Intelligent body control method, electronic device, vehicle, storage medium and program product

By determining the environment information of the on-board intelligent agent in the intelligent chat system and generating description text, and controlling the output using text optimization and control models, the problem that the intelligent agent cannot start automatically is solved, and intelligent automatic start and rich interaction are achieved, improving User experience.

CN118514708BActive Publication Date: 2024-11-19BYD CO LTD
View PDF 8 Cites 0 Cited by

Patent Information

Application Number
CN202411003613.0
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-07-25
Publication Date
2024-11-19
Estimated Expiration
2044-07-25

AI Technical Summary

Technical Problem

The existing intelligent chat system cannot automatically start based on environmental information, resulting in a single startup method and lack of intelligence and user experience improvement.

Method used

By determining the corresponding environmental information of the vehicle-mounted agent, including the vehicle's internal, external environment and network information, generating description text, and controlling the agent to output chat information based on the text optimization model and control model, to achieve automatic start-up and rich interaction methods.

Benefits of technology

It realizes that the agent automatically starts and outputs chat information based on environmental information, enriches the agent's startup method and improves user experience and interactivity.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118514708B_ABST
    Figure CN118514708B_ABST
Patent Text Reader

Abstract

The present disclosure relates to the field of vehicles, and more particularly to an intelligent body control method, an electronic device, a vehicle, a storage medium, and a program product, wherein the intelligent body control method includes: determining the environmental information corresponding to the on-board intelligent body; and controlling the on-board intelligent body to output chat information according to the environmental information. Through the above technical scheme, the environmental information corresponding to the on-board intelligent body can be determined, and the on-board intelligent body can be controlled to output chat information according to the environmental information. Thus, the on-board intelligent body can be automatically controlled to output chat information according to the environmental information, that is, the present disclosure can automatically start the on-board intelligent body and output chat information according to the environmental information. Compared with the related art in which the user starts the intelligent body and outputs the chat information, the present disclosure improves the user experience while enriching the intelligent body startup method.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present disclosure relates to the field of vehicles, and in particular, to an intelligent body control method, an electronic device, a vehicle, a storage medium, and a program product. Background Art

[0002] With the development of intelligent technology, intelligent agents are widely used in life, such as smart speakers, smart home systems, intelligent agent chat groups or smart phone assistants, etc. However, the intelligent agents in related technologies are generally actively started by users, which has the problem of a single starting method. Summary of the invention

[0003] The purpose of the present disclosure is to provide an intelligent body control method, an electronic device, a vehicle, a storage medium and a program product to solve the above-mentioned technical problems.

[0004] In order to achieve the above object, the present disclosure provides, in a first aspect, an intelligent agent control method, the intelligent agent control method comprising:

[0005] Determine the environmental information corresponding to the on-board intelligent agent;

[0006] According to the environmental information, the vehicle-mounted intelligent agent is controlled to output chat information.

[0007] Optionally, the determining of the environmental information corresponding to the vehicle-mounted intelligent agent includes:

[0008] The environmental information corresponding to the vehicle-mounted intelligent agent is determined according to at least one of the vehicle internal environmental information, the vehicle external environmental information and the network information.

[0009] Optionally, determining the environmental information corresponding to the vehicle-mounted intelligent agent according to at least one of the vehicle internal environmental information, the vehicle external environmental information and network information includes:

[0010] Determine the internal environment of the vehicle according to the internal temperature information and / or the internal humidity information of the vehicle;

[0011] Determine the external environment of the vehicle based on weather information;

[0012] The environmental information corresponding to the vehicle-mounted intelligent body is determined according to at least one of the vehicle internal environmental information, the vehicle external environmental information and network information of a preset category.

[0013] Optionally, controlling the in-vehicle intelligent agent to output chat information according to the environmental information includes:

[0014] Determining, according to the change of the environmental information, a description text for describing the change of the environmental information;

[0015] According to the description text, the vehicle-mounted intelligent agent is controlled to output chat information.

[0016] Optionally, determining, according to the change of the environmental information, a description text for describing the change of the environmental information includes:

[0017] Determining, according to the change of the environmental information, an initial description text for describing the change of the environmental information;

[0018] The initial description text is optimized based on a text optimization model to obtain a description text for describing changes in the environmental information, wherein the text optimization model is used to output an optimized description text based on the input initial description text.

[0019] Optionally, the vehicle-mounted intelligent agent includes a plurality of agents, and controlling the vehicle-mounted intelligent agent to output chat information according to the description text includes:

[0020] The description text is used as the initial target prompt text, and the following process is performed in a loop:

[0021] Determine a target vehicle-mounted intelligent agent related to the target prompt text from the multiple vehicle-mounted intelligent agents, and control the target vehicle-mounted intelligent agent to output chat information according to the target prompt text, at least using the chat information output by the target vehicle-mounted intelligent agent as a new target prompt text until a preset condition for stopping outputting the chat information is met.

[0022] Optionally, determining a target on-board intelligent agent related to the target prompt text from the plurality of on-board intelligent agents comprises:

[0023] Inputting the target prompt text into the control model to obtain the speech order information of the vehicle-mounted intelligent agent for the target prompt text, wherein the control model is used to output the speech order information according to the input target prompt text;

[0024] The first vehicle-mounted intelligent agent speaking in the speaking sequence information is determined as the target vehicle-mounted intelligent agent associated with the target prompt text.

[0025] Optionally, controlling the target vehicle-mounted intelligent agent to output chat information according to the target prompt text includes:

[0026] According to the target prompt text, determine a target document block from a preset knowledge base, wherein the preset knowledge base is used to store at least the document block of the target vehicle-mounted intelligent body, and the document block is obtained by splitting the user manual document of the target vehicle-mounted intelligent body;

[0027] Determine a new prompt text based on the target document block and the target prompt text;

[0028] The new prompt text is input into the target vehicle-mounted intelligent agent, and the target vehicle-mounted intelligent agent is controlled to output chat information.

[0029] Optionally, the preset knowledge base further stores a document block vector corresponding to the document block, and determining the target document block from the preset knowledge base according to the target prompt text includes:

[0030] Determining a text vector of the target prompt text;

[0031] Determine, based on the text vector, from the preset knowledge base a target document block vector whose similarity to the text vector is greater than a preset similarity threshold;

[0032] The document block corresponding to the target document block vector is determined as the target document block.

[0033] Optionally, controlling the in-vehicle intelligent agent to output chat information according to the environmental information includes:

[0034] Generate chat information according to the environmental information;

[0035] When the chat information passes the information verification, the vehicle-mounted intelligent agent is controlled to output the chat information.

[0036] Optionally, when the chat information passes the information verification, controlling the in-vehicle intelligent agent to output the chat information includes:

[0037] When the chat information does not include preset sensitive words and the chat information matches a user manual document of the in-vehicle intelligent body that outputs the chat information, the in-vehicle intelligent body is controlled to output the chat information.

[0038] Optionally, when the chat information passes the information verification, controlling the in-vehicle intelligent agent to output the chat information includes:

[0039] Based on the chat information and the information verification model, determining whether the chat information passes the verification result of the information verification, wherein the information verification model is used to output the verification result of the chat information according to the input chat information;

[0040] After the verification result indicates that the chat information has passed the information verification, the in-vehicle intelligent agent is controlled to output the chat information.

[0041] Optionally, it also includes:

[0042] When the chat information fails to pass the information verification, the vehicle-mounted intelligent agent that outputs the chat information is controlled to re-output the chat information.

