Data processing method and device based on scenic spot agent, electronic equipment and medium

Abstract

This application discloses a data processing method, apparatus, electronic device, and storage medium based on scenic area intelligent agents, comprising: receiving interaction information submitted by a user terminal after the intelligent agent interface configured in the scenic area service is triggered, wherein the scenic area service is entered after the user terminal completes geolocation matching and performs near-field perception interaction with the scenic area IoT sensing node; invoking the scenic area intelligent agent of the scenic area service based on the interaction information, so that the scenic area intelligent agent determines the target sub-intelligent agent corresponding to the interaction information among multiple configured sub-intelligent agents; if the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent, performing scenic area data collaborative processing through the real-scene question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interaction information to obtain a service result; generating service response data according to the service result and sending it to the user terminal.

Description

Technical Field

[0001] This application relates to the field of artificial intelligence technology, specifically to a data processing method, device, electronic device, and storage medium based on scenic area intelligent agents. Background Technology

[0002] With the digital upgrade of the cultural tourism industry, scenic spot services are gradually developing towards the integration of "online + offline". Users are constantly increasing their requirements for the real-world interactivity, geographical accuracy, real-time data and service collaboration of scenic spot services. Currently, most online services for scenic spots are built on a general cultural tourism service framework, which has several pain points: First, the access methods for scenic spot services are limited, mostly relying on near-field communication tags or keyword searches, without combining scenic spot geolocation and IoT sensing technologies, making it difficult to accurately match the user's real-world location with online services; Second, the integration of scenic spot intelligent services with real-world scenarios is low, with traditional question-and-answer intelligent agents only providing textual responses and failing to provide users with real-world interactive services by combining scenic spot real-world models and real-time data; Third, various functional services within scenic spots are fragmented, with sub-services such as ticket verification, scenic spot navigation, facility inquiry, and visitor flow warnings lacking effective intelligent collaboration mechanisms, requiring users to repeatedly operate through different service entry points, resulting in a poor interactive experience; Fourth, the response data for scenic spot services lacks real-world presentation, with service results mostly displayed in plain text or simple links, unable to be linked with real-world images, videos, and real-world interactive entry points, making it difficult to meet users' real-world consultation needs during offline visits; Fifth, real-time data of scenic spots (such as real-world visitor flow, facility status, and attraction opening status) is not deeply integrated with intelligent agent services, and the intelligent agent's processing results cannot reflect the real-time status of the scenic spot, resulting in insufficient service accuracy. Summary of the Invention

[0003] This application provides a data processing method, electronic device, apparatus, and storage medium based on scenic area intelligent agents, which can improve the intelligence, realism, and accuracy of scenic area services and meet users' diversified and realism-based interactive needs during their visit to the scenic area.

[0004] In a first aspect, embodiments of this application provide a data processing method based on a scenic area intelligent agent, including: The system receives interactive information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and performs near-field perception interaction with the scenic area's IoT sensing node. Based on the interaction information, the scenic area intelligent agent is invoked to provide the scenic area service, so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information among the multiple configured sub-intelligent agents; If the target sub-agent is a real-scene question-and-answer sub-agent, the service result is obtained by the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information through scenic area data collaborative processing. Based on the service result, service response data is generated and sent to the user terminal.

[0005] Optionally, in some embodiments of this application, the following approach is adopted: The access address of the scenic area service is obtained by parsing the scenic area identification data matched by the geolocation and the sensing access data transmitted by the scenic area IoT sensing node, and the scenic area service is accessed based on the access address.

[0006] Optionally, in some embodiments of this application, the step of obtaining service results through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and a sub-agent adapted to the interaction information includes: Based on the scenic area interaction data contained in the interaction information, candidate sub-intelligent agents that are adapted to the scenic area interaction data are selected, and the real-scene question-and-answer sub-intelligent agent and the scenic area data collaboration method of the candidate sub-intelligent agents are determined. The service results are obtained by performing collaborative data processing on scenic area data according to the described scenic area data collaboration method.

[0007] Optionally, in some embodiments of this application, obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaborative method includes: The real-scene question-answering sub-agent inputs the scenic area interaction data from the interaction information into the big language model and combines it with the scenic area real-scene model to process the data and generate service data, and extracts the scenic area keywords and real-scene feature keywords contained in the service data; Based on the scenic area keywords and real-scene feature keywords, the candidate sub-agent is invoked to process scenic area data and obtain real-scene information of the scenic area. The service data, the real-scene information of the scenic area, and the configuration information of the agent interface of the candidate sub-agent are used as the service result.

[0008] Optionally, in some embodiments of this application, obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaborative method includes: The candidate sub-agents process scenic area data based on interactive information to obtain the access address of the scenic area sub-service corresponding to the candidate sub-agents. The real-scene question-and-answer sub-agent generates service data based on the access address and the scenic area's real-scene data, and uses the access address and the service data as the service result.

[0009] Optionally, in some embodiments of this application, the step of obtaining service results through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and a sub-agent adapted to the interaction information includes: The scenic area intelligent agent performs intent recognition and keyword extraction on the scenic area interaction data in the interactive information to obtain multiple scenic area-related keywords corresponding to each intent category. Each scenic spot-related keyword is input into the candidate sub-agent corresponding to the intent category to process the scenic spot data, and the processing results and the service data generated by the real-scene question-and-answer sub-agent are obtained. The service result is obtained by aggregating the service data, processing results, and configuration information of the agent interface of the candidate sub-agents.

[0010] Optionally, in some embodiments of this application, after determining the target sub-agent operation corresponding to the interaction information among the configured plurality of sub-agents, the method further includes: If the target sub-agent is a candidate sub-agent, the service result is obtained by collaboratively processing scenic area data through the candidate sub-agent and the real-scene question-and-answer sub-agent.

[0011] Optionally, in some embodiments of this application, the step of obtaining service results through collaborative processing of scenic area data by the candidate sub-agent and the real-scene question-answering sub-agent includes: The scenic area agent calls the candidate sub-agent to process scenic area data, obtains the access address of the scenic area subroutine corresponding to the candidate sub-agent, and queries the scenic area candidate interactive text preset by the candidate sub-agent in the scenic area knowledge base. Generate the agent interface of the real-scene question-and-answer sub-agent corresponding to the candidate interactive text of the scenic area, input the agent information of the candidate sub-agent into the real-scene question-and-answer sub-agent and combine it with the real-time data of the scenic area to generate service text, and generate service results according to the configuration information of the agent interface corresponding to the candidate interactive text of the scenic area and the service text.

[0012] Optionally, in some embodiments of this application, before the step of receiving the interactive information submitted after the intelligent agent interface configured by the user terminal on the scenic area service page is triggered is executed, the method further includes: The encrypted sensing access data submitted by the user terminal calling the data transmission interface of the scenic area service is obtained; the data transmission interface obtains the encrypted sensing access data transmitted by the scenic area IoT sensing node by parsing the data and combining it with the geolocation verification result. The encrypted access data is decrypted to obtain the access instruction for the scenic area service page, and the scenic area service page is sent to the user terminal in response to the access instruction.

[0013] Optionally, in some embodiments of this application, generating service response data based on the service result includes: The scenic area data included in the service results are classified and processed according to the data type to obtain the scenic area service data corresponding to each data type. The categorized scenic area service data is written into the scenic area data template to obtain the service response data.

[0014] Optionally, in some embodiments of this application, after determining the target sub-agent operation corresponding to the interaction information among the configured plurality of sub-agents and before performing the operation of obtaining service results through scenic area data collaborative processing by the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information, the method further includes: Search the scenic area knowledge base to see if there is a preset scenic area response text that matches the scenic area interaction data in the interactive information; If it exists, obtain the preset scenic spot response text and write the preset scenic spot response text into the service result.

[0015] Optionally, in some embodiments of this application, obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaborative method includes: The scenic area intelligent agent extracts keywords from the interactive information to obtain scenic area ticketing keywords, and inputs the scenic area ticketing keywords into the ticketing verification sub-intelligent agent to query and verify scenic area ticketing information to obtain ticketing verification information; the ticketing verification information is obtained by calling the data query and verification interface of the ticketing verification subroutine using the scenic area ticketing keywords as input data. The ticket verification information and the interaction information are input into the real-world question-and-answer sub-agent to generate a service text, and the ticket verification information, the service text, and the access address of the ticket verification sub-program are used as the service result.

[0016] Optionally, in some embodiments of this application, generating service response data based on the service result includes: Extract the configuration information of the agent interface of the real-world question-and-answer sub-agent corresponding to the response text and the candidate interactive text of the scenic area contained in the service result, and generate a service response message based on the response text and the preset scenic area service elements; Based on the configuration information, the intelligent agent interface corresponding to the candidate interactive text of the scenic area is configured in the service response message, and the configured service response message is used as the service response data.

[0017] Optionally, in some embodiments of this application, generating service response data based on the service result includes: Extract the access address of the scenic area subroutine corresponding to the candidate sub-agent and the response text generated by the real-world question-and-answer sub-agent from the service results; Rich text is generated based on the access address and the response text, combined with real-life images / videos of the scenic area, and used as the service response data.

[0018] Secondly, embodiments of this application provide a data processing device based on a scenic area intelligent agent, comprising: The receiving module is used to receive the interactive information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is entered after the user terminal completes the geolocation matching and performs near-field perception interaction with the scenic area IoT sensing node. The invocation module is used to invoke the scenic area intelligent agent for the scenic area service based on the interaction information, so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information among the multiple configured sub-intelligent agents; The processing module is used to obtain service results by performing scenic area data collaborative processing through the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information if the target sub-agent is a real-scene question-and-answer sub-agent. The delivery module is used to generate service response data based on the service result and deliver it to the user terminal.

[0019] Accordingly, this application also provides an electronic device, including a memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as described in any of the methods above.

[0020] This application also provides a storage medium storing a processor program that, when executed by a processor, implements any of the methods described above.

