamental support for the scalability of vehicle functions.

[0115] In this embodiment, the second instruction first determines the target application from among the vehicle's multiple applications. Then, the target application is called to execute the function corresponding to the target atomic capability. This precise matching of atomic capability dimensions and a dynamic calling mechanism enable on-demand activation of application resources, improving function execution efficiency and system responsiveness, and enhancing the user interaction experience.

[0116] See Figure 5 , Figure 5 This is a flow chart of a vehicle control method provided in an embodiment of the present application. Figure 5 As shown, the method of the embodiment of the present application is applied to the server and may include the following steps S401-S404.

[0117] S401, receiving user perception data collected by the vehicle.

[0118] Specifically, the user perception data collected by the vehicle has been explained in the above embodiment and will not be repeated here. The vehicle sends the user perception data it collects to the server, and correspondingly, the server receives the user perception data collected by the vehicle.

[0119] In some possible implementations, the server includes an agent management platform that has the ability to establish a persistent connection with the vehicle. Therefore, the agent management platform can receive user perception data collected by the vehicle based on the persistent connection.

[0120] S402, calling the target agent in the server to generate a first instruction of a target atomic capability according to the user perception data, where the target atomic capability is determined from the first atomic capability set reported by the vehicle to the server.

[0121] Specifically, in some possible implementations, the server may further include an agent central system, which includes an agent hub and multiple agents.

[0122] Because user perception data is collected and sent directly by the vehicle and has not been standardized, the server can convert non-standardized user perception data into standardized user perception data. In some possible implementations, the server includes an intelligent agent management platform that supports multimodal data standardization conversion functions. It can receive non-standardized user perception data sent by the vehicle, and perform format unification, data cleaning, and feature extraction on the non-standardized user perception data to generate standardized user perception data that meets the input requirements of the intelligent agent central system.

[0123] After the agent management platform obtains standardized user perception data, it can send the standardized user perception data to the agent central system. The agent central system matches the characteristics of the standardized user perception data to the corresponding target agent for processing, thereby obtaining the first instruction of the target atomic capability generated by the target agent. In response, the agent central system will return the first instruction of the target atomic capability to the agent management platform.

[0124] S403: Convert the first instruction into a second instruction recognizable by the vehicle.

[0125] Specifically, the first instruction is a function execution instruction described by the server in a standard format, and the second instruction is an execution instruction that complies with the vehicle's preset instruction parsing specifications.

[0126] The agent management platform has command conversion capabilities. Based on pre-configured conversion rules, it maps the target atomic capability identifier in the first command to the corresponding atomic capability code on the vehicle side, and simultaneously converts and adapts the execution parameters to generate a second command. In some possible implementations, the conversion rules can include the corresponding relationship between the atomic capability identifiers on the server side and the vehicle side, parameter type conversion rules, and data verification mechanisms.

[0127] S404: Send a second instruction to the vehicle, so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

[0128] Specifically, after receiving the second instruction, the agent management platform can send the second instruction to the vehicle based on the communication relationship of the long link, so that the vehicle can execute the function corresponding to the target atomic capability according to the second instruction. For details about the vehicle executing the function corresponding to the target atomic capability according to the second instruction, please refer to the above embodiment and will not be repeated here.

[0129] In this embodiment, first, the server receives the user perception data collected by the vehicle; second, based on the user perception data, the target agent in the server is called to generate the first instruction of the target atomic capability, wherein the target atomic capability is determined from the first atomic capability set reported by the vehicle to the server, ensuring that the instruction generation matches the actual capability of the vehicle; then the first instruction is converted into a second instruction that the vehicle can recognize; finally, the second instruction is sent to the vehicle, so that the vehicle can execute the function corresponding to the target atomic capability according to the second instruction. In this way, by establishing an interaction mechanism of the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed, and the decoupling design of the atomic capability dimension is realized. Without relying on the strong coupling of the execution agent and the vertical domain agent, the vehicle can support the execution of specific functions based on user intentions, effectively improving the user's interactive experience.

[0130] See Figure 6 , provides a flow chart of generating a first instruction for an embodiment of the present application, such as Figure 6 As shown, the method of the embodiment of the present application may include the following steps S501-S504, and steps S501-S504 may be used as Figure 5 The detailed steps of step S402 in the embodiment are shown.