[0043] Optionally, controlling the in-vehicle intelligent agent to output chat information according to the environmental information includes:

[0044] Generate chat information according to the environmental information;

[0045] Control the in-vehicle intelligent agent to output the chat information in the form of audio and / or text.

[0046] Optionally, controlling the vehicle-mounted intelligent agent to output the chat information in the form of audio includes:

[0047] Determine the tone information of the chat information based at least on the chat information and a tone determination model, wherein the tone determination model is at least used to output the tone information of the chat information according to the input chat information;

[0048] Based on the tone information, the in-vehicle intelligent agent is controlled to output the chat information in the form of audio.

[0049] Optionally, the method further comprises:

[0050] In the process of controlling the in-vehicle intelligent agent to output chat information according to the environmental information, if user feedback information is received, the in-vehicle intelligent agent is controlled to output chat information in response to the feedback information, or, according to the feedback information, the in-vehicle intelligent agent is controlled to perform a target action.

[0051] Optionally, controlling the in-vehicle intelligent agent to output chat information in response to the feedback information includes:

[0052] If the feedback information is received after the vehicle-mounted intelligent agent outputs the chat information, and the chat information includes a preset vehicle-mounted control instruction, when the feedback information indicates that the vehicle-mounted control instruction is not executed, the vehicle-mounted intelligent agent is controlled to output the chat information for the feedback information; or,

[0053] If the feedback information is received after the in-vehicle intelligent agent outputs the chat information, and the chat information does not include a preset in-vehicle control instruction, the in-vehicle intelligent agent is controlled to output the chat information for the feedback information.

[0054] Optionally, controlling the vehicle-mounted intelligent agent to perform a target action according to the feedback information includes:

[0055] If the feedback information is received after the in-vehicle intelligent agent outputs the chat information, and the chat information includes a preset in-vehicle control instruction, then when the feedback information represents the execution of the in-vehicle control instruction, the in-vehicle intelligent agent is controlled to execute the action corresponding to the in-vehicle control instruction.

[0056] Optionally, the on-board intelligent body corresponds to the control device of the vehicle, and the on-board intelligent body includes a window on-board intelligent body, a seat on-board intelligent body and / or an air-conditioning on-board intelligent body, the window on-board intelligent body corresponds to the window control device of the vehicle, the seat on-board intelligent body corresponds to the seat control device of the vehicle, and the air-conditioning on-board intelligent body corresponds to the air-conditioning control device of the vehicle.

[0057] A second aspect of the present disclosure provides an electronic device, the electronic device comprising:

[0058] processor;

[0059] a memory for storing processor-executable instructions;

[0060] Wherein, the processor is configured to: execute the steps of any method described in the first aspect.

[0061] A third aspect of the present disclosure provides a vehicle, comprising a processor and a memory for storing processor executable instructions, wherein the processor is configured to: execute the steps of any one of the methods described in the first aspect.

[0062] A fourth aspect of the present disclosure provides a non-temporary computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0063] A fifth aspect of the present disclosure provides a computer program product, comprising a computer program, which, when executed by a processor, implements the steps of any one of the methods described in the first aspect.

[0064] Through the above technical solution, the environmental information corresponding to the vehicle-mounted intelligent body can be determined, and the vehicle-mounted intelligent body can be controlled to output chat information according to the environmental information. Therefore, the vehicle-mounted intelligent body can be automatically controlled to output chat information according to the environmental information, that is, the present disclosure can automatically start the vehicle-mounted intelligent body and output chat information according to the environmental information. Compared with the related art, in which the intelligent body is started and the chat information is output by the user, the present disclosure can automatically start the vehicle-mounted intelligent body according to the environmental information, thereby enriching the intelligent body startup method and improving the user experience.

[0065] Other features and advantages of the present disclosure will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] The accompanying drawings are used to provide a further understanding of the present disclosure and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the present disclosure but do not constitute a limitation of the present disclosure. In the accompanying drawings:

[0067] Figure 1 is a flow chart of an intelligent agent control method according to an exemplary embodiment of the present disclosure;

[0068] Figure 2 is a flow chart showing a target vehicle-mounted intelligent agent outputting chat information according to an exemplary embodiment of the present disclosure;

[0069] Figure 3 is a flowchart of another intelligent agent control method according to an exemplary embodiment of the present disclosure;

[0070] Figure 4 is a structural block diagram of an intelligent body control device according to an exemplary embodiment of the present disclosure;

[0071] Figure 5 It is a schematic diagram of a functional block diagram of a vehicle according to an exemplary embodiment of the present disclosure. DETAILED DESCRIPTION

[0072] The specific implementation of the present disclosure is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described herein is only used to illustrate and explain the present disclosure, and is not used to limit the present disclosure.

[0073] It should be understood that the various steps described in the method embodiments of the present disclosure may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present disclosure is not limited in this respect.

[0074] The term "including" and its variations used herein are open inclusions, i.e., "including but not limited to". The term "based on" means "based at least in part on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". The relevant definitions of other terms will be given in the following description.

[0075] It should be noted that the concepts such as "first" and "second" mentioned in the present disclosure are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.

[0076] It should be noted that the modifications of "one" and "plurality" mentioned in the present disclosure are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise clearly indicated in the context, it should be understood as "one or more".

[0077] As mentioned in the background technology, with the development of intelligent technology, intelligent agents are widely used in life, such as smart speakers, smart home systems, smart phone assistants or intelligent agent chat groups. Among them, the intelligent agent chat group refers to a chat group composed of intelligent agents, each of which is an autonomous software program that can simulate the process of human dialogue and communication. In the intelligent agent chat group, each intelligent agent plays a virtual role and has an independent identity, characteristics and capabilities. They interact in real time through text, voice or images, and simulate the form of real dialogue.

[0078] However, the intelligent agents in the related art are generally started by the user, but cannot be started automatically according to the needs of the scene. For example, they cannot be started automatically according to environmental changes, thus there are problems such as lack of intelligence and a single starting method.

[0079] For example, in the invention patent with publication number CN109407925A, an interactive method, device and related equipment based on a chat robot are disclosed. In this patent, the robot cannot be automatically started according to changes in the environment.

[0080] In view of this, the present disclosure provides an intelligent body control method, an electronic device, a vehicle, a storage medium and a program product to solve the above technical problems.

[0081] The embodiments of the present disclosure are further explained below with reference to the accompanying drawings.

[0082] Figure 1 is a flow chart of an intelligent agent control method according to an exemplary embodiment of the present disclosure, referring to Figure 1 , the method may include the following steps:

[0083] S101: Determine the environmental information corresponding to the vehicle-mounted intelligent agent.

[0084] The vehicle-mounted intelligent body may include one or more. When the vehicle-mounted intelligent body includes one, the vehicle-mounted intelligent body may correspond to multiple control devices of the vehicle; when the vehicle-mounted intelligent body includes multiple, one vehicle-mounted intelligent body may correspond to one control device of the vehicle or may correspond to multiple control devices of the vehicle, and the embodiments of the present disclosure do not impose any limitation on this.

[0085] For example, when the onboard intelligent body includes multiple onboard intelligent bodies, and one onboard intelligent body corresponds to a control device of the vehicle, the onboard intelligent body may include a window onboard intelligent body, a seat onboard intelligent body and / or an air conditioning onboard intelligent body. Accordingly, the window onboard intelligent body may correspond to the window control device of the vehicle, the seat onboard intelligent body may correspond to the seat control device of the vehicle, and the air conditioning onboard intelligent body may correspond to the air conditioning control device of the vehicle. When the onboard intelligent body includes multiple onboard intelligent bodies, and one onboard intelligent body may correspond to multiple control devices of the vehicle, the onboard intelligent body may include a temperature adjustment onboard intelligent body and a ventilation adjustment intelligent body. Accordingly, the temperature adjustment onboard intelligent body may correspond to the air conditioning control device and the window control device of the vehicle, and the ventilation adjustment intelligent body may correspond to the air conditioning control device, the window control device and the sunroof control device of the vehicle.