[0021] This application provides a data processing method, apparatus, electronic device, and storage medium based on scenic area intelligent agents. It receives interaction information submitted by a user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and near-field perception interaction with the scenic area's IoT sensing nodes. Then, based on the interaction information, the scenic area intelligent agent of the service is invoked, so that the scenic area intelligent agent determines the target sub-intelligent agent corresponding to the interaction information from among multiple configured sub-intelligent agents. If the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent, the real-scene question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interaction information perform scenic area data collaborative processing to obtain service results. Finally, service response data is generated based on the service results and sent to the user terminal. In the data processing scheme based on scenic area intelligent agents provided in this application, by integrating technologies such as geolocation, IoT sensing, and real-scene interaction, a scenic area-specific intelligent agent system is constructed, realizing collaborative processing of various sub-intelligent agents in the scenic area. This allows for deep integration of online services and offline real-scene experiences, improving the intelligence, real-scene nature, and accuracy of scenic area services, and meeting the diversified and real-scene interaction needs of users during their scenic area visits. Attached Figure Description

[0022] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 This is a schematic diagram of the implementation environment of the data processing method based on scenic area intelligent agents provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the data processing method based on scenic area intelligent agents provided in an embodiment of this application; Figure 3 This is a schematic diagram of the data processing device based on scenic area intelligent agents provided in the embodiments of this application; Figure 4 This is a schematic diagram of the structure of the electronic device provided in the embodiments of this application. Detailed Implementation

[0024] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] In the description of this application, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," and "outer," etc., indicating orientation or positional relationships based on the orientation or positional relationships shown in the accompanying drawings, are only for the convenience of describing this application and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of this application. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of indicated technical features. Thus, a feature defined with "first" or "second" may explicitly or implicitly include one or more of the stated features. In the description of this application, "a plurality of" means two or more, unless otherwise explicitly specified.

[0026] In this application, the term "exemplary" is used to mean "serving as an example, illustration, or description." Any embodiment described as "exemplary" in this application is not necessarily to be construed as being more preferred or advantageous than other embodiments. The following description is provided to enable any person skilled in the art to make and use this application. Details are set forth in the following description for purposes of explanation. It should be understood that those skilled in the art will recognize that this application can be made without using these specific details. In other instances, well-known structures and processes are not described in detail to avoid obscuring the description of this application with unnecessary detail. Therefore, this application is not intended to be limited to the embodiments shown, but is consistent with the broadest scope of the principles and features disclosed in this application.

[0027] This application provides a data processing method, apparatus, server, and computer-readable storage medium based on scenic area intelligent agents, which will be described in detail below.

[0028] The data processing method based on the scenic area intelligent agent in this application embodiment can be applied to the service operation environment of scenic area "online + offline" integration. Referring to Figure 1, the implementation environment includes at least: user terminal 101, scenic area service server 102, scenic area intelligent agent 103, and scenic area Internet of Things sensing node 104. The core feature of this implementation environment is the deep integration of offline real-scene data and online intelligent services. All data interactions between devices revolve around the real-scene tour scenario of the scenic area, ensuring the accuracy and realism of the service.

[0029] The user terminal 101 is used to complete the geolocation matching of the scenic area, conduct near-field perception interaction with the IoT sensing node 104 of the scenic area, and then access the scenic area services. It sends interactive information with real-scene features to the server 102 and receives real-scene service response data sent by the server 102. It also displays and links the real-scene data of the scenic area on the scenic area service page. The user terminal 101 can install the client of the scenic area service. The client can be an independent application, a subroutine of a third-party application, or a service module. Specifically, the user terminal 101 can be a smartphone, tablet, scenic area-specific guide terminal, AR / VR-based real-scene interaction device, wearable smart device, etc., and must have high-precision positioning components (GPS / BeiDou dual-mode) and near-field perception components (NFC / Bluetooth / UWB) to support meter-level positioning and near-field data interaction within the scenic area.

[0030] The scenic area service server 102 is a dedicated service processing server for the scenic area. It is used to receive interactive information sent by user terminal 101, call the scenic area intelligent agent 103 based on the geographic location and real-scene feature data in the interactive information, generate service response data according to the real-scene service results generated by the scenic area intelligent agent 103, and send it to user terminal 101. The server 102 can be a standalone server, a server cluster composed of several servers, or one or more cloud servers in a cloud computing platform. It also establishes a real-time data communication link with the scenic area IoT sensing node 104, and can synchronously obtain real-time operation data and real-scene data of the scenic area.

[0031] The scenic area intelligent agent 103 is a dedicated intelligent agent system for serving the scenic area. It is used to determine the target sub-intelligent agent from multiple configured sub-intelligent agents based on interactive information with real-world features. If it is determined to be a real-world question-and-answer sub-intelligent agent, the real-world question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interactive information will perform collaborative processing of scenic area data to obtain service results. The scenic area intelligent agent 103 can be deployed independently on a dedicated server, and the server 102 can interact with the scenic area intelligent agent 103 through intelligent agent invocation. Alternatively, it can be integrated and deployed within the scenic area service server 102. The scenic area intelligent agent 103 establishes data connections with the scenic area real-world model library, the scenic area knowledge base, and the scenic area real-time data platform, and can obtain real-time data such as the scenic area's 3D real-world model, preset real-world response text, and the real-time status of the scenic area's visitor flow, facilities, and attractions.

[0032] The scenic area IoT sensing node 104 is a distributed sensing device deployed within the scenic area. It is mainly deployed in key locations such as various attractions, entrances and exits, amusement facilities, and service facilities. It is used to collect real-time real-scene data of the scenic area (such as visitor flow, facility operation status, attraction opening status, and environmental data) and to conduct near-field sensing interaction with user terminal 101 to transmit encrypted sensing access data. The scenic area IoT sensing node 104 includes IoT sensors, near-field communication base stations, real-scene acquisition cameras, visitor flow statistics equipment, etc., and all devices are synchronized with the scenic area service server 102 at the millisecond level to ensure the real-time nature of the collected data.

[0033] The core working logic of this implementation environment is as follows: After user terminal 101 completes the scenic area's geolocation matching and near-field perception interaction with the scenic area's IoT sensing node 104, it enters the scenic area service page that accurately matches the user's offline real-world location. During the visit, the user triggers the intelligent agent interface configured in the scenic area service, submits interactive information with geolocation and real-world features, and sends it to server 102. Server 102 calls the scenic area intelligent agent 103 based on the interactive information. The scenic area intelligent agent 103 responds to the call by determining the target sub-intelligent agent. If it is a real-world question-and-answer sub-intelligent agent, it performs collaborative processing of scenic area data with the adapted sub-intelligent agent, integrating the real-world model and real-time data to generate service results. Server 102 generates real-world service response data based on the service results and sends it to user terminal 101. User terminal 101 displays the service response data on the scenic area service page and links with the scenic area's real-world data to achieve AR / VR real-world display. In this way, during the user's visit to the scenic area, the collaborative processing of the scenic area intelligent agent achieves deep integration of online services and offline real-world, providing users with real-world, accurate, and intelligent scenic area services.

[0034] The data processing method (server-side) based on scenic area intelligent agents provided in this application embodiment is as follows: Referring to Figure 2, the data processing method based on scenic area intelligent agents provided in this embodiment includes steps S201 to S204. This method is the core processing flow on the server side. All steps revolve around the real-world data of the scenic area, which is different from the undifferentiated processing of general cultural and tourism services, and realizes the customization and precision of scenic area services.

[0035] Step S201: Receive the interaction information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered.

[0036] The scenic area service described in this embodiment is an online + offline integrated service specifically designed for offline real-world tours of scenic areas. It is provided by the scenic area service server 102 and can be deployed through applications and / or application subroutines. It provides users with diverse functions such as real-world Q&A, ticket verification, real-world navigation, facility inquiry, visitor flow warning, and real-world attraction introductions. All functions are highly bound to the user's offline real-world location, realizing the core needs of "real-world questioning, real-world answering, and real-world interaction".

[0037] Specifically, the scenic area service can be configured with a scenic area intelligent agent 103. All functions of the scenic area service are implemented through the collaboration of the scenic area intelligent agent 103 and its sub-intelligent agents. The scenic area service generates and configures corresponding trigger controls and / or intelligent agent interfaces for each real-world function. This intelligent agent interface is a dedicated interface for the scenic area service, corresponding one-to-one with each sub-intelligent agent of the scenic area. It can be configured on various page formats such as the regular page, AR real-world page, and VR panoramic page of the scenic area service. During the offline tour of the scenic area, users can trigger the intelligent agent interface through touch, voice, real-world shooting, etc., to submit interactive information related to the offline real-world scene, realizing immersive access to the scenic area service.

[0038] Among them, the scenic area intelligent agent 103 is equipped with multiple scenic area-specific sub-intelligent agents. Each sub-intelligent agent corresponds to the real-scene function of the scenic area service. It can independently realize a single function or cooperate with other sub-intelligent agents to handle complex scenic area service requests. The processing data of each sub-intelligent agent is combined with the scenic area's geographic location data and real-scene feature data to ensure the real-scene nature of the processing results. Optionally, the sub-intelligent agents include: real-scene question and answer sub-intelligent agents, ticket verification sub-intelligent agents, real-scene tour sub-intelligent agents, facility query sub-intelligent agents, visitor flow real-scene sub-intelligent agents, and attraction introduction sub-intelligent agents, etc. Each sub-intelligent agent has a built-in scenic area real-scene data parsing module, which supports real-time data interaction with the scenic area real-scene model library and real-time data platform.

[0039] The core functions of each sub-agent are as follows: The Real-Scene Question Answering sub-agent serves as the core collaborative hub for the scenic area's agents, combining the scenic area's real-scene model and large language model to provide users with realistic text / voice responses and to collaborate with other sub-agents on data. The Ticketing Verification sub-agent is used to query, verify, and redeem scenic area tickets, amusement park tickets, and guided tour tickets, and provides a redemption entry point matched with the user's real-scene location. The Real-Scene Guide sub-agent generates AR / VR real-scene navigation routes for the scenic area based on the user's geolocation, providing real-scene guided tour services. The Visitor Flow Real-Scene sub-agent queries real-time visitor flow data for each scenic spot and provides real-scene images / videos of visitor flow. The Facility Query sub-agent queries the real-scene locations and operational status of service facilities such as restrooms, drinking fountains, sightseeing vehicles, and parking lots within the scenic area. The Scenic Spot Introduction sub-agent combines the user's real-scene location to provide realistic introductions to scenic spots, historical and cultural explanations, and other services.

[0040] In practice, to achieve accurate matching between scenic area services and users' offline real-world locations, scenic area services adopt a dual verification method of geolocation matching and near-field sensing interaction with scenic area IoT sensing nodes. This method differs from the traditional single near-field communication access method of cultural tourism services. Through dual verification, it ensures that the user terminal's access request is highly consistent with the scenic area's real-world location, laying a data foundation for subsequent real-world service processing. Optionally, scenic area services can be accessed after the user terminal completes geolocation matching and near-field sensing interaction with the scenic area's IoT sensing nodes.