[0131] S501, determining at least one available agent on the server side, where the available agent is an agent configured to be online;

[0132] S502, determining the available atomic capability of each available agent in at least one available agent, where the available atomic capability is the atomic capability configured to be online;

[0133] S503, determining at least one candidate agent from at least one available agent based on the first atomic capability set reported by the vehicle to the server, where the available atomic capabilities of the candidate agent include at least one atomic capability in the first atomic capability set;

[0134] S504: Call a target agent among at least one candidate agent according to user perception data to generate a first instruction for a target atomic capability.

[0135] Specifically, the online state involved in this embodiment refers to the state in which the intelligent agent or atomic capability is set to be callable by the intelligent agent central system of the server side. Correspondingly, the offline state refers to the state in which the intelligent agent or atomic capability is set to be not callable by the intelligent agent central system of the server side.

[0136] On the one hand, the multiple agents on the server side include at least one available agent configured to be online, and unavailable agents other than the available agents configured to be offline. Available agents can receive user perception data from the agent central system and generate corresponding atomic capability instructions; unavailable agents may be currently under maintenance, upgrade, or deactivated and cannot respond to scheduling requests from the agent central system.

[0137] On the other hand, the server's multiple atomic capabilities refer to the set of all atomic capabilities supported by the server. The server's multiple atomic capabilities include at least one available atomic capability configured to be online, and all unavailable atomic capabilities other than the available atomic capabilities configured to be offline.

[0138] In order to ensure that the server can accurately dispatch vehicles based on their actual capabilities, it is first necessary to determine at least one available agent on the server and the available atomic capabilities of each available agent in the at least one available agent. Regarding this process, in some possible implementations, the agent management platform on the server maintains an agent configuration database and an atomic capability configuration database; the agent configuration database records the identifiers, functional descriptions, and online / offline status of all agents on the server; the atomic capability configuration database records the identifiers, agents to which they belong, functional descriptions, and online / offline status of all atomic capabilities on the server; the agent management platform regularly queries these two databases to filter out agents and atomic capabilities marked as online; for each available agent, the agent management platform further queries its associated available atomic capabilities and establishes a mapping relationship table between agents and atomic capabilities.

[0139] Furthermore, at least one candidate agent is determined from at least one available agent based on the first atomic capability set reported by the vehicle to the service end. The candidate agent refers to an available agent whose available atomic capabilities intersect with the first atomic capability set of the vehicle. Regarding this process, in some possible implementations, the agent management platform compares the first atomic capability set reported by the vehicle with a pre-established agent-atomic capability mapping relationship table; for each available agent, it is checked whether its associated available atomic capabilities contain at least one atomic capability in the first atomic capability set; available agents that meet the conditions are marked as candidate agents, and the matching atomic capability information is recorded; at the same time, available agents whose available atomic capabilities have no intersection with the first atomic capability set are excluded to ensure that subsequent scheduling decisions are based on the capabilities actually supported by the vehicle.

[0140] Finally, based on the user perception data, the target agent among at least one candidate agent is called to generate the first instruction of the target atomic capability. Regarding this process, in some possible implementations, the agent central system receives standardized user perception data and a list of candidate agents from the agent management platform; based on the characteristic dimensions of the user perception data (such as interaction scenario, functional field or intent type), an agent suitable for processing the data is selected from at least one candidate agent as the target agent; the user perception data is forwarded to the target agent for intent analysis and instruction generation; based on the analysis results, the target agent selects the target atomic capability that best matches the user intent from its available atomic capabilities and generates the corresponding first instruction.

[0141] In this embodiment, at least one available agent on the server side is first determined to ensure that the scheduling scope is limited to the set of agents that have been online; then the available atomic capabilities of each available agent in the at least one available agent are determined to determine the functional boundaries of each available agent; then, based on the first atomic capability set reported by the vehicle to the server side, at least one candidate agent is determined in the at least one available agent, and the available atomic capabilities of the candidate agent include at least one atomic capability in the first atomic capability set, so as to achieve dynamic matching between the server side capability and the actual capability of the vehicle; finally, based on the user perception data, the target agent in the at least one candidate agent is called to generate the first instruction of the target atomic capability, so as to ensure that the instruction generation is based on the appropriate agent selection. In this way, through the hierarchical screening mechanism, the offline agents and atomic capabilities are first excluded, and then the candidate agents are further screened based on the actual capability of the vehicle, and finally accurate scheduling is achieved, which effectively improves the accuracy and efficiency of agent scheduling and avoids the waste of resources caused by invalid scheduling.