[0086] S102: Control the vehicle-mounted intelligent agent to output chat information according to environmental information.

[0087] Through the above technical solution, the environmental information corresponding to the vehicle-mounted intelligent body can be determined, and the vehicle-mounted intelligent body can be controlled to output chat information according to the environmental information. Therefore, the vehicle-mounted intelligent body can be automatically controlled to output chat information according to the environmental information, that is, the present disclosure can automatically start the vehicle-mounted intelligent body and output chat information according to the environmental information. Compared with the related art, in which the intelligent body is started and the chat information is output by the user, the present disclosure can automatically start the vehicle-mounted intelligent body according to the environmental information, thereby enriching the intelligent body startup method and improving the user experience.

[0088] Among possible ways, determining the environment information corresponding to the onboard intelligent agent may include:

[0089] The environmental information corresponding to the vehicle-mounted intelligent agent is determined according to at least one of the vehicle internal environmental information, the vehicle external environmental information and the network information.

[0090] The vehicle internal environment information may include the vehicle temperature, vehicle humidity and / or vehicle noise, etc., the vehicle external environment may include the vehicle temperature, vehicle humidity and / or visibility, etc., and the network information may include network information of a preset category and / or network information with a network heat value greater than a preset network heat value, etc. The preset category of network information may be entertainment network information, sports network information or science and technology network information, etc., which may be set according to actual conditions, and the embodiment of the present disclosure does not impose any restrictions on this. Accordingly, determining the environment information corresponding to the vehicle-mounted intelligent body may be determining at least one of the vehicle temperature, vehicle humidity, vehicle noise, vehicle temperature, vehicle humidity, visibility, network information of a preset category and network information with a heat value greater than a preset heat value as the environment information corresponding to the vehicle-mounted intelligent body. For example, the vehicle temperature outside may be determined as the environment information corresponding to the vehicle-mounted intelligent body, the vehicle temperature outside and the vehicle temperature inside may be determined as the environment information corresponding to the vehicle-mounted intelligent body, or the vehicle-related network information may be determined as the environment information corresponding to the vehicle-mounted intelligent body.

[0091] In a possible manner, determining the environment information corresponding to the vehicle-mounted intelligent agent according to at least one of the vehicle internal environment information, the vehicle external environment information and the network information may include:

[0092] Determine the vehicle's internal environment based on the vehicle's internal temperature information and / or the vehicle's internal humidity information; determine the vehicle's external environment based on weather information; determine the environmental information corresponding to the on-board intelligent body based on at least one of the vehicle's internal environment information, the vehicle's external environment information and network information of a preset category.

[0093] For example, the in-vehicle temperature information may be acquired through an in-vehicle temperature sensor, and the in-vehicle temperature information acquired by the in-vehicle temperature sensor may be determined as the environmental information corresponding to the in-vehicle intelligent agent.

[0094] For example, the network information in the media platform can be obtained through the in-vehicle infotainment system in the vehicle, and when network information of a preset category appears in the network information, the network information of the preset category is determined as the environmental information corresponding to the in-vehicle intelligent body.

[0095] For example, the preset category of network information may be entertainment network information that the user is interested in. Thus, when network information in the media platform is obtained through the in-vehicle infotainment system in the vehicle, keywords can be extracted from the obtained network information through keyword extraction technology to determine whether the obtained network information is entertainment network information or whether it contains entertainment network information. When the obtained network information is entertainment network information or contains entertainment network information, the corresponding network information is determined as the environmental information corresponding to the in-vehicle intelligent body.

[0096] For example, weather information may be acquired through an in-vehicle infotainment system in a vehicle, and the acquired weather information may be determined as environmental information corresponding to the in-vehicle intelligent agent.

[0097] In a possible manner, controlling the vehicle-mounted intelligent agent to output chat information according to the environmental information may include:

[0098] According to the changes in the environmental information, a description text for describing the changes in the environmental information is determined; according to the description text, the vehicle-mounted intelligent agent is controlled to output chat information.

[0099] According to the change of the environmental information, the description text used to describe the change of the environmental information can be determined by detecting the change of the environmental information in real time or periodically, and once the change of the environmental information is detected, the description text used to describe the change of the environmental information is determined. Alternatively, the change of the environmental information can be detected in real time or periodically, and when the environmental information changes and the change meets the preset change, the description text used to describe the change of the environmental information is determined, and the embodiments of the present disclosure do not impose any restrictions on this.

[0100] For example, the temperature sensor collects the temperature inside the car in real time. When it is detected that the temperature inside the car rises by 3°C within 30 minutes, a description text can be determined according to the change of the temperature inside the car. For example, the description text is determined to be "the temperature inside the car rises by 3°C within 30 minutes".

[0101] For example, network information in a media platform can be acquired in real time. When network information of a preset category appears in the preset media platform, such as network information of a vehicle category, a description text can be determined based on changes in the network information in the preset media platform. For example, the description text is determined to be "Network information appears on the media platform: XXXX". Here, "XXXX" can be the title of network information of a preset category that appears on the media platform, or it can be a summary obtained by extracting content from network information of a preset category that appears on the media platform using keyword extraction technology, or it can be other, and the embodiments of the present disclosure do not impose any restrictions on this.

[0102] In a possible manner, according to the change of the environmental information, determining the description text used to describe the change of the environmental information may include:

[0103] According to the changes in environmental information, an initial description text for describing the changes in environmental information is determined; the initial description text is optimized based on a text optimization model to obtain a description text for describing the changes in environmental information, wherein the text optimization model is used to output an optimized description text based on the input initial description text.

[0104] The text optimization model may be an existing model in the related art, or an improved model of an existing model, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the text optimization model may be a large language model, a natural language understanding model, or a convolutional neural network.

[0105] For example, when the temperature sensor detects that the temperature inside the car rises by 3°C within 30 minutes, an initial description text "The temperature inside the car rises by 3°C within 30 minutes" can be generated based on the change in the temperature inside the car. Afterwards, the initial description text "The temperature inside the car rises by 3°C within 30 minutes" is input into the large language model, and the optimized description text "During the half-hour journey, the temperature inside the car rose gently by 3°C" is output.

[0106] Through the above technical solution, the initial description text can be optimized, thereby reducing grammatical errors, spelling errors, and logical errors in the initial description text, so that when the vehicle-mounted intelligent agent is subsequently controlled to output chat information according to the description text, the vehicle-mounted intelligent agent can understand the semantics of the description text more quickly and accurately.

[0107] In a possible manner, the vehicle-mounted intelligent agent may include multiple ones. Accordingly, according to the description text, controlling the vehicle-mounted intelligent agent to output the chat information may include:

[0108] The description text is used as the initial target prompt text, and the following process is executed in a loop: a target vehicle-mounted intelligent agent related to the target prompt text is determined from multiple vehicle-mounted intelligent agents, and according to the target prompt text, the target vehicle-mounted intelligent agent is controlled to output chat information, and at least the chat information output by the target vehicle-mounted intelligent agent is used as the new target prompt text until the preset condition for stopping outputting the chat information is met.

[0109] The condition for stopping the output of chat information can be set according to actual conditions, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the condition for stopping the output of chat information can be that the single chat duration reaches a preset chat duration, it is recognized through semantic recognition or intention recognition that the user has no intention to continue chatting, and / or it is recognized that the output chat information has solved the user's problem, etc.