[0041] Specifically, the user terminal first completes the scenic area geolocation matching through a high-precision positioning component, that is, parsing the positioning data to determine the user's specific location within the scenic area (such as the core area of ​​XX scenic spot, the entrance of XX scenic spot, or the scenic area visitor center), generating scenic area identification data containing location codes and regional characteristics; then, through a near-field sensing component, it performs near-field sensing interaction with the scenic area IoT sensing nodes deployed at that location to obtain sensing access data containing scenic area service access information transmitted by the sensing nodes; in an optional implementation provided in this embodiment, the scenic area service is accessed in the following way: parsing the scenic area identification data matched by geolocation and the sensing access data transmitted by the scenic area IoT sensing nodes to obtain the access address of the scenic area service, and accessing the scenic area service based on the access address. This access address is the exclusive sub-page address of the scenic area service corresponding to the user's current real-world location**, rather than the general address of the scenic area service, to ensure that the service page entered by the user is accurately matched with the offline real-world location. For example, if the user is at the waterfall scenic spot, they will enter the exclusive real-world service page of the waterfall scenic spot, which only displays service functions and data related to that scenic spot.

[0042] In practical applications, to enhance the access security of scenic area services and prevent unauthorized access by users outside the scenic area, the sensing access data transmitted by the scenic area's IoT sensing nodes is encrypted. This involves encrypting the access command for the scenic area services by combining the user's geographic location data and the sensing node's location code, and then writing it to the sensing node. The user terminal parses the encrypted sensing access data obtained through near-field sensing interaction, verifies the data by combining it with its own geographic location verification results, and obtains the data transmission interface for the scenic area services after successful verification. Correspondingly, before receiving the interaction information sent by the user terminal, the server must first receive the encrypted sensing access data submitted by the user terminal, decrypt and verify it, obtain the access command for the scenic area service page, and then send a dedicated service page matching the user's actual location to the user terminal. In one optional implementation of this embodiment, before receiving the interaction information, the process further includes: obtaining the encrypted sensing access data submitted by the user terminal calling the data transmission interface for the scenic area services; the data transmission interface is obtained by parsing the encrypted sensing access data transmitted by the scenic area's IoT sensing nodes and combining it with the geographic location verification results; decrypting the encrypted sensing access data to obtain the access command for the scenic area service page; and sending the scenic area service page to the user terminal in response to the access command.

[0043] In the specific execution process, after entering the dedicated sub-page of the scenic area service, the user can trigger the intelligent agent interface at any real-world location within the scenic area to submit interactive information. This interactive information is a service request for the scenic area based on the user's offline real-world location. Unlike the general interactive information of traditional cultural and tourism services, the interactive information in this embodiment has strong real-world characteristics and includes the user's current geographic location data and real-world feature data, ensuring that the server and the scenic area intelligent agent can accurately identify the user's real-world needs. Among them, the interactive information refers to the information generated after the user triggers the intelligent agent interface configured in the scenic area service to describe the user's real-world interactive operation. The interactive information may include the configuration information of the intelligent agent interface triggered by the user, the scenic area interactive data entered by the user, the geographic location data of the user's terminal, and the real-world feature data. Optionally, the interactive information includes the configuration information of the intelligent agent interface, the scenic area interactive data, the geographic location data, and the real-world feature data. Among them, the scenic area interactive data includes interactive text, interactive voice, and real-world interactive media data (such as real-world pictures / videos taken by the user), and the real-world feature data includes real-world elements taken by the user, scenic area features of the area, and surrounding facility information.

[0044] As mentioned above, the intelligent agent interface refers to a dedicated interface configured on the scenic area service and / or the scenic area service real-scene page for calling the scenic area intelligent agent. The configuration information of this interface includes sub-intelligent agent identifiers, service function types, and real-scene adaptation ranges. After the intelligent agent interface is triggered, the corresponding scenic area sub-intelligent agent can be called based on interactive information with real-scene features. The interactive information can be submitted to the server through the user terminal after the user triggers the scenic area service's intelligent agent interface. Specifically, the interactive information can be sent after triggering the intelligent agent interface of the real-scene question-and-answer sub-intelligent agent, including the interface configuration information of the real-scene question-and-answer sub-intelligent agent, the real-scene question-and-answer data entered by the user, and the user's location and real-scene feature data; alternatively, the interactive information can also be sent after triggering the intelligent agent interfaces of other candidate sub-intelligent agents, including the interface configuration information of the candidate sub-intelligent agents, the user's corresponding scenic area function request data, and the user's location and real-scene feature data.

[0045] In practical applications, to improve the response efficiency of scenic area services and reduce the resource consumption of intelligent agent collaborative processing, after receiving the interaction information, the server can first perform a matching query for preset response texts in the scenic area knowledge base. This scenic area knowledge base is a knowledge base exclusive to the scenic area, containing a large number of preset real-world response texts related to various attractions, facilities, and services within the scenic area. All texts are written in conjunction with the real-world characteristics of the scenic area and are adapted to common real-world questions asked by tourists. If a preset scenic area response text matching the interaction data is found, the text can be directly written into the service result without complex intelligent agent collaborative processing. In an optional implementation of this embodiment, after determining the target sub-intelligent agent operation corresponding to the interaction information among the configured multiple sub-intelligent agents and before the operation of obtaining the service result through scenic area data collaborative processing by the real-world question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interaction information, the method further includes: querying the scenic area knowledge base to see if there is a preset scenic area response text matching the scenic area interaction data in the interaction information; if it exists, obtaining the preset scenic area response text and writing it into the service result; if it does not exist, then executing the subsequent scenic area intelligent agent collaborative processing steps.

[0046] Step S202: Invoke the scenic area intelligent agent for the scenic area service based on the interaction information.

[0047] In specific implementation, a scenic area intelligent agent is specifically invoked based on interactive information with real-scene features sent by the user terminal. This differs from the general intelligent agent invocation in traditional cultural tourism services. In this embodiment, the invocation process uses geographic location data and real-scene feature data from the interactive information as core invocation parameters, rather than solely relying on the user's interactive text. This ensures that the processing results of the scenic area intelligent agent accurately match the user's offline real-world experience. Correspondingly, in response to the invocation, the scenic area intelligent agent determines the target sub-intelligent agent corresponding to the interactive information from among multiple configured sub-intelligent agents. The target sub-intelligent agent then collaborates with the sub-intelligent agent adapted to the interactive information to perform scenic area data collaborative processing to obtain the service result. Optionally, the scenic area intelligent agent determines the target sub-intelligent agent corresponding to the interactive information from among multiple configured sub-intelligent agents.

[0048] S203. If the target sub-agent is a real-scene question-and-answer sub-agent, the service result is obtained by the scenic area data collaborative processing through the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information.

[0049] Specifically, during the process of calling the scenic area intelligent agent, the full amount of interactive information (including intelligent agent interface configuration information, scenic area interactive data, geolocation data, and real-world feature data) can be used as the core input for calling the scenic area intelligent agent. The application programming interface (API) of the scenic area intelligent agent is called to send the full amount of interactive data to the scenic area intelligent agent, ensuring that the scenic area intelligent agent can perform sub-intelligent agent matching and collaborative processing based on multi-dimensional real-world data, and avoiding the disconnect between the processing results and the real-world data due to data loss.

[0050] After the scenic area intelligent agent is invoked based on the interactive information, it responds to the invocation and processes the data. The core of this service collaboration involves sub-intelligent agents adapted to the interactive information to collaboratively process scenic area data and obtain service results. The core of this service collaboration is the real-scene question-and-answer sub-intelligent agent, which acts as the central collaborative hub, fusing scenic area data with various candidate sub-intelligent agents. All collaborative processing combines the scenic area's real-scene model, real-time real-scene data, and user geolocation data to ensure the realism and accuracy of the service results. Specifically, the scenic area intelligent agent determines the target sub-intelligent agent from among multiple sub-intelligent agents based on the real-scene characteristics of the interactive information and user intent data. The target sub-intelligent agent and the sub-intelligent agents adapted to the interactive information then collaboratively process scenic area data to obtain service results. During the service collaboration process, the target sub-intelligent agent can act as the master agent for collaborative processing, or the scenic area intelligent agent can act as the main coordinator to schedule collaborative processing among the sub-intelligent agents. Alternatively, the target sub-intelligent agent and candidate sub-intelligent agents can invoke each other at the same level. The specific collaboration method can be flexibly switched according to the real-time load of the scenic area service.

[0051] The service result refers to the real-scene service collaboration result obtained by each sub-agent after collaborative processing of scenic area data. Unlike the general service results of traditional cultural tourism services, the service result in this embodiment has strong real-scene characteristics of the scenic area and contains multi-dimensional scenic area data, which can be directly used to generate real-scene service response data. The service result may include the real-scene processing results of each sub-agent, the real-scene service data generated by the real-scene question-and-answer sub-agent, and the aggregated data of the processing results of each sub-agent. Optionally, the service result may include: service text / real-scene reply data generated by the real-scene question-and-answer sub-agent, real-scene professional data of candidate sub-agents (ticket verification information / passenger flow real-scene data / facility real-scene information), access address of scenic area sub-program / sub-service, configuration information of scenic area agent interface, association address of scenic area real-scene model / image / video, real-time operation data of scenic area, etc.

[0052] Specifically, in the process of determining the target sub-agent, the scenic area agent determines the target sub-agent from among multiple sub-agents based on the real-scene feature data, user intent data, and agent interface configuration information contained in the interaction information, ensuring that the target sub-agent highly matches the user's real-scene interaction needs. In one optional implementation of this embodiment, determining the target sub-agent corresponding to the interaction information from among the configured multiple sub-agents includes: detecting whether the interaction information contains real-scene interaction data entered by the user (such as real-scene Q&A, real-scene shooting query, real-scene guide request); if so, determining the real-scene Q&A sub-agent as the target sub-agent from among the sub-agents, prioritizing the use of the real-scene Q&A sub-agent to achieve multi-dimensional real-scene responses; if not, directly using the agent interface configuration information contained in the interaction information to determine the sub-agent corresponding to the configuration information as the target sub-agent. This determination method can prioritize matching the real-scene Q&A sub-agent, aligning with the core real-scene interaction needs of scenic area users, and achieving the service effect of "one question, multi-dimensional response".

[0053] Specifically, in the collaborative processing of scenic area data, based on the scenic area-specific sub-intelligent agents provided above, the collaborative processing methods between the target sub-intelligent agent and different sub-intelligent agents revolve around real-world rendering. The core is divided into two cases: the collaborative processing of the real-world question-and-answer sub-intelligent agent as the target sub-intelligent agent and the candidate sub-intelligent agent, and the collaborative processing of the candidate sub-intelligent agent as the target sub-intelligent agent and the real-world question-and-answer sub-intelligent agent. These will be explained in detail below. All collaborative processing processes must combine the data from the scenic area real-world model library and the scenic area real-time data platform to ensure the real-world rendering and real-time nature of the processing results.