[0142] In one embodiment, combining Figure 6 In the embodiment shown, the method of the embodiment of the present application may further include the following steps:

[0143] In response to the first configuration instruction, at least one of the plurality of agents on the server is configured as an online available agent, and agents other than the available agents in the plurality of agents on the server are configured as offline unavailable agents; and / or,

[0144] In response to the second configuration instruction, at least one of the multiple atomic capabilities of the server is configured as an online available atomic capability, and the atomic capabilities other than the available atomic capabilities of the multiple atomic capabilities of the server are configured as offline unavailable atomic capabilities.

[0145] Specifically, the server involved in this embodiment includes an intelligent agent background management system, which is used to manage the online and offline of intelligent agents and atomic capabilities.

[0146] The first configuration instruction refers to a configuration command issued by the server-side administrator through the server-side agent background management system to adjust the agent's online / offline status; the second configuration instruction refers to a configuration command issued by the server-side administrator through the server-side agent background management system to adjust the atomic capability's online / offline status. The online status indicates that the agent or atomic capability can be called by the server-side agent central system, while the offline status indicates that the agent or atomic capability is temporarily unavailable to the server-side agent central system.

[0147] Regarding the execution process of the first configuration instruction, in some possible implementation methods, the server-side intelligent agent background management system provides an intelligent agent configuration interface, which displays a list of all intelligent agents on the server and their current status; the server-side administrator selects the intelligent agent whose status needs to be adjusted through the interface, and specifies the target status (online or offline); after the intelligent agent background management system receives the first configuration instruction submitted by the server-side administrator, it first verifies the legality of the instruction and the validity of the authority; after the verification is passed, the intelligent agent background management system updates the status mark of the corresponding intelligent agent in the intelligent agent configuration database; for available intelligent agents configured to be online, the intelligent agent background management system adds them to the schedulable list of the intelligent agent central system; for unavailable intelligent agents configured to be offline, the intelligent agent background management system removes them from the schedulable list of the intelligent agent central system; at the same time, the intelligent agent background management system returns a configuration operation result notification to the server-side administrator.

[0148] Regarding the execution process of the second configuration instruction, in some possible implementation methods, the server-side intelligent agent background management system provides an atomic capability configuration interface, which displays the atomic capabilities associated with each intelligent agent and its current status; the server-side administrator can view the atomic capabilities associated with a specific intelligent agent and select the atomic capabilities whose status needs to be adjusted; after the intelligent agent background management system receives the second configuration instruction submitted by the server-side administrator, it executes a verification process similar to the first configuration instruction; after the verification is passed, the intelligent agent background management system updates the status mark of the corresponding atomic capability in the atomic capability configuration database; for the available atomic capability configured to be online, the intelligent agent background management system activates the capability in the capability set of the intelligent agent to which it belongs; for the unavailable atomic capability configured to be offline, the intelligent agent background management system disables it from the capability set of the intelligent agent to which it belongs; after the configuration change is completed, the intelligent agent background management system synchronously updates the intelligent agent-atomic capability mapping relationship table.

[0149] It should be noted that the first configuration instruction and the second configuration instruction can be executed independently or in combination.

[0150] In this embodiment, first, the first configuration instruction is used to achieve flexible control of the online / offline status of the server-side intelligent agent, allowing the server-side administrator to dynamically adjust the available intelligent agent set according to actual needs; secondly, the second configuration instruction is used to achieve flexible control of the online / offline status of atomic capabilities, supporting the activation or deactivation of functions based on the atomic capability dimension; in addition, the two configuration mechanisms can operate independently or in collaboration. In this way, through hierarchical configuration management, it is possible to control the overall functional module at the intelligent agent dimension and to control specific functions at the atomic capability dimension, significantly improving the flexibility and adaptability of system management, while ensuring that the server-side scheduling decisions are always based on the latest configuration status.