[0110] It should be understood that a single chat message provides limited information, so only a single output of the vehicle-mounted intelligent body is used as the target prompt text, and the target vehicle-mounted intelligent body is determined based on the target prompt text. There is a problem that the chat message output by the target vehicle-mounted intelligent body is incoherent with the previously output chat message. Therefore, in a possible way, the chat message output by the vehicle-mounted intelligent body can be recorded during the startup of the vehicle-mounted intelligent body, and the recorded chat record can be used as a new target prompt text, so that the target prompt text can provide richer context information to maintain the continuity of the conversation.

[0111] In a possible manner, determining a target onboard intelligent agent related to the target prompt text from a plurality of onboard intelligent agents may include:

[0112] The target prompt text is input into the control model to obtain the speech order information of the on-board intelligent agent for the target prompt text, wherein the control model is used to output the speech order information according to the input target prompt text; the on-board intelligent agent that speaks first in the speech order information is determined as the target on-board intelligent agent related to the target prompt text.

[0113] The control model may be an existing model in the related art, or may be an improved model of the existing model, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the control model may be a large language model.

[0114] In order to obtain the speech order information of the on-board intelligent agent for the target prompt text based on the target prompt text and the control model, training data can also be constructed to fine-tune or train the control model.

[0115] For example, for each vehicle-mounted intelligent agent, the user manual document of the vehicle-mounted intelligent agent can be split to obtain multiple document blocks, and then the document blocks including the core functions of the vehicle-mounted intelligent agent and the name of the vehicle-mounted intelligent agent are extracted from the multiple document blocks, thereby obtaining a sample library including multiple core functions of the vehicle-mounted intelligent agent and multiple names of the vehicle-mounted intelligent agent. After that, a sample prompt text is constructed, and the control model is trained based on the sample prompt text, the label for indicating the speech order information, and the sample library until a preset number of iterations is reached. Therefore, after the target prompt text is input into the control model, the control model can infer the speech order information of the vehicle-mounted intelligent agent for the target prompt text based on the information in the sample library.

[0116] In a possible manner, according to the target prompt text, controlling the target vehicle-mounted intelligent agent to output chat information may include:

[0117] According to the target prompt text, a target document block is determined from a preset knowledge base, wherein the preset knowledge base is at least used to store the document block of the target vehicle-mounted intelligent body, and the document block is obtained by splitting the user manual document of the target vehicle-mounted intelligent body; a new prompt text is determined based on the target document block and the target prompt text; the new prompt text is input into the target vehicle-mounted intelligent body, and the target vehicle-mounted intelligent body is controlled to output chat information.

[0118] The target document block may be a document block with similar semantics to the target prompt text, or a document block in which keywords in the target document block match keywords in the target prompt text, or other documents, which are not limited in any way by the embodiments of the present disclosure.

[0119] Among them, determining a new prompt text based on the target document block and the target prompt text may be to pre-set a prompt text template, and after obtaining the target document block, the prompt text is obtained by filling the target document block and the target prompt text into the prompt text template.

[0120] For example, the prompt text template may be set as: Please answer the target prompt text "XXXX" based on the content in the target document block "XXX".

[0121] The splitting method of the manual document can be set according to the actual situation, and the embodiment of the present disclosure does not impose any restrictions on this. For example, the first-level title in the manual document can be used as a splitting point to split the manual document to obtain multiple document blocks. Each functional module in the manual document can also be used as a splitting point to split the manual document to obtain multiple document blocks.

[0122] In a possible manner, the preset knowledge base also stores document block vectors corresponding to the document blocks. Accordingly, determining the target document block from the preset knowledge base according to the target prompt text may include:

[0123] Determine a text vector of a target prompt text; determine a target document block vector whose similarity with the text vector is greater than a preset similarity threshold from a preset knowledge base based on the text vector; and determine a document block corresponding to the target document block vector as a target document block.

[0124] The text vector of the target prompt text and / or the target document block vector can be obtained by inputting the target prompt text and / or the target document block vector into a text embedding model (Text Embedding Models), a bag of words model (Bag of Words, BoW) or a word embedding model (Word2Vec).

[0125] For example, Figure 2As shown, the user manual document of each vehicle-mounted intelligent agent can be split in advance to obtain multiple document blocks, and then each document block is input into a text embedding model to obtain a document block vector. Finally, all document block vectors and document blocks are stored in a database to obtain a preset knowledge base for each vehicle-mounted intelligent agent. Afterwards, when the target prompt text is transmitted to the target vehicle-mounted intelligent agent, the target prompt text is first input into the text embedding model to obtain a text vector of the target prompt text, and then a vector search is performed based on the text vector and the preset knowledge base of the target vehicle-mounted intelligent agent to obtain a target document block vector whose similarity with the text vector is greater than a preset similarity threshold from the preset knowledge base of the target vehicle-mounted intelligent agent, and the target document block is determined based on the target document block vector. Finally, a new prompt text is generated based on the target document block and the target prompt text, and the new prompt text is input into the target vehicle-mounted intelligent agent to generate a chat message.

[0126] The similarity between the text vector and the document block vector may be calculated based on Euclidean distance, cosine similarity or Manhattan distance, which is determined according to actual conditions and is not limited in any way in the embodiments of the present disclosure.

[0127] In a possible manner, controlling the vehicle-mounted intelligent agent to output chat information according to the environmental information may include:

[0128] Generate chat information based on environmental information; when the chat information passes information verification, control the on-board intelligent agent to output the chat information.

[0129] Among them, the information verification can be used to verify whether there are preset sensitive words in the chat information, whether the logic of the chat information is smooth, and whether the grammar of the chat information is correct, etc., which is determined according to the actual situation, and the embodiments of the present disclosure do not impose any restrictions on this. For example, after the chat information is generated, the chat information can be semantically verified. When the semantic verification result indicates that the chat information does not contain hate information, biased information, and radical information, it is determined that the chat information passes the information verification, and the vehicle-mounted intelligent body is controlled to output the chat information.

[0130] Through the above technical solution, the chat information can be verified, and when the chat information passes the information verification, the vehicle-mounted intelligent body is controlled to output the chat information to the outside, thereby reducing the spread of erroneous information and improving user experience.

[0131] In a possible manner, when the chat information passes the information verification, controlling the vehicle-mounted intelligent agent to output the chat information may include:

[0132] When the chat information does not include preset sensitive words and the chat information matches the user manual document of the in-vehicle intelligent agent that outputs the chat information, the in-vehicle intelligent agent is controlled to output the chat information.

[0133] The preset sensitive words may be set according to actual conditions, and the embodiments of the present disclosure do not impose any limitation on this.

[0134] Among them, the matching of the chat information with the user manual document of the on-board intelligent agent that outputs the chat information means that the chat information output by the on-board intelligent agent must correspond to the existing functions in the user manual document.

[0135] For example, the air-conditioning vehicle intelligent agent has the function of controlling the air-conditioning, but does not have the function of controlling devices such as windows or wipers. Therefore, if the chat information output by the air-conditioning vehicle intelligent agent is "Do you need me to help you adjust the air-conditioning to 25°C?", it is determined that the chat information output by the air-conditioning vehicle intelligent agent matches the user manual document of the air-conditioning vehicle intelligent agent. If the chat information output by the air-conditioning vehicle intelligent agent is "Do you need me to help you open the window?", it is determined that the chat information output by the air-conditioning vehicle intelligent agent does not match the user manual document of the air-conditioning vehicle intelligent agent.