[0054] (1) Collaborative processing of scenic area data between real-scene question-answering sub-agents and sub-agents corresponding to interactive information. If the target sub-agent is a real-scene question-and-answer sub-agent, the service result is obtained through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information. This process is the core scenario of the scenic area agent's service processing and is adapted to the real-scene question-and-answer requests made by users, such as "How do I verify the ticket for this attraction?", "Are there many tourists nearby?", "Where is the restroom?", etc. Specifically, the scenic area agent determines the target sub-agent corresponding to the interaction information as the real-scene question-and-answer sub-agent from among the configured sub-agents. Through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and the candidate sub-agent adapted to the interaction information, the real-scene data of both parties is integrated to generate the service result, realizing the core requirement of "natural language question + real-scene professional answer".

[0055] Specifically, in the process of collaborative processing of scenic area data, the scenic area intelligent agent first selects suitable candidate sub-intelligent agents based on the scenic area interaction data, real-scene feature data, and user geolocation data contained in the interaction information. Then, it determines the scenic area data collaboration method between the real-scene question-and-answer sub-intelligent agent and the candidate sub-intelligent agents. Finally, it processes the data according to the collaboration method to obtain the service result. In one optional implementation of this embodiment, obtaining the service result through collaborative processing of scenic area data by the real-scene question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interaction information includes: selecting candidate sub-intelligent agents adapted to the scenic area interaction data based on the scenic area interaction data contained in the interaction information, and determining the scenic area data collaboration method of the real-scene question-and-answer sub-intelligent agent and the candidate sub-intelligent agents; and obtaining the service result by performing collaborative processing of scenic area data according to the collaborative processing method.

[0056] Optionally, the scenic area data collaboration methods include: using the real-scene question-and-answer sub-agent as the main agent to call candidate sub-agents for collaborative processing; having the real-scene question-and-answer sub-agent and candidate sub-agents call each other at the same level for collaborative processing; and using the scenic area agent as the main coordinator to call the real-scene question-and-answer sub-agent and candidate agent for collaborative processing. All three collaboration methods are based on "real-scene data fusion," differing only in the data scheduling entity. They can be flexibly switched according to the real-time load of the scenic area service and the complexity of user requests: simple requests use the collaboration method with the real-scene question-and-answer sub-agent as the main agent, resulting in higher processing efficiency; complex, multi-dimensional requests use the coordination method with the scenic area agent as the main agent, resulting in more comprehensive processing results.

[0057] Specifically, in the process of selecting candidate sub-agents adapted to the interactive information, the scenic area agent uses a large language model combined with a scenic area real-scene model library to perform intent recognition on the scenic area interactive data and real-scene feature data in the interactive information, obtains the user's real-scene service intent category, and then determines the corresponding candidate sub-agents based on the intent category. In an optional implementation provided in this embodiment, the sub-agents adapted to the interactive information are determined in the following way: intent recognition is performed on the scenic area interactive data in the interactive information to obtain the intent category, and the sub-agents corresponding to the intent categories are determined as candidate sub-agents. Among them, the intent categories are all categories related to scenic area real-scene services, including scenic area ticket verification, real-scene guidance, facility real-scene query, visitor flow real-scene query, attraction real-scene introduction, amusement project reservation, etc., which correspond one-to-one with the functions of each sub-agent of the scenic area.

[0058] Specifically, in the intent recognition process, the large language model combines with the scenic area real-scene model library to perform real-scene intent recognition. Unlike general text-based intent recognition, this process first jointly analyzes interaction data, real-scene feature data, and user geolocation data to obtain the user's real-scene service intent. Then, it matches the user intent with the preset real-scene intent categories of each sub-agent, selecting the preset intent category corresponding to the matching result as the intent recognition result. For example, if the user interaction data is "How do I verify the ticket for this attraction?" and the real-scene feature data is "XX attraction entrance" and the geolocation data is "scenic area south gate," the intent recognition result is "scenic area ticket verification," and the corresponding candidate sub-agent is the ticket verification sub-agent. If the user interaction data is "Are there many visitors nearby?" and the real-scene feature data is "XX waterfall attraction" and the geolocation data is "waterfall viewing platform," the intent recognition result is "visitor flow real-scene query," and the corresponding candidate sub-agent is the visitor flow real-scene sub-agent. If the user interaction data is "Where is the restroom?" and the geolocation data is "scenic area visitor center," the intent recognition result is "facility real-scene query," and the corresponding candidate sub-agent is the facility query sub-agent.

[0059] Based on the three scenic area data collaboration methods provided above, in the specific execution process, the real-scene question-and-answer sub-intelligent agent and the candidate sub-intelligent agent are processed in accordance with the collaboration method to achieve real-scene data collaboration. The following details the processing procedures corresponding to the three collaboration methods. All processes combine scenic area real-scene data and user geolocation data to ensure that the processing results are highly consistent with the user's offline real-scene data.

[0060] (a) Using the real-world question-answering sub-agent as the master agent to call candidate sub-agents for collaborative processing This method is the mainstream collaborative approach for scenic area services, suitable for single real-scene question-and-answer requests from users. It boasts high processing efficiency and fast response speed. The real-scene question-and-answer sub-agent acts as the main agent, leading the entire data collaborative processing process. In specific execution, the real-scene question-and-answer sub-agent first generates real-scene service data based on user interaction data, then extracts scenic area real-scene keywords from the service data, calls candidate sub-agents to obtain professional real-scene scenic area data, and finally merges the data from both sides to generate the service result. In the first optional implementation method provided in this embodiment, the service collaborative processing to obtain the service result according to the service collaboration method includes: the real-scene question-and-answer sub-agent inputs the scenic area interaction data from the interaction information into a large language model and combines it with the scenic area real-scene model to process the data and generate service data, and extracts the scenic area area keywords and real-scene feature keywords contained in the service data; based on the scenic area area keywords and real-scene feature keywords, it calls the candidate sub-agents to process the scenic area data to obtain scenic area real-scene information, and uses the service data, the scenic area real-scene information, and the configuration information of the agent interface of the candidate sub-agents as the service result.

[0061] The scenic area's real-world model is a pre-constructed 3D model containing precise 3D data of various attractions, facilities, and roads. The service data generated by the large language model, combined with this model, features strong real-world characteristics rather than generic text. For example, it might say, "You are currently at the entrance of XX Waterfall. Ticket verification for this attraction can be completed through offline gates or online AR verification portals," rather than simply "Tickets can be verified online or offline." The scenic area keywords are the specific areas the user is currently located in (e.g., XX attraction, XX entrance / exit, XX viewing platform), and the real-world feature keywords are the real-world elements corresponding to the user's interaction (e.g., tickets, visitor flow, restrooms, sightseeing vehicles). Based on these keywords, candidate sub-agents combine real-time data from the scenic area's IoT sensing nodes and the scenic area's real-world model library to generate real-world information. For example, the ticket verification sub-agent provides the online verification portal for XX attraction tickets, the real-world location of the offline verification gates, and navigation routes; the visitor flow real-world sub-agent provides information on XX... The system provides real-time visitor flow data for attractions, real-scene photos of visitor flow, and recommendations for nearby attractions with lower visitor flow. The facility query sub-agent provides the real-scene location, distance, and navigation route for restrooms.

[0062] (b) Enable real-world question-answering sub-agents and candidate sub-agents to call each other and collaborate at the same level. This method is suitable for scenarios where user interaction data directly corresponds to the function of candidate sub-agents. For example, if a user directly asks, "How do I redeem tickets for XX attraction?", the candidate sub-agent first generates a real-world sub-service access address based on the interaction information. The real-world question-and-answer sub-agent then generates service data based on the access address and real-world data of the scenic area. Finally, the access address and service data are merged to generate a service result. In the second optional implementation provided in this embodiment, service collaboration processing is performed to obtain the service result, including: the candidate sub-agent processes scenic area data based on the interaction information to obtain the access address of the scenic area sub-service corresponding to the candidate sub-agent; the real-world question-and-answer sub-agent generates service data based on the access address and real-world data of the scenic area, and uses the access address and the service data as the service result.

[0063] Specifically, the scenic area intelligent agent acts as a data relay, inputting interactive information into candidate sub-intelligent agents. These candidate sub-intelligent agents, based on the user's geolocation data and real-world feature data, generate a unique access address for the scenic area's sub-services that matches the user's real-world location (e.g., the AR real-world verification page address for XX attraction ticket redemption, or the VR viewing address for XX attraction visitor flow), rather than a generic sub-service address. This access address is then input into the real-world question-and-answer sub-intelligent agent, which combines the scenic area's real-world data to generate a real-world service text containing the access address (e.g., "You can click the link below to enter the AR real-world verification page for XX attraction tickets; simply scan the ticket QR code to complete the verification"). Finally, the access address and service data are used as the service result, achieving integrated processing of "service address + real-world guidance."

[0064] (c) Using the scenic area agent as the main coordinator, call upon the real-world question-answering sub-agent and candidate sub-agents for collaborative processing. This approach is suitable for scenarios where user interaction information contains multi-dimensional real-world needs, such as a user asking, "Where can I verify my tickets nearby? Is it crowded? Are there any restrooms?" This request simultaneously includes needs for ticket verification, visitor flow inquiry, and facility inquiry, requiring collaborative processing from multiple sub-agents. The scenic area agent acts as the main coordinator, uniformly scheduling each sub-agent for data processing, and finally aggregating all processing results to generate a service result. In the third optional implementation provided in this embodiment, service results are obtained through service collaboration processing, including: the scenic area agent performing intent recognition and keyword extraction on the scenic area interaction data in the interaction information to obtain multiple scenic area-related keywords corresponding to each intent category; inputting each scenic area-related keyword into the candidate sub-agent corresponding to the intent category for scenic area data processing to obtain processing results and service data generated by the real-world question-and-answer sub-agent; and aggregating the service data, processing results, and configuration information of the agent interfaces of the candidate sub-agents to obtain the service result.

[0065] In the specific execution process, the scenic area intelligent agent first uses a large language model to perform multi-intent recognition on the interaction data, breaking it down into multiple real-scene service intent categories. Then, it extracts corresponding scenic area-related keywords for each intent category and sends the keywords to the corresponding candidate sub-intelligent agents. Each candidate sub-intelligent agent independently processes the scenic area data and generates corresponding real-scene professional data. At the same time, the real-scene question-and-answer sub-intelligent agent generates real-scene service text that integrates all needs based on the user's full interaction data. Finally, the scenic area intelligent agent aggregates the service data of the real-scene question-and-answer sub-intelligent agent, the processing results of each candidate sub-intelligent agent, and the intelligent agent interface configuration information of each candidate intelligent agent to generate a service result containing multi-dimensional information, realizing "one question, multi-dimensional real-scene response," which greatly improves the user's interactive experience.