[0151] Based on the above Figures 2 to 6 For the illustrated embodiment, see Figure 7 , Figure 7 This is an example schematic diagram of implementing functional control based on atomic capabilities provided in an embodiment of the present application.

[0152] Specifically, when the vehicle is started, the vehicle's applications A and B, as second applications configured to start automatically, complete automatic loading and operation; application C, as a first application configured to start passively, remains in an unstarted state. The capability registration center of the vehicle's capability management module performs an initialization operation, including the following steps: first, scanning the atomic capabilities of each first application in all applications of the vehicle, specifically performing a static scan of application C, obtaining the atomic capability D it supports and establishing a mapping relationship; then, in response to the registration instructions of each second application, receiving the registration instruction containing atomic capability A sent by application A, and the registration instruction containing atomic capability A and atomic capability B sent by application B, to complete the dynamic registration process. The capability registration center summarizes and removes the collected atomic capability information, and determines that the first atomic capability set includes atomic capability A, atomic capability B, and atomic capability D. Since this first atomic capability set is the first atomic capability set obtained after the vehicle is started, the capability registration center determines that the preset reporting conditions are met and reports the first atomic capability set to the intelligent body management platform on the server side through long link communication.

[0153] The server-side intelligent agent background management system is pre-configured: vertical domain intelligent agent A and vertical domain intelligent agent B are configured to be online, and vertical domain intelligent agent C and vertical domain intelligent agent D are configured to be offline; atomic capabilities A and atomic capabilities B are configured to be online, and atomic capabilities C and atomic capabilities D are configured to be offline. After the intelligent agent management platform receives the first atomic capability set reported by the vehicle, it executes the available intelligent agent screening process: first, it queries the intelligent agent background management system to determine that the currently online available intelligent agents are vertical domain intelligent agent A and vertical domain intelligent agent B; then it determines that the available atomic capability of vertical domain intelligent agent A is atomic capability B, and the available atomic capabilities of vertical domain intelligent agent B are atomic capability A, atomic capability B, and atomic capability C; then it compares the first atomic capability set with the available atomic capabilities of each available intelligent agent, and determines that the atomic capability B of vertical domain intelligent agent A matches the atomic capability B in the first atomic capability set, and the atomic capability A and atomic capability B of vertical domain intelligent agent B match the atomic capability A and atomic capability B in the first atomic capability set, so vertical domain intelligent agent A and vertical domain intelligent agent B are marked as candidate intelligent agents.

[0154] When the user perception unit of the vehicle detects the user's interactive behavior, it collects and generates user perception data and sends it to the intelligent agent management platform on the server side through a long link. The intelligent agent management platform performs multimodal data standardization conversion on the received user perception data, generates standardized user perception data and forwards it to the intelligent agent hub of the intelligent agent central system. The intelligent agent hub selects the vertical domain intelligent agent B from the candidate intelligent agents as the target intelligent agent for processing based on the characteristic dimensions of the user perception data. The vertical domain intelligent agent B performs intent analysis on the standardized user perception data, determines that the atomic capability A needs to be called, generates a first instruction described in the standard format of the server side and returns it to the intelligent agent management platform. The intelligent agent management platform converts the first instruction into a second instruction that can be recognized by the vehicle, specifically including mapping the atomic capability A identifier to the corresponding encoding format on the vehicle side, and sending it to the capability scheduling center of the vehicle's capability management module through a long link.

[0155] After receiving the second instruction, the capability dispatch center of the vehicle performs the following operations: parses the target atomic capability identifier in the second instruction to determine it as atomic capability A; queries the atomic capability and application mapping relationship stored in the capability registration center to determine application A and application B that support atomic capability A; gives priority to application B that is already in operation according to the preset selection strategy; sends a function call request containing execution parameters to application B. Application B executes the specific functional operation corresponding to atomic capability A according to the received parameters to complete the response to the user's intention. If the subsequent user perception unit detects that a new interaction instruction requires the call of atomic capability D, since the atomic capability D of the server is offline, the intelligent agent management platform will exclude the vertical domain intelligent agent C and vertical domain intelligent agent D containing atomic capability D in the screening process to ensure that the dispatch instruction is always consistent with the actual capability of the vehicle and the configuration status of the server. When the application installation status of the vehicle changes, the capability registration center detects the change of the atomic capability set through periodic scanning, and reports the updated first atomic capability set to the server to achieve dynamic capability synchronization.