[0136] In a possible manner, when the chat information passes the information verification, controlling the vehicle-mounted intelligent agent to output the chat information may include:

[0137] Based on the chat information and the information verification model, determine whether the chat information passes the verification result of the information verification, wherein the information verification model is used to output the verification result of the chat information based on the input chat information; after the verification result indicates that the chat information passes the information verification, control the on-board intelligent agent to output the chat information.

[0138] The information verification model may be an existing model in the related art, or an improved model of an existing model, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the information verification model may be a grammar verification model, a semantic verification model, a logic verification model, or a large language model.

[0139] Among the possible methods, they also include:

[0140] When the chat information fails to pass the information verification, the vehicle-mounted intelligent agent that outputs the chat information is controlled to re-output the chat information.

[0141] For example, the air-conditioning vehicle intelligent agent has the function of controlling the air-conditioning, but does not have the function of controlling the windows or wipers. Therefore, if the chat message output by the air-conditioning vehicle intelligent agent is "Do you need me to open the window for you?", it is determined that the chat message output by the air-conditioning vehicle intelligent agent does not match the user manual document of the air-conditioning vehicle intelligent agent, and the target prompt text can be transmitted to the air-conditioning vehicle intelligent agent again to control the air-conditioning vehicle intelligent agent to re-output the chat message.

[0142] In a possible manner, controlling the vehicle-mounted intelligent agent to output chat information according to the environmental information may include:

[0143] Generate chat information based on environmental information; control the on-board intelligent agent to output the chat information in the form of audio and / or text.

[0144] When the vehicle-mounted intelligent agent outputs the chat information in the form of audio, the generated chat information can be sent to the vehicle-mounted speaker to output the chat information through the vehicle-mounted speaker. When the vehicle-mounted intelligent agent outputs the chat information in the form of text, the generated chat information can be sent to the central control screen to display the chat information through the central control screen.

[0145] Among possible ways, controlling the vehicle-mounted intelligent agent to output chat information in the form of audio may include:

[0146] At least based on the chat information and the tone determination model, determine the tone information of the chat information, wherein the tone determination model is at least used to output the tone information of the chat information according to the input chat information; based on the tone information, control the on-board intelligent body to output the chat information in the form of audio.

[0147] The tone determination model may be an existing model in the related art, or may be an improved model of an existing model, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the tone determination model may be a sentiment analysis model or a large language model.

[0148] For example, after the chat information is generated, the chat information can be input into a sentiment analysis model to obtain sentiment tendencies, such as humor, maturity, or shyness, and the sentiment tendencies are determined as tone information of the chat information.

[0149] It should be understood that this is only an illustrative description and does not constitute a limitation to the solution. In a possible manner, the tone information of the chat message can also be randomly assigned by a tone determination model.

[0150] Through the above technical solution, the tone information of the chat information can be determined before the chat information is output, so that the vehicle-mounted intelligent body can be controlled to output the chat information in the form of audio based on the tone information. In this way, different chat messages can have different tones, which increases the fun of the chat process and further improves the user experience.

[0151] In a possible manner, the intelligent agent control method may further include:

[0152] In the process of controlling the on-board intelligent agent to output chat information according to environmental information, if user feedback information is received, the on-board intelligent agent is controlled to output chat information in response to the feedback information, or the on-board intelligent agent is controlled to perform a target action according to the feedback information.

[0153] For example, when receiving feedback information from a user, the feedback information may be subjected to intent recognition. When the intent recognition result indicates that the user is inclined to chat, the in-vehicle intelligent agent is controlled to output chat information in response to the feedback information. When the intent recognition result indicates that the user is inclined to have the in-vehicle intelligent agent perform a target action, the in-vehicle intelligent agent is controlled to perform the target action. Alternatively, when receiving feedback information from a user, it may be determined whether there is a preset in-vehicle control instruction in the feedback information. When there is a preset in-vehicle control instruction in the feedback information, the in-vehicle intelligent agent is controlled to perform the target action corresponding to the in-vehicle control instruction. When there is no preset in-vehicle control instruction in the feedback information, the in-vehicle intelligent agent is controlled to output chat information in response to the feedback information.

[0154] In a possible manner, controlling the vehicle-mounted intelligent agent to output chat information for feedback information may include:

[0155] If feedback information is received after the on-board intelligent body outputs chat information, and the chat information includes a preset on-board control instruction, then when the feedback information indicates that the on-board control instruction is not executed, the on-board intelligent body is controlled to output chat information in response to the feedback information; or, if feedback information is received after the on-board intelligent body outputs chat information, and the chat information does not include a preset on-board control instruction, then the on-board intelligent body is controlled to output chat information in response to the feedback information.

[0156] The vehicle control instruction may be determined according to actual conditions, and the embodiments of the present disclosure do not impose any restrictions on this. For example, the vehicle control instruction may include opening a window, turning on an air conditioner, or adjusting the air conditioner temperature.

[0157] For example, after the air-conditioning in-vehicle intelligent agent outputs "Do you need me to help you adjust the air-conditioning temperature to 25℃?" and receives the user's feedback information "No", the in-vehicle intelligent agent can be controlled to output a chat message in response to the feedback information. For example, the window intelligent agent can be controlled to output "Do you need me to help you open the window?"

[0158] For example, after the on-board intelligent agent outputs "The air quality has been really good in the past week" through the car window, and receives feedback from the user "Yes, even my mood has become happier", the on-board intelligent agent can be controlled to output a chat message in response to the feedback information. For example, the multimedia intelligent agent can be controlled to output "It would be even better to play a cheerful song at this time."

[0159] Among them, it is worth mentioning that if the on-board intelligent body receives feedback information from the user when outputting chat information, then after the chat information is output, the on-board intelligent body can be controlled to output chat information for the feedback information, or the output of the chat information can be interrupted and the on-board intelligent body can be controlled to output chat information for the feedback information. The embodiments of the present disclosure do not impose any restrictions on this.

[0160] Among the possible ways, controlling the vehicle-mounted intelligent agent to perform the target action according to the feedback information may include:

[0161] If feedback information is received after the in-vehicle intelligent agent outputs the chat information, and the chat information includes a preset in-vehicle control instruction, then when the feedback information represents the execution of the in-vehicle control instruction, the in-vehicle intelligent agent is controlled to execute the action corresponding to the in-vehicle control instruction.

[0162] For example, after the air-conditioning vehicle-mounted intelligent agent outputs "Do you need me to help you adjust the air-conditioning temperature to 25°C?" and receives the user's feedback information "Yes", the air-conditioning vehicle-mounted intelligent agent can be controlled to adjust the air-conditioning temperature to 25°C.

[0163] It should be understood that this is only an illustrative description and is not intended to limit the pairing scheme. In a possible manner, in order to more accurately execute the target action for the feedback information, after receiving the feedback information, it is also possible to determine whether the feedback information contains a preset vehicle control instruction. When the feedback information contains the vehicle control instruction, the target action corresponding to the vehicle control instruction in the feedback information is executed.

[0164] For example, after the air-conditioning vehicle-mounted intelligent agent outputs "the temperature has risen from 27°C to 30°C", and receives the user's feedback information "Yes, it's so hot, please help me open the window", the window vehicle-mounted intelligent agent can be controlled to open the window.

[0165] For example, after the air-conditioning vehicle-mounted intelligent agent outputs "The temperature has risen to 32°C, do you need to turn on the air conditioner?", and receives the user's feedback information "Please help me open the window", the window vehicle-mounted intelligent agent can be controlled to open the window.