[0066] In practical applications, the collaborative processing of the real-scene question-answering sub-agent and each candidate sub-agent requires real-time data interaction with the scenic area's IoT sensing nodes and real-time data platform to obtain real-time real-scene data of the scenic area. The following details the specific collaborative processing process of the real-scene question-answering sub-agent and candidate sub-agent for the two core application scenarios of scenic area services. Other scenarios can refer to this logic for execution.

[0067] (i) Real-world question-and-answer sub-agent and ticket verification sub-agent collaborate on scenic area data processing. This scenario represents the core service scenario at the entrance of the scenic area and the entrances of various attractions. The core user need is real-time Q&A related to ticket verification. The collaborative processing combines the scenic area ticketing system and real-time location data to achieve a one-stop service of "Q&A + verification". In one optional implementation of this embodiment, the service collaborative processing to obtain the service result includes: the scenic area agent extracting keywords from the interactive information to obtain scenic area ticketing keywords; inputting the scenic area ticketing keywords into the ticket verification sub-agent to query and verify scenic area ticketing information to obtain ticket verification information; obtaining the ticket verification information by calling the data query and verification interface of the ticket verification sub-program using the scenic area ticketing keywords as input data; inputting the ticket verification information and the interactive information into the real-time Q&A sub-agent to generate text to obtain service text; and using the ticket verification information, the service text, and the access address of the ticket verification sub-program as the service result.

[0068] The ticketing keywords include ticket type, attraction name, verification code, and ticket purchaser information; ticket verification information includes ticket validity, verification status, remaining usage count, online verification portal, real-world location of offline verification gates, and navigation route; the access address for the ticket verification subroutine is an AR real-world verification page address that matches the user's real-world location. Users can click the address to directly enter the real-world verification page without having to search for the verification portal, achieving a one-stop interaction of "question and answer - verification" and improving the efficiency of ticket verification.

[0069] (ii) Collaborative processing of scenic area data between real-scene question-and-answer sub-agent and real-scene visitor flow sub-agent. This scenario represents the core service scenario for various attractions within the scenic area. The core user need is to query the real-time visitor flow. The collaborative processing combines real-time visitor flow data and real-time data collected from the scenic area's IoT sensing nodes to provide users with intuitive real-time visitor flow information. In another optional implementation of this embodiment, the service collaborative processing to obtain service results includes: the scenic area intelligent agent extracting keywords from the interactive information to obtain scenic area visitor flow keywords; inputting the scenic area visitor flow keywords into the visitor flow real-time sub-intelligent agent to query scenic area visitor flow information to obtain real-time visitor flow information; obtaining the real-time visitor flow information by calling the real-time data interface of the visitor flow real-time sub-program using the scenic area visitor flow keywords as input data (including real-time visitor flow data, visitor flow real-time images / videos, and visitor flow warning levels); inputting the visitor flow real-time information and the interactive information into the real-time question-and-answer sub-intelligent agent to generate service text; and using the visitor flow real-time information, the service text, and the access address of the visitor flow real-time sub-program as the service result.

[0070] In addition to the two core scenarios mentioned above, the real-scene question-answering sub-agent can also perform similar real-scene data collaborative processing with the real-scene navigation sub-agent, facility query sub-agent, and attraction introduction sub-agent. The core logic is "keyword extraction - candidate sub-agents obtain real-scene data - real-scene question-answering sub-agents generate fused text - aggregated data generates service results". All processing results combine the user's current geographic location and scenic area real-scene data to ensure the accuracy and real-scene nature of the answers.

[0071] (2) Collaborative processing of scenic area data between candidate sub-agents and sub-agents corresponding to the interaction information. If the target sub-agent is a candidate sub-agent (such as a ticketing verification sub-agent, a visitor flow scene sub-agent, or a facility query sub-agent), the service result is obtained through collaborative processing of scenic area data by the candidate sub-agent and the sub-agent adapted to the interactive information (the core being the scene-based question-and-answer sub-agent). This process is suitable for scenarios where users directly trigger the candidate sub-agent interface, such as when a user directly clicks the "visitor flow query" button instead of submitting a request through scene-based question-and-answer. Specifically, the scenic area agent determines the target sub-agent corresponding to the interactive information as a candidate sub-agent from among the configured sub-agents. The scenic area data is then collaboratively processed by the candidate sub-agent and the scene-based question-and-answer sub-agent. The candidate sub-agent provides professional scene-based data, while the scene-based question-and-answer sub-agent transforms the professional data into scene-based service text that is easy for users to understand. The results are then generated by integrating these elements, achieving the transformation of "professional data + popular scene-based interpretation".

[0072] In one optional implementation of this embodiment, after determining the target sub-agent operation corresponding to the interaction information among the configured multiple sub-agents, the method further includes: if the target sub-agent is a candidate sub-agent, obtaining service results by performing scenic area data collaborative processing between the candidate sub-agent and the real-scene question-and-answer sub-agent.

[0073] Specifically, this collaborative processing process uses the scenic area intelligent agent as the main coordinator to uniformly schedule candidate sub-intelligent agents and real-scene question-and-answer sub-intelligent agents for data processing, ensuring effective integration of data from both sides. In one optional implementation of this embodiment, the service result is obtained through collaborative processing of scenic area data by the candidate sub-intelligent agents and the real-scene question-and-answer sub-intelligent agents, including: the scenic area intelligent agent calling the candidate sub-intelligent agents to process scenic area data to obtain the access address of the scenic area subroutine corresponding to the candidate sub-intelligent agent, and querying the scenic area candidate interactive text preset by the candidate sub-intelligent agent in the scenic area knowledge base; generating the intelligent agent interface of the real-scene question-and-answer sub-intelligent agent corresponding to the scenic area candidate interactive text, inputting the intelligent agent information of the candidate intelligent agent into the real-scene question-and-answer sub-intelligent agent to generate service text in combination with real-time scenic area data, and generating service result according to the configuration information of the intelligent agent interface corresponding to the scenic area candidate interactive text and the service text.

[0074] For example, the target intelligent agent is the real-time visitor flow intelligent agent. The user directly clicks the "Visitor Flow Inquiry" button on the scenic area's service page. The scenic area intelligent agent first calls the real-time visitor flow intelligent agent to obtain the access address of the XX scenic spot's real-time visitor flow sub-program based on the user's geolocation data. Then, it queries the scenic area's knowledge base to find the corresponding scenic area candidate interactive text for "Visitor Flow Inquiry" (such as "Is XX scenic spot currently crowded?", "What are some scenic spots with less visitor flow around XX scenic spot?", "When is the peak visitor flow for XX scenic spot?"). This candidate interactive text consists of common questions related to visitor flow inquiries, pre-configured by the scenic area. Next, it generates the intelligent agent interface of the real-time question-and-answer sub-intelligent agent corresponding to this candidate interactive text, providing the user with a secondary real-time question-and-answer entry point. Then, it inputs the intelligent agent information of the real-time visitor flow intelligent agent (such as function type, data range, and real-time adaptation area) into the real-time question-and-answer sub-intelligent agent. The real-time question-and-answer sub-intelligent agent combines the scenic area's real-time visitor flow data to generate easy-to-understand real-time service text (such as "XX scenic spot currently has 800 visitors, which is crowded. What are some scenic spots with less visitor flow around XX scenic spot?"). The bamboo forest scenic spot only had 200 visitors, which is a small number. You can click the link below to view the real-time visitor flow map of the XX scenic spot. Finally, the intelligent agent interface configuration information and service text corresponding to the candidate interactive text are integrated into the service result, which not only provides users with professional real-time visitor flow data, but also provides real-time service text and secondary interaction entry points, thereby improving the extensibility of the service.

[0075] It should be noted that the collaborative processing method between the candidate sub-agent and the real-scene question-and-answer sub-agent can be flexibly adopted according to the actual needs of the scenic area services. The three collaborative methods mentioned above, in which the real-scene question-and-answer sub-agent is the main agent, are all based on the core principle that "the candidate sub-agent provides professional real-scene data, and the real-scene question-and-answer sub-agent performs real-scene text conversion and fusion". This ensures that the service results are both professional and in line with the user's real-scene interaction needs, allowing the user to quickly understand and use the scenic area services.

[0076] Step S204: Generate service response data based on the service result and send it to the user terminal.

[0077] In practice, the server extracts multi-dimensional real-scene scenic area data from the service results, generates customized real-scene service response data based on the real-scene display requirements of the scenic area service, and sends it to the user terminal. Unlike the general response data of traditional cultural tourism services, the service response data in this embodiment is customized data adapted to the offline real-scene tour of the scenic area, including various forms such as text, real-scene pictures / videos, AR interactive entrances, and subroutine access addresses, which can be displayed on the AR / VR real-scene page of the user terminal. Correspondingly, the user terminal receives the service response data and renders and displays it on the scenic area service page, while linking with the real-scene data of the scenic area to achieve real-scene display, allowing users to obtain an immersive service experience.

[0078] Service response data refers to exclusive response data in response to real-world interactive information sent by the user terminal. The user terminal can display this data on the regular page, AR real-world page, and VR panoramic page of the scenic area service. The display formats include real-world messages, AR rich text, real-world banners, and real-world pop-ups. All display formats are linked to the user's current offline real-world scene. For example, AR rich text can be directly overlaid on the real-world scene image taken by the user. Optionally, service response data includes: real-world text data, access addresses of scenic area subprograms / subservices, AR / VR rich text, configuration information of intelligent agent interfaces, associated addresses of scenic area real-world images / videos, configuration information of real-world interaction entry points, and real-time operation data of the scenic area.

[0079] Specifically, in the service response data generation process, the server first classifies the scenic area data included in the service result according to the data type, and then writes the classified real-scene service data into a scenic area-specific data template to generate standardized service response data. This method is suitable for regular scenic area service requests and has high generation efficiency. In an optional implementation method provided in this embodiment, generating service response data based on the service result includes: classifying the scenic area data included in the service result according to the data type to obtain scenic area service data corresponding to each data type; and writing the classified scenic area service data into a scenic area data template to obtain the service response data. The data types include real-scene service text, professional real-scene data, subroutine / subservice access addresses, intelligent agent interface configuration information, and real-scene media data (images / videos). The scenic area data template is a pre-configured real-scene display template for the scenic area, containing multiple display styles that can automatically adapt to the user's geographic location data and real-scene feature data. For example, if the user is at an outdoor scenic spot, the outdoor real-scene display style will be adapted; if the user is at an indoor venue, the indoor real-scene display style will be adapted.