[0156] In this embodiment, by establishing an interaction mechanism of the atomic capability dimension, the function execution logic between the vehicle and the server is reconstructed, and the decoupling design of the atomic capability dimension is realized. Without relying on the strong coupling between the execution agent and the vertical domain agent, the vehicle can support the execution of specific functions based on user intentions, effectively improving the user's interactive experience.

[0157] The following will be combined Figure 8 The vehicle control device provided in the embodiment of the present application is introduced in detail. It should be noted that, Figure 8 The vehicle control device is used to execute the present application Figure 2 - Figure 4 For the convenience of explanation, only the part related to the embodiment of the present application is shown. For the specific technical details not disclosed, please refer to the present application. Figure 2 - Figure 4 Specifically, the vehicle control device 600 is applied to a vehicle and may include a sending unit 601, a receiving unit 602, and an execution unit 603, as follows:

[0158] A sending unit 601 is configured to send user perception data collected by the vehicle to a server, so that the server can invoke a target agent in the server to generate a first instruction for a target atomic capability based on the user perception data, and convert the first instruction into a second instruction recognizable by the vehicle, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server.

[0159] Receiving unit 602, configured to receive a second instruction sent by the server;

[0160] The execution unit 603 is configured to execute the function corresponding to the target atomic capability according to the second instruction.

[0161] Optionally, in some embodiments, the vehicle control device 600 can be used to: obtain the atomic capabilities of each application in multiple applications of the vehicle; determine a first atomic capability set based on the atomic capabilities of each application; and report the first atomic capability set to the server when preset reporting conditions are met.

[0162] Optionally, in some embodiments, the vehicle control device 600 can be used to: scan the atomic capabilities of each first application in multiple applications of the vehicle, and the first application is configured to be passively started; respond to the registration instructions of each second application in multiple applications, determine the atomic capabilities of each second application according to the registration instructions of each second application, and the second application is configured to be automatically started.

[0163] Optionally, in some embodiments, the vehicle control device 600 can be used to: if the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, determine that the preset reporting conditions are met; if the first atomic capability set is different from the atomic capability set obtained by the vehicle last time, determine that the preset reporting conditions are met.

[0164] Optionally, in some embodiments, the execution unit 603 may be configured to: determine a target application from a plurality of applications of the vehicle according to the second instruction, and call the target application to execute a function corresponding to the target atomic capability.

[0165] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned vehicle control method, which will not be repeated here.

[0166] The following will be combined Figure 9 The vehicle control device provided in the embodiment of the present application is introduced in detail. It should be noted that, Figure 9 The vehicle control device is used to execute the present application Figure 5 - Figure 6 For the convenience of explanation, only the part related to the embodiment of the present application is shown. For the specific technical details not disclosed, please refer to the present application. Figure 5 - Figure 6 Specifically, the vehicle control device 700 is applied to the server and may include a receiving unit 701, a generating unit 702, a converting unit 703, and a sending unit 704, as follows:

[0167] A receiving unit 701 is configured to receive user perception data collected by the vehicle;

[0168] A generating unit 702 is configured to call a target agent in a server to generate a first instruction for a target atomic capability based on user perception data, where the target atomic capability is determined from a first atomic capability set reported by the vehicle to the server;

[0169] A conversion unit 703, configured to convert the first instruction into a second instruction recognizable by the vehicle;

[0170] The sending unit 704 is used to send a second instruction to the vehicle, so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

[0171] Optionally, in some embodiments, the generation unit 702 can be used to: determine at least one available agent on the server side, where the available agent is an agent configured to go online; determine the available atomic capabilities of each available agent in the at least one available agent, where the available atomic capabilities are atomic capabilities configured to go online; determine at least one candidate agent in the at least one available agent based on the first atomic capability set reported by the vehicle to the server side, where the available atomic capabilities of the candidate agent include at least one atomic capability in the first atomic capability set; and call the target agent in the at least one candidate agent based on the user perception data to generate a first instruction for the target atomic capability.