[0166] Among them, it is worth mentioning that if the on-board intelligent body receives feedback information from the user when outputting the chat information, the on-board intelligent body can be controlled to perform the target action for the feedback information after the chat information is output, or the output of the chat information can be interrupted and the on-board intelligent body can be controlled to perform the target action for the feedback information. The embodiments of the present disclosure do not impose any restrictions on this.

[0167] Through the above technical solution, the on-board intelligent body can be controlled to perform target actions. Therefore, during driving, the user can control the on-board intelligent body through voice to operate the on-board equipment, thereby improving driving safety and further improving the user experience.

[0168] To facilitate understanding of the intelligent body control method provided by the present disclosure, an implementation method of the intelligent body control method provided by the present disclosure is described below.

[0169] For example, Figure 3As shown, when the in-vehicle intelligent body is not started, there are two ways to start the intelligent body: Method 1: the in-vehicle intelligent body automatically starts according to environmental information; Method 2: the user actively starts the in-vehicle intelligent body.

[0170] For method 1, after the on-board intelligent agent perceives the change of environmental information, it generates an initial description text for describing the change of environmental information according to the change of environmental information, and transmits the initial description text to the text optimization model for optimization to obtain the optimized description text. Secondly, the description text is transmitted to the control model, and the control model determines the target on-board intelligent agent that needs to output chat information for the description text, and transmits the description text to the target on-board intelligent agent. Thirdly, the target on-board intelligent agent generates chat information according to the description text, and transmits the chat information to the information verification model to verify whether the chat information is compliant. If the chat information is not compliant, the target on-board intelligent agent is controlled to regenerate chat information according to the description text. If the chat information is compliant, the chat information is transmitted to the tone determination model to obtain the tone information for the chat information, and the chat information is output based on the tone information.

[0171] After the chat information is output, in order to be able to control the vehicle-mounted intelligent body to perform accurate actions (output chat information or control vehicle-mounted equipment) based on the user's feedback information, after the chat information is output, it is also possible to judge whether the chat information contains a preset vehicle-mounted control instruction. If the chat information includes a vehicle-mounted control instruction, it is possible to determine whether the user agrees to execute the vehicle-mounted control instruction within a preset time. If the user agrees to execute the vehicle-mounted control instruction, the control instruction is converted into a control instruction text, and the control instruction text is input into the target vehicle-mounted control instruction to control the target vehicle-mounted control to execute the action corresponding to the control instruction text. If the user does not agree to execute the vehicle-mounted control instruction, the control management model re-determines the target vehicle-mounted intelligent body to speak based on the user's feedback information and / or the historical chat information record (the chat information output during the process of starting the vehicle-mounted intelligent body this time). If the chat information does not include a vehicle-mounted control instruction, it is possible to determine whether the preset condition for stopping the output of the chat information is met. When the preset condition for stopping the output of the chat information is not met, the management model re-determines the target vehicle-mounted intelligent body to speak based on the output chat information and / or the historical chat information record (the chat information output during the process of starting the vehicle-mounted intelligent body this time). If the preset conditions for stopping outputting chat information are met, the on-board intelligent agent is controlled to end the chat.

[0172] The difference between method 2 and method 1 is that the initial description text is not obtained according to the change of environmental information, but is obtained by converting the user's audio signal into a text signal. Therefore, the control process of method 2 can refer to the control process of method 1, and will not be repeated here.

[0173] Based on the same concept, the embodiment of the present disclosure also provides an intelligent body control device, such as Figure 4 As shown, the intelligent body control device 400 may include:

[0174] A determination module 401 is used to determine the environment information corresponding to the vehicle-mounted intelligent agent;

[0175] The first control module 402 is used to control the vehicle-mounted intelligent agent to output chat information according to environmental information.

[0176] Through the above-mentioned intelligent agent control device 400, the environmental information corresponding to the vehicle-mounted intelligent agent can be determined, and the vehicle-mounted intelligent agent can be controlled to output chat information according to the environmental information. Therefore, the vehicle-mounted intelligent agent can be automatically controlled to output chat information according to the environmental information, that is, the present disclosure can automatically start the vehicle-mounted intelligent agent and output chat information according to the environmental information. Compared with the related art, in which the user starts the intelligent agent and outputs the chat information, the present disclosure can automatically start the vehicle-mounted intelligent agent according to the environmental information, thereby enriching the intelligent agent startup method and improving the user experience.

[0177] In a possible manner, the determination module 401 may include:

[0178] The first determination submodule is used to determine the environmental information corresponding to the vehicle-mounted intelligent body according to at least one of the vehicle internal environmental information, the vehicle external environmental information and the network information.

[0179] In a possible manner, the first determining submodule may include:

[0180] A first determining unit, configured to determine the internal environment of the vehicle according to the in-vehicle temperature information and / or the in-vehicle humidity information;

[0181] A second determining unit, used to determine the external environment of the vehicle according to weather information;

[0182] The third determination unit is used to determine the environmental information corresponding to the vehicle-mounted intelligent body according to at least one of the vehicle internal environmental information, the vehicle external environmental information and the network information of a preset category.

[0183] In a possible manner, the first control module 402 may include:

[0184] The second determination submodule is used to determine a description text for describing the change of the environmental information according to the change of the environmental information;

[0185] The first control submodule is used to control the vehicle-mounted intelligent agent to output chat information according to the description text.

[0186] In a possible manner, the second determining submodule may include:

[0187] A fourth determining unit, used to determine an initial description text for describing the change of the environmental information according to the change of the environmental information;

[0188] The optimization unit is used to optimize the initial description text based on the text optimization model to obtain the description text used to describe the change of the environmental information, wherein the text optimization model is used to output the optimized description text according to the input initial description text.

[0189] In a possible manner, the on-board intelligent agent includes multiple, and the first control submodule can be used to use the description text as the initial target prompt text, and cyclically execute the following process: determine the target on-board intelligent agent related to the target prompt text from multiple on-board intelligent agents, and control the target on-board intelligent agent to output chat information according to the target prompt text, at least use the chat information output by the target on-board intelligent agent as the new target prompt text until the preset condition for stopping outputting the chat information is met.

[0190] In a possible manner, the first control submodule may include:

[0191] An input unit, used to input the target prompt text into the control model, and obtain the speech order information of the vehicle-mounted intelligent agent for the target prompt text, wherein the control model is used to output the speech order information according to the input target prompt text;

[0192] The fifth determining unit is used to determine the first vehicle-mounted intelligent agent that speaks in the speaking order information as the target vehicle-mounted intelligent agent associated with the target prompt text.

[0193] In a possible manner, the first control submodule may include:

[0194] A sixth determination unit is used to determine a target document block from a preset knowledge base according to the target prompt text, wherein the preset knowledge base is used to store at least a document block of a target vehicle-mounted intelligent body, and the document block is obtained by splitting a user manual document of the target vehicle-mounted intelligent body;

[0195] A seventh determining unit, configured to determine a new prompt text based on the target document block and the target prompt text;

[0196] The first control unit is used to input new prompt text into the target vehicle-mounted intelligent body and control the target vehicle-mounted intelligent body to output chat information.

[0197] In a possible manner, the preset knowledge base further stores document block vectors corresponding to the document blocks. Accordingly, the sixth determining unit may include:

[0198] A first determination subunit, used to determine a text vector of a target prompt text;

[0199] A second determination subunit is used to determine, based on the text vector, from a preset knowledge base a target document block vector whose similarity with the text vector is greater than a preset similarity threshold;

[0200] The third determining subunit is used to determine the document block corresponding to the target document block vector as the target document block.