[0080] In practical applications, the server pre-stores scenic area-specific service elements (such as scenic area real-scene icons, AR navigation controls, real-scene interactive buttons, scenic area logos, etc.). The service results also include the intelligent agent interface configuration information of the sub-intelligent agents. These service elements and interface configuration information can be integrated with the service results to generate a service response message with a real-scene interactive entry point, providing users with a one-stop interactive experience. In one optional implementation of this embodiment, generating service response data based on the service results includes: extracting the response text contained in the service results and the configuration information of the intelligent agent interface of the real-scene question-and-answer sub-intelligent agent corresponding to the scenic area candidate interactive text; generating a service response message based on the response text and / or preset scenic area service elements; configuring the intelligent agent interface corresponding to the scenic area candidate interactive text in the service response message according to the configuration information; and using the configured service response message as the service response data. This service response message not only contains the response text of the user's request but also configures the relevant intelligent agent interface. Users can directly click on the interface to ask a second question without returning to the main service page, improving interaction efficiency.

[0081] In addition, the service results also include the access address of the scenic area sub-program / sub-service and the real-world response text generated by the real-world question-and-answer sub-agent. The server can merge the access address with the real-world images / videos of the scenic area to generate AR / VR rich text as service response data. This rich text can be directly displayed on the AR real-world page of the user's terminal, integrating text, links, and real-world images to achieve a deep integration of "text + real-world + interaction". In another optional implementation provided in this embodiment, generating service response data based on the service results includes: extracting the access address of the scenic area sub-program corresponding to the candidate sub-agents included in the service results and the response text generated by the real-world question-and-answer sub-agent; and generating rich text based on the access address and the response text combined with the real-world images / videos of the scenic area and using it as the service response data. For example, to meet the needs of ticket verification, the generated AR rich text can include the response text "Click here for AR verification", the access address of the ticket verification subroutine, and a real-world image of the scenic area's verification gate. After clicking the link, users can directly enter the AR verification page, scan the ticket QR code to complete the verification, and achieve an immersive ticket verification experience.

[0082] It should be noted that the three service response data generation methods provided above can be flexibly combined according to the actual display needs of scenic area services and the type of user requests. For example, for multi-dimensional real-world Q&A requests, the reply text, multiple subroutine access addresses, multiple smart agent interface configuration information, and real-world image / video addresses in the service results can be extracted and integrated to generate AR rich text with multiple interactive entry points and real-world images as service response data. This ensures that the content displayed on the user terminal is both rich and comprehensive, and also fits the needs of real-world tours, achieving "one-time response, multi-dimensional interaction".

[0083] After generating service response data, the server sends it to the user terminal through an encrypted communication link established with the user terminal. During the sending process, the server can adaptively adjust the data transmission format according to the network status of the user terminal (such as Wi-Fi / 5G / 4G). For example, when the network status is good, it sends AR rich text + high-definition real-scene video, and when the network status is poor, it sends a simplified version of real-scene text + thumbnail, ensuring the stability and timeliness of the service response.

[0084] In summary, the data processing method (server-side) based on scenic area intelligent agents provided in this embodiment ensures accurate binding between scenic area services and users' offline real-world experiences through dual access verification of "geographic location matching + scenic area IoT sensing interaction." Using the real-world question-and-answer sub-intelligent agent as the core collaborative hub, it links various professional sub-intelligent agents to achieve real-world collaborative processing of scenic area data, integrating scenic area real-world models and real-time operational data to generate multi-dimensional service results. Based on the service results, it generates real-world service response data such as AR / VR rich text, and ensures efficient reception by user terminals through an adaptive distribution mechanism. The entire method is permeated with the core logic of "real-world integration," solving the pain points of insufficient accuracy, poor interactive experience, and low real-world integration in traditional scenic area services, providing users with immersive and intelligent scenic area services.

[0085] The following example demonstrates the complete application of the data processing method based on scenic area intelligent agents provided in this embodiment in ticket verification and real-world question-and-answer scenarios. The entire process of the method is broken down in detail. This scenario is a high-frequency service scenario for scenic areas, covering the core ticketing needs of users from entering the scenic area to visiting attractions: In step S302, the user terminal calls the positioning component to complete the scenic area geolocation matching, determines that the user is at the entrance of the scenic spot XX (meter-level positioning), and generates scenic area identification data containing the scenic spot code and entrance location; at the same time, it calls the near-field sensing component to perform near-field sensing interaction with the scenic area IoT sensing node deployed at the entrance, and obtains the encrypted sensing access data transmitted by the node. Step S304: The user terminal parses the encrypted sensing access data, combines the geolocation verification result to obtain the data transmission interface of the scenic area service, calls the interface with the encrypted sensing access data as input, and sends a service access request to the server. Step S306: The server receives the encrypted sensing access data from the user terminal, decrypts it to obtain the access instruction for the scenic area service page, and sends the exclusive scenic area service page for the XX scenic spot entrance to the user terminal based on the user's geolocation data and scenic spot identification data. In step S308, the user triggers the agent interface of the real-scene question-and-answer sub-agent on the service page, submits interaction information (including interface configuration information, interaction text "How to verify the ticket of XX attraction?", user location data, and real-scene feature data of the attraction entrance), and sends it to the server; Step S310: The server receives the interaction information, first checks the scenic area knowledge base to see if there is a matching preset scenic area response text. If the query result is not found, the scenic area intelligent agent is invoked. In step S312, the server calls the scenic area intelligent agent based on the interaction information. The scenic area intelligent agent detects that the interaction information contains real-world question and answer data and determines that the target sub-intelligent agent is the real-world question and answer sub-intelligent agent. The server performs intent recognition on the interaction text by combining the large language model with the scenic area real-world model library to obtain the intent category of "scenic area ticket verification" and determines the appropriate candidate sub-intelligent agent as the ticket verification sub-intelligent agent. In step S314, the scenic area intelligent agent determines the collaboration method as "the real-scene question-and-answer sub-intelligent agent acts as the main intelligent agent and calls the candidate sub-intelligent agents"; the real-scene question-and-answer sub-intelligent agent inputs the interactive text into the large language model and combines it with the scenic area real-scene model to generate service data ("You are currently at the entrance of XX scenic spot, and you can verify your ticket through online AR verification or offline gate verification"), and extracts the keywords "XX scenic spot, ticket verification"; Step S316: The real-scene question-and-answer sub-agent calls the ticket verification sub-agent based on keywords. The ticket verification sub-agent queries the user's bound ticket information through the data query and verification interface of the ticket verification sub-program, and obtains the ticket verification information (ticket valid, 1 use remaining, online AR verification entrance, offline gate No. 3 real-scene location and navigation route); at the same time, it generates the access address of the AR real-scene verification page of the ticket verification sub-program. In step S318, the ticket verification sub-agent feeds back the ticket verification information and AR real-scene verification address to the real-scene question-and-answer sub-agent. The real-scene question-and-answer sub-agent integrates the service data and ticket verification information to generate a complete service text. The scenic area agent aggregates the service text, ticket verification information, AR real-scene verification address, and ticket verification sub-agent interface configuration information into a service result. Step S320: The server extracts the data from the service results, classifies it according to data type, and generates a service response message by combining scenic area service elements (real-scene gate icon, AR verification button), and configures the interface of the ticket verification sub-intelligent agent in the message; Step S322: The server sends the configured service response message to the user terminal; In step S324, the user terminal receives the service response message, displays the service text, the real-world location of the offline gate and navigation route on the scenic area service page, and displays the AR verification entry button; the user clicks the AR verification button and is redirected to the AR real-world verification page of the ticket verification subroutine; In step S326, the user scans the ticket QR code on the AR page to complete the verification. The ticket verification sub-agent synchronizes the verification result to the server, and the server sends a real-world notification message (including verification success text and attraction visit prompts) to the user terminal.

[0086] The following example demonstrates the practicality of the data processing method based on scenic area intelligent agents provided in this embodiment in a real-time visitor flow query scenario within a scenic area. This scenario covers the peak-avoidance needs of users during their visit: Step S402: The user terminal enters the scenic area service through the regular entry point (application search), calls the positioning component to complete the geolocation matching, and determines that the user is at the viewing platform of the XX Waterfall scenic spot in the scenic area. In step S404, the user triggers the real-time visitor flow sub-intelligent agent interface configured on the scenic area service page and submits interactive information (including interface configuration information, user location data, and real-time feature data of the viewing platform). Step S406: The user terminal sends the interaction information to the server; In step S408, the server receives the interaction information and calls the scenic area intelligent agent based on the interface configuration information. The scenic area intelligent agent determines that the target sub-intelligent agent is the real-scene sub-intelligent agent of passenger flow and starts collaborative processing with the real-scene question-and-answer sub-intelligent agent. Step S410: The scenic area intelligent agent calls the real-time visitor flow sub-intelligent agent, queries the real-time visitor flow data of XX Waterfall scenic spot based on user location data, obtains the real-time visitor flow real-time images and visitor flow statistics (currently 800 people, congestion level 3) of the scenic area through the scenic area IoT sensing node, and generates the access address of the visitor flow real-time sub-program. Step S412: The scenic area agent queries the scenic area knowledge base for the corresponding candidate interactive text for "visitor flow query" ("Which attractions in the surrounding area have less visitor flow?" "When will the peak visitor flow end?"). Step S414: The scenic area intelligent agent generates the real-scene question-and-answer sub-intelligent agent interface corresponding to the candidate interactive text, inputs the intelligent agent information (function type, data range) of the real-scene question-and-answer sub-intelligent agent into the real-scene question-and-answer sub-intelligent agent, and generates service text ("XX Waterfall Scenic Spot currently has 800 visitors and is in a crowded state, while the surrounding XX Bamboo Forest Scenic Spot has 200 visitors and is less crowded. You can click to view the real-scene visitor flow map"). In step S416, the scenic area intelligent agent aggregates the service text, real-time visitor flow information, real-time visitor flow subroutine address, and interface configuration information corresponding to the candidate interaction text into a service result. In step S418, the server extracts the data from the service results and combines it with the scenic area's real-scene images to generate AR rich text (including service text, real-scene images of passenger flow, candidate interactive text interfaces, and passenger flow subroutine addresses), which is then sent out as service response data. In step S420, the user terminal receives AR rich text and displays it on the scenic area service AR page, overlaying the real-life images of visitor flow onto the real-life images of the viewing platform taken by the user; the user triggers the "What are some nearby attractions with less visitor flow?" interface and submits secondary interaction information; In step S422, the server receives the secondary interaction information, calls the scenic area intelligent agent, and through the collaborative processing of the real-scene question-and-answer sub-intelligent agent and the scenic spot recommendation sub-intelligent agent, generates a list of surrounding low-traffic scenic spots and navigation routes, which are then distributed in rich text format. In step S424, the user terminal displays secondary response data, and the user adjusts the tour route based on the data to meet the peak passenger flow avoidance needs.