[0172] Optionally, in some embodiments, the vehicle control device 700 can be used to: in response to a first configuration instruction, configure at least one of the multiple intelligent agents on the server side as an available intelligent agent that is online, and configure the intelligent agents other than the available intelligent agents among the multiple intelligent agents on the server side as unavailable intelligent agents that are offline; and / or, in response to a second configuration instruction, configure at least one of the multiple atomic capabilities on the server side as an available atomic capability that is online, and configure the atomic capabilities other than the available atomic capabilities among the multiple atomic capabilities on the server side as unavailable atomic capabilities that are offline.

[0173] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned vehicle control method, which will not be repeated here.

[0174] See Figure 10 , provides a structural diagram of a vehicle according to an embodiment of the present application. Figure 10 As shown, the vehicle 800 includes a processor 801 and a memory 802. The processor 801 is electrically connected to the memory 802.

[0175] Processor 801 is the control center of vehicle 800 and may include one or more processing cores. Using various interfaces and circuits, processor 801 connects various components of vehicle 800. By running or invoking computer programs stored in memory 802 and accessing data stored in memory 802, it executes various functions of vehicle 800 and processes data, thereby providing overall control over vehicle 800. Optionally, processor 801 may be implemented using at least one of the following hardware forms: digital signal processing (DSP), field programmable gate array (FPGA), or programmable logic array (PLA). Processor 801 may integrate one or a combination of a CPU, a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interfaces, and applications; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into processor 801 and may instead be implemented via a separate communications chip.

[0176] Memory 802 can be used to store software programs and modules. Processor 801 executes various functional applications and data processing by running the computer programs and modules stored in memory 802. Memory 802 may primarily include a program storage area and a data storage area. The program storage area may store an operating system and computer programs required for at least one function, while the data storage area may store data generated based on the use of vehicle 800.

[0177] In addition, the memory 802 may include a high-speed random access memory and a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 802 may also include a memory controller to provide the processor 801 with access to the memory 802.

[0178] In this embodiment, the processor 801 in the vehicle 800 loads instructions corresponding to one or more computer program processes into the memory 802 according to the following steps, and the processor 801 runs the computer program stored in the memory 802 to implement various functions as follows:

[0179] Sending user perception data collected by the vehicle to a server, so that the server can call a target agent in the server to generate a first instruction of a target atomic capability based on the user perception data, and converting the first instruction into a second instruction recognizable by the vehicle, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server;

[0180] Receive the second instruction sent by the server;

[0181] The function corresponding to the target atomic capability is executed according to the second instruction.

[0182] Optionally, the processor 801 may also execute: obtaining the atomic capabilities of each application in the vehicle's multiple applications; determining a first atomic capability set based on the atomic capabilities of each application; and reporting the first atomic capability set to the server when preset reporting conditions are met.

[0183] Optionally, when executing to obtain the atomic capabilities of each application in multiple applications of the vehicle, the processor 801 specifically performs: scanning the atomic capabilities of each first application in multiple applications of the vehicle, and the first application is configured to be passively started; responding to the registration instructions of each second application in the multiple applications, determining the atomic capabilities of each second application according to the registration instructions of each second application, and the second application is configured to be automatically started.

[0184] Optionally, the processor 801 can also execute: if the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, it is determined that the preset reporting conditions are met; if the first atomic capability set is different from the atomic capability set obtained by the vehicle last time, it is determined that the preset reporting conditions are met.

[0185] Optionally, when executing the function corresponding to the target atomic capability according to the second instruction, the processor 801 specifically performs: determining the target application among multiple applications of the vehicle according to the second instruction, and calling the target application to execute the function corresponding to the target atomic capability.

[0186] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned vehicle control method, which will not be repeated here.

[0187] See Figure 11 , provides a schematic diagram of the structure of a server for the embodiment of this application. Figure 11 As shown, the server 900 includes a processor 901 and a memory 902. The processor 901 is electrically connected to the memory 902.