[0201] In a possible manner, the first control module 402 may include:

[0202] The first generation submodule is used to generate chat information according to the environmental information;

[0203] The second control submodule is used to control the vehicle-mounted intelligent agent to output the chat information when the chat information passes the information verification.

[0204] In a possible manner, the second control submodule may include:

[0205] The second control unit is used to control the in-vehicle intelligent body to output the chat information when the chat information does not include preset sensitive words and the chat information matches the user manual document of the in-vehicle intelligent body that outputs the chat information.

[0206] In a possible manner, the second control submodule may include:

[0207] an eighth determination unit, configured to determine whether the chat information passes the verification result of the information verification based on the chat information and the information verification model, wherein the information verification model is configured to output the verification result of the chat information according to the input chat information;

[0208] The third control unit is used to control the vehicle-mounted intelligent agent to output the chat information after the verification result indicates that the chat information has passed the information verification.

[0209] In a possible manner, the second control submodule may further include:

[0210] The fourth control unit is used to control the vehicle-mounted intelligent entity that outputs the chat information to re-output the chat information when the chat information fails to pass the information verification.

[0211] In a possible manner, the first control module 402 may include:

[0212] The first generation submodule is used to generate chat information according to the environmental information;

[0213] The third control submodule is used to control the vehicle-mounted intelligent agent to output chat information in the form of audio and / or text.

[0214] In a possible manner, the third control submodule may include:

[0215] a ninth determination unit, configured to determine the tone information of the chat information based at least on the chat information and the tone determination model, wherein the tone determination model is configured to output the tone information of the chat information based on at least the input chat information;

[0216] The fifth control unit is used to control the in-vehicle intelligent agent to output chat information in the form of audio based on the tone information.

[0217] In a possible manner, the intelligent body control device 400 may further include:

[0218] The second control module is used to control the on-board intelligent body to output chat information according to environmental information. If feedback information from the user is received during the process of controlling the on-board intelligent body to output chat information, or to control the on-board intelligent body to perform a target action according to the feedback information.

[0219] In a possible manner, the second control module may include:

[0220] A fourth control submodule is used for, if feedback information is received after the vehicle-mounted intelligent agent outputs the chat information, and the chat information includes a preset vehicle-mounted control instruction, when the feedback information indicates that the vehicle-mounted control instruction is not executed, controlling the vehicle-mounted intelligent agent to output the chat information for the feedback information; or,

[0221] The fifth control submodule is used to control the on-board intelligent body to output chat information in response to the feedback information if feedback information is received after the on-board intelligent body outputs the chat information, and the chat information does not include a preset on-board control instruction.

[0222] In a possible manner, the second control module may further include:

[0223] The sixth control submodule is used to control the on-board intelligent body to execute the action corresponding to the on-board control instruction if feedback information is received after the on-board intelligent body outputs the chat information, and the chat information includes a preset on-board control instruction, when the feedback information represents the execution of the on-board control instruction.

[0224] In a possible manner, the on-board intelligent body corresponds to the control device of the vehicle, and the on-board intelligent body includes a window on-board intelligent body, a seat on-board intelligent body and / or an air-conditioning on-board intelligent body. The window on-board intelligent body corresponds to the window control device of the vehicle, the seat on-board intelligent body corresponds to the seat control device of the vehicle, and the air-conditioning on-board intelligent body corresponds to the air-conditioning control device of the vehicle.

[0225] Regarding the intelligent body control device 400 in the above embodiment, the specific manner in which each module performs operations has been described in detail in the embodiment of the method, and will not be elaborated here.

[0226] Based on the same concept, an embodiment of the present disclosure further provides an electronic device, the electronic device comprising:

[0227] processor;

[0228] a memory for storing processor-executable instructions;

[0229] Wherein, the processor is configured to: execute the steps of any of the above-mentioned intelligent agent control methods.

[0230] Based on the same concept, an embodiment of the present disclosure also provides a vehicle, the vehicle comprising a processor and a memory for storing instructions executable by the processor, wherein the processor is configured to: execute the steps of any of the above-mentioned intelligent agent control methods.

[0231] Based on the same concept, the embodiment of the present disclosure also provides a non-temporary computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, the steps of any of the above-mentioned intelligent agent control methods are implemented.

[0232] Based on the same concept, an embodiment of the present disclosure also provides a computer program product, including a computer program, which implements the steps of any of the above-mentioned intelligent agent control methods when executed by a processor.

[0233] Please refer to Figure 5 , Figure 5 5 is a functional block diagram of a vehicle according to an exemplary embodiment. The vehicle 500 may include various subsystems, such as an infotainment system 510, a perception system 520, a decision control system 530, a drive system 540, and a computing platform 550. The vehicle 500 may also include more or fewer subsystems, and each subsystem may include multiple components. In addition, each subsystem and each component of the vehicle 500 may be interconnected by wire or wirelessly.

[0234] In some embodiments, the infotainment system 510 may include a communication system, an entertainment system, and a navigation system, among others.

[0235] The perception system 520 may include several sensors for sensing information about the environment around the vehicle 500. For example, the perception system 520 may include a global positioning system (the global positioning system may be a GPS system, or a Beidou system or other positioning systems), an inertial measurement unit (IMU), a laser radar, a millimeter wave radar, an ultrasonic radar, and a camera.

[0236] The decision control system 530 may include a computing system, a vehicle controller, a steering system, a throttle, and a braking system.

[0237] The drive system 540 may include components that provide powered motion for the vehicle 500. In one embodiment, the drive system 540 may include an engine, an energy source, a transmission system, and wheels. The engine may be an internal combustion engine, an electric motor, an air compression engine, or a combination of one or more thereof. The engine is capable of converting energy provided by the energy source into mechanical energy.

[0238] Some or all functions of the vehicle 500 are controlled by a computing platform 550. The computing platform 550 may include at least one processor 551 and a memory 552, and the processor 551 may execute instructions 553 stored in the memory 552.

[0239] The processor 551 may be any conventional processor, such as a commercially available CPU. The processor may also include a graphics processor (Graphic Process Unit, GPU), a field programmable gate array (Field Programmable Gate Array, FPGA), a system on chip (System on Chip, SOC), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) or a combination thereof.

[0240] The memory 552 may be implemented by any type of volatile or nonvolatile memory device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk or optical disk.

[0241] In addition to instructions 553 , memory 552 may also store data, such as road maps, route information, vehicle location, direction, speed, etc. The data stored in memory 552 may be used by computing platform 550 .

[0242] In the embodiment of the present disclosure, the processor 551 can execute instruction 553 to complete all or part of the steps of the above-mentioned intelligent agent control method.

[0243] The preferred embodiments of the present disclosure are described in detail above in conjunction with the accompanying drawings; however, the present disclosure is not limited to the specific details in the above embodiments. Within the technical concept of the present disclosure, a variety of simple modifications can be made to the technical solution of the present disclosure, and these simple modifications all fall within the protection scope of the present disclosure.

[0244] It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the present disclosure will not further describe various possible combinations.

[0245] In addition, various embodiments of the present disclosure may be arbitrarily combined, and as long as they do not violate the concept of the present disclosure, they should also be regarded as the contents disclosed by the present disclosure.