[0087] Furthermore, the data processing method of this application specifically includes steps S502 to S506. This method is the core processing flow on the user terminal side, which works in conjunction with the server-side method to realize the real-world access, interaction and display of scenic area services. All steps revolve around the "user's offline real-world experience" to ensure ease of operation and intuitive display.

[0088] Step S502: Obtain the interaction information submitted by the user after triggering the intelligent agent interface for the scenic area service configuration.

[0089] The scenic area service described in this embodiment is an online + offline integrated service exclusive to the scenic area. It is provided by the scenic area service server and deployed in the application or subroutine on the user's terminal. The core functions include real-scene Q&A, ticket verification, real-scene navigation, visitor flow query, facility location, etc. All functions are accurately matched with the user's current offline real-scene location, support AR / VR real-scene interaction, and meet the user's immersive service needs during the scenic area visit.

[0090] Specifically, the scenic area service is configured with a scenic area intelligent agent and multiple real-world sub-intelligent agents. The scenic area service configures corresponding intelligent agent interfaces for the functions of each sub-intelligent agent. These interfaces are deployed in various forms, such as the scenic area service's regular pages, AR real-world pages, and VR panoramic pages. During their visit to the scenic area, users can activate the intelligent agent interfaces through touch clicks, voice activation, or triggering real-world photography to submit interactive information related to the offline real-world environment. For example, after a user takes a picture of a facility in the scenic area, the interface is triggered to query the facility's usage status.

[0091] Among them, the scenic area intelligent agent is a dedicated intelligent agent for scenic area services. Its sub-intelligent agents include real-scene question-and-answer sub-intelligent agents, ticket verification sub-intelligent agents, real-scene tour guide sub-intelligent agents, etc. The processing results of each sub-intelligent agent are fed back to the user terminal in a real-scene form. The user terminal accesses the scenic area services through a dual method of "geographic location matching + near-field perception interaction of scenic area IoT sensing nodes". This access method is the foundation for realizing the real-scene service, ensuring that the service page entered by the user is highly consistent with the offline real scene and avoiding information redundancy on the general service page.

[0092] In specific implementation, the core process for user terminals to access scenic area services is as follows: First, a high-precision positioning component is used to complete the scenic area geolocation matching to determine the user's specific location within the scenic area (such as attractions, entrances, or surrounding facilities), generating scenic area identification data containing location codes and regional characteristics; then, a near-field sensing component interacts with the scenic area IoT sensing node deployed at that location to obtain encrypted sensing access data transmitted by the node; finally, the encrypted sensing access data is parsed and combined with the geolocation verification result to obtain the data transmission interface for the scenic area service, and the interface is called to send an access request to the server, receiving and displaying the exclusive scenic area service page issued by the server; in an optional implementation provided in this embodiment, before obtaining the interaction information, the process further includes: calling the positioning component to complete the scenic area geolocation matching, and calling the near-field sensing component to interact with the scenic area IoT sensing node to obtain the encrypted sensing access data transmitted by the scenic area IoT sensing node; parsing the encrypted sensing access data and combining it with the geolocation verification result to obtain the data transmission interface for the scenic area service, using the encrypted sensing access data as the interface call input to call the data transmission interface; receiving and displaying the scenic area service page issued by the server.

[0093] In addition, user terminals also support other access methods, such as searching for the scenic spot name + keywords (e.g., "XX Scenic Spot XX Attraction Service") through the application to obtain the service access address and enter the scenic spot service page based on the address; or scanning the QR code in the scenic spot and parsing the access address contained in the QR code to enter the service page; regardless of the access method, the user terminal will automatically complete the geolocation matching to ensure the accurate binding between the service page and the user's real-world location.

[0094] In real-world scenarios, user terminals can access scenic area services whether the screen is unlocked or locked. When unlocked, users can directly access services through near-field sensing interaction or application search. When locked, user terminals can receive wake-up signals from scenic area IoT sensing nodes through near-field sensing components, and automatically enter the scenic area service page after unlocking. The scenic area service page supports a floating window display mode, allowing users to view the service page while it is floating during their visit, without affecting their view of the actual scenery and improving ease of operation.

[0095] In this step, the interactive information acquired by the user terminal is a service request made by the user based on the offline real scene, which has strong real scene characteristics and is different from the general interactive information of traditional cultural and tourism services. The interactive information specifically includes: the configuration information of the intelligent agent interface (sub-intelligent agent identifier, function type), the scenic area interaction data entered by the user (text / voice / real scene media data), the geographic location data of the user terminal (meter-level accuracy), and real scene feature data (scenic area features of the area where the user is located, surrounding facility information, and real scene image features captured). For example, if the user triggers the interface near the scenic area restroom, the submitted interactive information may include the text data of "Is there a stall available in the restroom?", the location data of the surrounding area of ​​the restroom, and the real scene image of the restroom entrance captured by the user.

[0096] In the specific process of acquiring interactive information, the user terminal can collect user input in various ways: text input (keyboard, handwriting), voice input (real-time speech recognition into text), and real-world media input (taking pictures / videos and extracting feature data). The user terminal will also automatically supplement geolocation data and real-world feature data, eliminating the need for manual input by the user, reducing operation steps, and improving interaction efficiency. For example, if the user only enters "ticket verification", the user terminal will automatically supplement the current location data and surrounding real-world feature data to generate complete interactive information.

[0097] Step S504: Send the interaction information to the server.

[0098] In practice, the user terminal sends complete interactive information (including intelligent agent interface configuration information, scenic area interactive data, geolocation data, and real-world feature data) to the server through an encrypted communication link established with the server, ensuring the security and integrity of data transmission. During the transmission process, the user terminal compresses the interactive information to reduce the amount of data transmitted and adapt to the possible weak network environment within the scenic area. If the network is interrupted, the user terminal caches the interactive information and automatically resends it after the network is restored, ensuring that service requests are not lost.

[0099] After the interactive information is sent, the user terminal will display a real-world prompt of "Service Response in Progress" (such as overlaying an animation on an AR page) to inform the user of the service processing progress and improve the user experience. At the same time, the user terminal can continue to receive other operations from the user and support multi-tasking in parallel. For example, while waiting for the passenger flow query response, the user can trigger the facility query interface to submit new interactive information.

[0100] After receiving the interaction information, the server will perform data verification and parsing. Based on the geolocation data and real-scene feature data in the interaction information, it will call the corresponding scenic area intelligent agent and sub-intelligent agents for collaborative processing to generate service results. If the server detects that there is missing data in the interaction information (such as the absence of geolocation data), it will send a supplementary data request to the user terminal. After receiving the request, the user terminal will automatically re-collect the missing data and send it to the server without the need for manual operation by the user.

[0101] Step S506: Receive the service response data generated and sent by the server based on the service result, and display it on the scenic area service page.

[0102] In practice, the user terminal receives service response data from the server. This data is real-world data specific to the scenic area, including various forms such as real-world text, real-world images / videos, AR / VR interactive entry points, and subprogram access addresses. The user terminal selects the corresponding display method according to the type of service response data. The core display principle is "real-world, intuitive, and convenient interaction" to ensure that users can quickly obtain key information during their visit.

[0103] The service response data is displayed in several core ways, which can be flexibly switched according to the data content and user scenario: (1) AR Real-Scene Overlay Display: Applicable to service response data containing real-scene images / videos and navigation routes. The user terminal uses AR components to overlay service text, interactive entry points, and navigation routes onto the real-scene images of the scenic area taken by the user, realizing the integrated display of "virtual information + real-scene images". For example, the service response data for ticket verification can overlay the AR verification entry point and the gate navigation route onto the real-scene image of the scenic spot entrance taken by the user. The user can directly click on the verification entry point in the AR image to complete the operation. (2) VR panoramic display: Applicable to service response data such as overall scenic area guide and attraction introduction. The user terminal calls the VR component to display the VR panoramic view of the scenic area or attraction, and overlays service text and interactive entry. Users can browse the panorama by sliding, zooming and other operations, and click the entry to trigger further services; for example, the service response data of attraction introduction can display the VR panorama of the attraction, overlay historical and cultural introduction text and "audio explanation" entry, and users can listen to the explanation while browsing the panorama; (3) Real-scene pop-up display: Applicable to simple service prompt data (such as successful verification, passenger flow warning). The user terminal will pop up a real-scene pop-up on the scenic area service page, which includes service text and real-scene icons of the scenic area. The pop-up does not obstruct the core sightseeing view and users can quickly view and close it. (4) Rich text list display: Suitable for multi-dimensional information data (such as facility list, attraction recommendation). The user terminal displays the information list in rich text form. Each list item contains real-scene pictures, core information and interactive entry points. Users can click on the list item to view details or trigger related services.

[0104] In one optional implementation of this embodiment, displaying service response data includes: parsing the service response message contained in the service response data to obtain service text and configuration information of the agent interfaces of candidate sub-agents; displaying the service text on the scenic area service page and, according to the configuration information, displaying the agent interfaces of the candidate sub-agents on the scenic area service page, while simultaneously loading corresponding scenic area real-scene data for linked display. For example, when a user queries the service response data for "nearby restrooms," the user terminal parses the data to obtain the location information, real-scene images, and agent interface configuration information of three restrooms. The distance, availability, and real-scene images of each restroom are displayed in a rich text list format, and "Navigation" and "View Real-Time Footage" interfaces are displayed in the list items. Clicking "Navigation" triggers the real-scene navigation sub-agent to obtain an AR navigation route.

[0105] In practical applications, user terminals also support secondary interaction of service response data. This means that users trigger the intelligent agent interface on the display page, submitting new interactive information to form a closed-loop interaction of "request-response-secondary request-secondary response." In one optional implementation of this embodiment, after receiving and displaying the service response data, the method further includes: obtaining the target interactive text submitted by the intelligent agent interface of the real-world question-and-answer sub-intelligent agent corresponding to the target interactive text configured by the user on the scenic area service page; sending the target interactive text to the server; receiving the rich text obtained by the server through the scenic area intelligent agent call for the scenic area service; and displaying the rich text on the scenic area service page and associating it with the corresponding scenic area real-world interactive entry. For example, after viewing visitor flow data, a user triggers the "Nearby Low-Flow Attractions" interface, submitting secondary interactive text. The server generates a rich text response through collaborative processing by the scenic area intelligent agent, and the user terminal displays the rich text and associates it with the "Navigate to this attraction" real-world interactive entry, achieving one-stop service.