[0188] Processor 901 is the control center of server 900 and may include one or more processing cores. Processor 901 utilizes various interfaces and circuits to connect the various components of server 900. By running or invoking computer programs stored in memory 902, as well as invoking data stored in memory 902, it executes various server 900 functions and processes data, thereby providing overall control over server 900. Optionally, processor 901 may be implemented using at least one of the following hardware forms: digital signal processing (DSP), field programmable gate array (FPGA), and programmable logic array (PLA). Processor 901 may integrate one or a combination of a CPU, a graphics processing unit (GPU), and a modem. The CPU primarily processes the operating system, user interfaces, and applications; the GPU is responsible for rendering and drawing display content; and the modem handles wireless communications. It is understood that the modem may not be integrated into processor 901 and may be implemented separately via a communication chip.

[0189] The memory 902 can be used to store software programs and modules. The processor 901 executes various functional applications and data processing by running the computer programs and modules stored in the memory 902. The memory 902 can mainly include a program storage area and a data storage area. The program storage area can store the operating system, at least one computer program required for the function, etc.; the data storage area can store data created based on the use of the server 900.

[0190] In addition, the memory 902 may include a high-speed random access memory and a non-volatile memory, such as at least one disk storage device, a flash memory device, or other volatile solid-state storage device. Accordingly, the memory 902 may also include a memory controller to provide the processor 901 with access to the memory 902.

[0191] In this embodiment, the processor 901 in the server 900 loads instructions corresponding to one or more computer program processes into the memory 902 according to the following steps, and the processor 901 runs the computer program stored in the memory 902 to implement various functions as follows:

[0192] Receive user perception data collected by the vehicle;

[0193] Invoke a first instruction of a target agent in a server to generate a target atomic capability according to user perception data, where the target atomic capability is determined from a first atomic capability set reported by the vehicle to the server;

[0194] converting the first instruction into a second instruction recognizable by the vehicle;

[0195] A second instruction is sent to the vehicle so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

[0196] Optionally, when the processor 901 executes the first instruction to call the target agent in the server side to generate the target atomic capability based on the user perception data, it specifically performs the following: determining at least one available agent in the server side, where the available agent is an agent configured to be online; determining the available atomic capability of each available agent in the at least one available agent, where the available atomic capability is the atomic capability configured to be online; determining at least one candidate agent in the at least one available agent based on the first atomic capability set reported by the vehicle to the server side, where the available atomic capability of the candidate agent includes at least one atomic capability in the first atomic capability set; calling the first instruction to generate the target atomic capability of the target agent in the at least one candidate agent based on the user perception data.

[0197] Optionally, the processor 901 may also execute: in response to a first configuration instruction, configuring at least one of the multiple agents on the server side as an available agent that is online, and configuring the multiple agents on the server side other than the available agents as unavailable agents that are offline; and / or, in response to a second configuration instruction, configuring at least one of the multiple atomic capabilities on the server side as an available atomic capability that is online, and configuring the atomic capabilities on the server side other than the available atomic capabilities as unavailable atomic capabilities that are offline.

[0198] The effects that can be achieved by this embodiment can be found in the relevant embodiments of the above-mentioned vehicle control method, which will not be repeated here.

[0199] It should be understood that the device provided in the embodiment of the present application is used to execute the above-mentioned vehicle control method, and therefore can achieve the same effect as the above-mentioned implementation method.

[0200] In the case of an integrated unit, the device may include a processing module and a storage module. When the device is used in a vehicle, the processing module may be used to control and manage the vehicle's movements, while the storage module may be used to support the vehicle's execution of relevant program codes.

[0201] The processing module may be a processor or controller that implements or executes the various exemplary logic blocks, modules, and circuits described in conjunction with the disclosure of this application. The processor may also be a combination that implements computing functions, such as a combination of one or more microprocessors, a combination of a digital signal processing system (DSP) and a microprocessor, and the storage module may be a memory.

[0202] In addition, the device provided in the embodiment of the present application can specifically be a chip, component or module, and the chip may include a connected processor and memory; wherein the memory is used to store instructions, and when the processor calls and executes the instructions, the chip can execute a vehicle control method provided in the above embodiment.

[0203] An embodiment of the present application also provides a computer-readable storage medium, which stores computer program code. When the computer program code runs on a computer, the computer executes the above-mentioned related method steps to implement a vehicle control method provided by the above-mentioned embodiment.

[0204] This embodiment also provides a computer program product. When the computer program product is run on a computer, it enables the computer to execute the above-mentioned related steps to implement a vehicle control method provided by the above embodiment.