Claims

1. An intelligent agent control method, characterized in that: The intelligent agent control method comprises: Determine environmental information corresponding to the vehicle-mounted intelligent agent, wherein the vehicle-mounted intelligent agent includes a plurality of vehicles; Determining, according to the change of the environmental information, a description text for describing the change of the environmental information; The description text is used as the initial target prompt text, and the following process is performed in a loop: The target prompt text is input into the control model to obtain the speech order information of the on-board intelligent agent for the target prompt text, wherein the control model is used to output the speech order information according to the input target prompt text; the on-board intelligent agent that speaks first in the speech order information is determined as the target on-board intelligent agent related to the target prompt text, and according to the target prompt text, the target on-board intelligent agent is controlled to output chat information, at least the chat information output by the target on-board intelligent agent is used as the new target prompt text until the preset condition for stopping the output of the chat information is met.

2. The intelligent agent control method according to claim 1, characterized in that: The determining of the environmental information corresponding to the vehicle-mounted intelligent agent includes: The environmental information corresponding to the vehicle-mounted intelligent agent is determined according to at least one of the vehicle internal environmental information, the vehicle external environmental information and the network information.

3. The intelligent agent control method according to claim 2, characterized in that: The determining, based on at least one of vehicle internal environment information, vehicle external environment information, and network information, environment information corresponding to the vehicle-mounted intelligent agent comprises: Determine the internal environment of the vehicle according to the internal temperature information and / or the internal humidity information of the vehicle; Determine the external environment of the vehicle based on weather information; The environmental information corresponding to the vehicle-mounted intelligent body is determined according to at least one of the vehicle internal environmental information, the vehicle external environmental information and network information of a preset category.

4. The intelligent agent control method according to claim 1, characterized in that: The step of determining, according to the change of the environmental information, a description text for describing the change of the environmental information includes: Determining, according to the change of the environmental information, an initial description text for describing the change of the environmental information; The initial description text is optimized based on a text optimization model to obtain a description text for describing changes in the environmental information, wherein the text optimization model is used to output an optimized description text based on the input initial description text.

5. The intelligent agent control method according to claim 1, characterized in that: The step of controlling the target vehicle-mounted intelligent agent to output chat information according to the target prompt text includes: According to the target prompt text, determine a target document block from a preset knowledge base, wherein the preset knowledge base is used to store at least the document block of the target vehicle-mounted intelligent body, and the document block is obtained by splitting the user manual document of the target vehicle-mounted intelligent body; Determine a new prompt text based on the target document block and the target prompt text; The new prompt text is input into the target vehicle-mounted intelligent agent, and the target vehicle-mounted intelligent agent is controlled to output chat information.

6. The intelligent agent control method according to claim 5, characterized in that: The preset knowledge base also stores a document block vector corresponding to the document block, and determining the target document block from the preset knowledge base according to the target prompt text includes: Determining a text vector of the target prompt text; Determine, based on the text vector, from the preset knowledge base a target document block vector whose similarity to the text vector is greater than a preset similarity threshold; The document block corresponding to the target document block vector is determined as the target document block.

7. The intelligent agent control method according to any one of claims 1 to 3, characterized in that: The step of controlling the target vehicle-mounted intelligent agent to output chat information according to the target prompt text includes: Generate chat information according to the target prompt text; When the chat information passes the information verification, the target vehicle-mounted intelligent agent is controlled to output the chat information.

8. The intelligent agent control method according to claim 7, characterized in that: When the chat information passes the information verification, controlling the target vehicle-mounted intelligent agent to output the chat information includes: When the chat information does not include preset sensitive words and the chat information matches the user manual document of the target vehicle-mounted intelligent body, the target vehicle-mounted intelligent body is controlled to output the chat information.

9. The intelligent agent control method according to claim 7, characterized in that: When the chat information passes the information verification, controlling the target vehicle-mounted intelligent agent to output the chat information includes: Based on the chat information and the information verification model, determining whether the chat information passes the verification result of the information verification, wherein the information verification model is used to output the verification result of the chat information according to the input chat information; After the verification result indicates that the chat information has passed the information verification, the target vehicle-mounted intelligent agent is controlled to output the chat information.

10. The intelligent agent control method according to claim 7, characterized in that: Also includes: When the chat information fails to pass the information verification, the target vehicle-mounted intelligent agent is controlled to re-output the chat information.

11. The intelligent agent control method according to any one of claims 1 to 3, characterized in that: The step of controlling the target vehicle-mounted intelligent agent to output chat information according to the target prompt text includes: Generate chat information according to the target prompt text; Control the target vehicle-mounted intelligent agent to output the chat information in the form of audio and / or text.

12. The intelligent agent control method according to claim 11, characterized in that: The controlling the target vehicle-mounted intelligent agent to output the chat information in the form of audio includes: Determine the tone information of the chat information based at least on the chat information and a tone determination model, wherein the tone determination model is at least used to output the tone information of the chat information according to the input chat information; Based on the tone information, the target vehicle-mounted intelligent agent is controlled to output the chat information in the form of audio.

13. The intelligent agent control method according to any one of claims 1 to 3, characterized in that: The method further comprises: In the process of controlling the target vehicle-mounted intelligent agent to output chat information according to the target prompt text, if user feedback information is received, the target vehicle-mounted intelligent agent is controlled to output chat information in response to the feedback information, or, according to the feedback information, the target vehicle-mounted intelligent agent is controlled to perform a target action.

14. The intelligent agent control method according to claim 13, characterized in that: The controlling the target vehicle-mounted intelligent agent to output chat information in response to the feedback information comprises: If the target vehicle-mounted intelligent agent receives the feedback information after outputting the chat information, and the chat information includes a preset vehicle-mounted control instruction, when the feedback information indicates that the vehicle-mounted control instruction is not executed, the target vehicle-mounted intelligent agent is controlled to output the chat information for the feedback information; or, If the feedback information is received after the target vehicle-mounted intelligent agent outputs the chat information, and the chat information does not include a preset vehicle-mounted control instruction, the target vehicle-mounted intelligent agent is controlled to output chat information for the feedback information.

15. The intelligent agent control method according to claim 13, characterized in that: The step of controlling the target vehicle-mounted intelligent agent to perform a target action according to the feedback information includes: If the target vehicle-mounted intelligent agent receives the feedback information after outputting the chat information, and the chat information includes a preset vehicle-mounted control instruction, then when the feedback information represents the execution of the vehicle-mounted control instruction, the target vehicle-mounted intelligent agent is controlled to execute the action corresponding to the vehicle-mounted control instruction.

16. The intelligent agent control method according to any one of claims 1 to 3, characterized in that: The on-board intelligent body corresponds to the control device of the vehicle, and the on-board intelligent body includes a window on-board intelligent body, a seat on-board intelligent body and / or an air-conditioning on-board intelligent body. The window on-board intelligent body corresponds to the window control device of the vehicle, the seat on-board intelligent body corresponds to the seat control device of the vehicle, and the air-conditioning on-board intelligent body corresponds to the air-conditioning control device of the vehicle.

17. An electronic device, characterized in that: The electronic device comprises: processor; a memory for storing processor-executable instructions; Wherein, the processor is configured to: execute the steps of the method described in any one of claims 1-16.

18. A vehicle, characterized in that: The vehicle comprises a processor and a memory for storing processor executable instructions, wherein the processor is configured to: perform the steps of the method of any one of claims 1-16.

19. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that: When the program is executed by a processor, the steps of the method described in any one of claims 1 to 16 are implemented.

20. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 16 are implemented.

Citation Information

Patent Citations

  • Chatting robot-based interaction method, device and related equipment

    CN109407925A

  • Interactive control method and device, electronic equipment and storage medium

    CN110053629A

  • Agent device, agent control method, and program

    CN112805182A

  • Intelligent agent device, intelligent agent method, and recording medium

    CN114312797A

  • Multi-agent collaborative reinforcement learning method, terminal and storage medium

    CN116226662A