[0205] Among them, the device, computer-readable storage medium, computer program product or chip provided in this embodiment are all used to execute the corresponding methods provided above. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0206] Through the description of the above implementation methods, technical personnel in the relevant field can understand that for the convenience and simplicity of description, only the division of the above-mentioned functional modules is used as an example. In actual applications, the above-mentioned functions can be distributed and completed by different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0207] In the embodiments provided in this application, it should be understood that the disclosed devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of modules or units is only a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another device, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0208] The above content is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of the present application. Therefore, the scope of protection of the present application should be based on the scope of protection of the claims.

Claims

1. A vehicle control method, characterized in that: Applied to a vehicle, the method comprises: Sending user perception data collected by the vehicle to a server, so that the server can invoke a target agent in the server to generate a first instruction for a target atomic capability based on the user perception data, and converting the first instruction into a second instruction recognizable by the vehicle, where the target atomic capability is determined from a set of first atomic capabilities reported by the vehicle to the server; receiving the second instruction sent by the server; Execute the function corresponding to the target atomic capability according to the second instruction.

2. The method according to claim 1, characterized in that The method further comprises: obtaining an atomic capability of each of a plurality of applications in the vehicle; Determining a first atomic capability set according to the atomic capabilities of each of the applications; When the preset reporting conditions are met, the first atomic capability set is reported to the server.

3. The method according to claim 2, characterized in that The obtaining of the atomic capability of each application in the plurality of applications of the vehicle includes: Scanning atomic capabilities of each first application in a plurality of applications of the vehicle, where the first application is configured to be passively started; In response to the registration instruction of each second application among the plurality of applications, the atomic capability of each second application is determined according to the registration instruction of each second application, and the second application is configured to be automatically started.

4. The method according to claim 2, characterized in that The method further comprises: If the first atomic capability set is the atomic capability set obtained for the first time after the vehicle is started, determining that the preset reporting condition is met; If the first atomic capability set is different from the atomic capability set last acquired by the vehicle, it is determined that the preset reporting condition is met.

5. The method according to claim 1, wherein The executing the function corresponding to the target atomic capability according to the second instruction includes: A target application is determined among multiple applications of the vehicle according to the second instruction, and the target application is called to execute a function corresponding to the target atomic capability.

6. A vehicle control method, characterized in that: Applied to the server, the method includes: Receive user perception data collected by the vehicle; a first instruction for invoking a target agent in the server to generate a target atomic capability according to the user perception data, where the target atomic capability is determined from a first atomic capability set reported by the vehicle to the server; converting the first instruction into a second instruction recognizable by the vehicle; The second instruction is sent to the vehicle so that the vehicle executes the function corresponding to the target atomic capability according to the second instruction.

7. The method according to claim 6, characterized in that The first instruction of calling the target agent in the server to generate a target atomic capability according to the user perception data includes: Determine at least one available agent on the server, where the available agent is an agent configured to be online; Determining an available atomic capability of each of the at least one available agent, wherein the available atomic capability is an atomic capability configured to be online; Determining at least one candidate agent from at least one of the available agents based on the first atomic capability set reported by the vehicle to the server, wherein the available atomic capabilities of the candidate agent include at least one atomic capability in the first atomic capability set; A first instruction is provided for calling a target agent among at least one of the candidate agents to generate a target atomic capability according to the user perception data.

8. The method according to claim 7, characterized in that The method further comprises: In response to the first configuration instruction, at least one of the plurality of agents on the server is configured as an online available agent, and agents other than the available agent in the plurality of agents on the server are configured as offline unavailable agents; and / or, In response to the second configuration instruction, at least one of the multiple atomic capabilities of the server is configured as an online available atomic capability, and the atomic capabilities of the multiple atomic capabilities of the server other than the available atomic capability are configured as offline unavailable atomic capabilities.

9. A vehicle, characterized in that: include: a memory for storing executable program code; A processor is configured to call and run the executable program code from the memory, so that the vehicle executes the method according to any one of claims 1 to 5.

10. A server, characterized in that: include: a memory for storing executable program code; A processor, configured to call and run the executable program code from the memory, so that the server executes the method as claimed in any one of claims 6 to 8.

Images