[0106] In addition, the user terminal supports caching and offline viewing of service response data. For information that users may view repeatedly (such as attraction introductions and tour routes), the user terminal automatically caches it locally, and users can view it offline later in an environment without network access. At the same time, the user terminal can automatically push relevant service response data based on the user's tour trajectory. For example, when the user reaches the exit of an attraction, the service information of nearby catering facilities will be automatically pushed, improving the proactiveness of the service.

[0107] In summary, the data processing method (user terminal) based on scenic area intelligent agents provided in this embodiment achieves accurate access to scenic area services through dual access verification, simplifying user operation processes; it meets users' convenient input needs during the tour through diverse interactive information collection methods; it enhances the intuitiveness and extensibility of services through realistic display and secondary interaction functions; and it works in conjunction with server-side methods to provide users with an immersive and intelligent scenic area tour experience.

[0108] To facilitate better implementation of the data processing method based on scenic area intelligent agents in this application embodiment, this application embodiment also provides a data processing device based on scenic area intelligent agents, wherein the meanings of the terms are the same as those in the above-described data processing method based on scenic area intelligent agents, and specific implementation details can be found in the description in the system embodiment.

[0109] Please see Figure 3 , Figure 3 This is a schematic diagram of the structure of a data processing device based on a scenic area intelligent agent provided in an embodiment of this application. Specifically, the data processing device may include a receiving module 310, a calling module 320, a processing module 330, and a sending module 340, as follows: The receiving module 310 is used to receive the interactive information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is entered after the user terminal completes the geolocation matching and performs near-field perception interaction with the scenic area IoT sensing node. The calling module 320 is used to call the scenic area intelligent agent for the scenic area service based on the interaction information, so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information among the multiple configured sub-intelligent agents; Processing module 330 is used to obtain service results by performing scenic area data collaborative processing through the real-scene question-and-answer sub-intelligent agent and the sub-intelligent agent adapted to the interaction information if the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent. The delivery module 340 is used to generate service response data based on the service result and deliver it to the user terminal.

[0110] This application provides a data processing device based on a scenic area intelligent agent. A receiving module 310 receives interaction information submitted by a user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and performs near-field sensing interaction with the scenic area's IoT sensing nodes. Then, a calling module 320 calls the scenic area intelligent agent based on the interaction information, so that the scenic area intelligent agent determines the target sub-intelligent agent corresponding to the interaction information among multiple configured sub-intelligent agents. If the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent, the processing module 330 uses the real-scene... The question-and-answer sub-agent and the sub-agent adapted to the interaction information perform collaborative processing of scenic area data to obtain service results. Finally, the distribution module 340 generates service response data based on the service results and distributes it to the user terminal. In the data processing scheme based on scenic area intelligent agents provided in this application, by integrating technologies such as geolocation, IoT sensing, and real-scene interaction, a scenic area-specific intelligent agent system is constructed to realize the collaborative processing of various sub-agents in the scenic area. This allows for a deep integration of online services and offline real-scene experiences, improving the intelligence, real-scene experience, and accuracy of scenic area services, and meeting the diversified and real-scene interaction needs of users during their visit to the scenic area.

[0111] Furthermore, embodiments of this application also provide an electronic device, such as... Figure 4 As shown, it illustrates a structural schematic diagram of the electronic device involved in the embodiments of this application, specifically: The electronic device may include components such as a processor 301 with one or more processing cores, a memory 302 with one or more processor-readable storage media, a power supply 303, and an input unit 304. Those skilled in the art will understand that... Figure 4 The electronic device structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or have different component arrangements. Wherein: Processor 301 is the control center of the electronic device. It connects various parts of the electronic device via various interfaces and lines. By running or executing software programs and / or modules stored in memory 302, and by calling data stored in memory 302, it performs various functions and processes data, thereby providing overall monitoring of the electronic device. Optionally, processor 301 may include one or more processing cores; preferably, processor 301 may integrate an application processor and a modem processor. The application processor mainly handles the operating system, user interface, and applications, while the modem processor mainly handles wireless data processing based on the scenic area intelligent agent. It is understood that the modem processor may not be integrated into processor 301.

[0112] The memory 302 can be used to store software programs and modules. The processor 301 executes various functional applications and data processing methods based on scenic area intelligent agents by running the software programs and modules stored in the memory 302. The memory 302 may mainly include a program storage area and a data storage area. The program storage area may store the operating system, application programs required for at least one function (such as sound playback function, image playback function, etc.), etc.; the data storage area may store data created according to the use of the electronic device, etc. In addition, the memory 302 may include high-speed random access memory, and may also include non-volatile memory, such as at least one disk storage device, flash memory device, or other volatile solid-state storage device. Accordingly, the memory 302 may also include a memory controller to provide the processor 301 with access to the memory 302.

[0113] The electronic device also includes a power supply 303 that supplies power to various components. Preferably, the power supply 303 can be logically connected to the processor 301 through a power management system, thereby enabling functions such as charging, discharging, and power consumption management through the power management system. The power supply 303 may also include one or more DC or AC power supplies, recharging systems, power fault detection circuits, power converters or inverters, power status indicators, and other arbitrary components.

[0114] The electronic device may also include an input unit 304, which can be used to receive input digital or character information and generate keyboard, mouse, joystick, optical or trackball signal inputs related to user settings and function control.

[0115] Although not shown, the electronic device may also include a display unit, etc., which will not be described in detail here. Specifically, in the embodiments of this application, the processing 301 in the electronic device loads the executable files corresponding to the processes of one or more applications into the memory 302 according to the following instructions, and the processing 301 runs the applications stored in the memory 302 to realize various functions, as follows: The system receives interaction information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and performs near-field perception interaction with the scenic area's IoT sensing nodes. Based on the interaction information, the system invokes the scenic area intelligent agent of the service so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information from among multiple configured sub-intelligent agents. If the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent, the system performs scenic area data collaborative processing with the sub-intelligent agent adapted to the interaction information to obtain a service result. Based on the service result, the system generates service response data and sends it to the user terminal.

[0116] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0117] This application embodiment integrates technologies such as geolocation, IoT sensing, and real-scene interaction to construct a scenic area-specific intelligent agent system. This system enables collaborative processing among various sub-intelligent agents within the scenic area, allowing for a deep integration of online services with offline real-scene experiences. This enhances the intelligence, realism, and precision of scenic area services, meeting users' diverse and real-scene interactive needs during their visit to the scenic area.

[0118] Those skilled in the art will understand that all or part of the steps in the various methods of the above embodiments can be performed by instructions, or by instructions controlling related hardware. These instructions can be stored in a processor-readable storage medium and loaded and executed by a processor.

[0119] Therefore, embodiments of this application provide a storage medium storing multiple instructions that can be loaded by a processor to execute steps in any of the data processing methods based on scenic area intelligent agents provided in embodiments of this application. For example, the instructions can execute the following steps: The system receives interaction information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and performs near-field perception interaction with the scenic area's IoT sensing nodes. Based on the interaction information, the system invokes the scenic area intelligent agent of the service so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information from among multiple configured sub-intelligent agents. If the target sub-intelligent agent is a real-scene question-and-answer sub-intelligent agent, the system performs scenic area data collaborative processing with the sub-intelligent agent adapted to the interaction information to obtain a service result. Based on the service result, the system generates service response data and sends it to the user terminal.

[0120] For details on the implementation of each of the above operations, please refer to the previous examples, which will not be repeated here.

[0121] The storage medium may include: read-only memory (ROM), random access memory (RAM), disk or optical disk, etc.

[0122] Since the instructions stored in the storage medium can execute the steps in any of the data processing methods based on scenic area intelligent agents provided in the embodiments of this application, the beneficial effects that any of the data processing methods based on scenic area intelligent agents provided in the embodiments of this application can achieve can be realized. For details, please refer to the previous embodiments, which will not be repeated here.

[0123] The above provides a detailed description of a data processing method, apparatus, electronic device, and storage medium based on a scenic area intelligent agent provided in the embodiments of this application. Specific examples have been used to illustrate the principles and implementation methods of this application. The description of the above embodiments is only for the purpose of helping to understand the method and core ideas of this application. At the same time, for those skilled in the art, there will be changes in the specific implementation methods and application scope based on the ideas of this application. Therefore, the content of this specification should not be construed as a limitation of this application.

Claims

1. A data processing method based on scenic area intelligent agents, characterized in that, include: The system receives interactive information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is accessed after the user terminal completes geolocation matching and performs near-field perception interaction with the scenic area's IoT sensing node. Based on the interaction information, the scenic area intelligent agent is invoked to provide the scenic area service, so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information among the multiple configured sub-intelligent agents; If the target sub-agent is a real-scene question-and-answer sub-agent, the service result is obtained by the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information through scenic area data collaborative processing. Based on the service result, service response data is generated and sent to the user terminal.

2. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, The scenic area services can be accessed in the following ways: The access address of the scenic area service is obtained by parsing the scenic area identification data matched by the geolocation and the sensing access data transmitted by the scenic area IoT sensing node, and the scenic area service is accessed based on the access address.

3. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, The process of obtaining service results through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and a sub-agent adapted to the interaction information includes: Based on the scenic area interaction data contained in the interaction information, candidate sub-intelligent agents that are adapted to the scenic area interaction data are selected, and the real-scene question-and-answer sub-intelligent agent and the scenic area data collaboration method of the candidate sub-intelligent agents are determined. The service results are obtained by performing collaborative data processing on scenic area data according to the described scenic area data collaboration method.

4. The data processing method based on scenic area intelligent agents according to claim 3, characterized in that, The process of obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaboration method includes: The real-scene question-answering sub-agent inputs the scenic area interaction data from the interaction information into the big language model and combines it with the scenic area real-scene model to process the data and generate service data, and extracts the scenic area keywords and real-scene feature keywords contained in the service data; Based on the scenic area keywords and real-scene feature keywords, the candidate sub-agent is invoked to process scenic area data and obtain real-scene information of the scenic area. The service data, the real-scene information of the scenic area, and the configuration information of the agent interface of the candidate sub-agent are used as the service result.

5. The data processing method based on scenic area intelligent agents according to claim 3, characterized in that, The process of obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaboration method includes: The candidate sub-agents process scenic area data based on interactive information to obtain the access address of the scenic area sub-service corresponding to the candidate sub-agents. The real-scene question-and-answer sub-agent generates service data based on the access address and the scenic area's real-scene data, and uses the access address and the service data as the service result.

6. The data processing method based on scenic area intelligent agents according to claim 3, characterized in that, The process of obtaining service results through collaborative processing of scenic area data by the real-scene question-and-answer sub-agent and a sub-agent adapted to the interaction information includes: The scenic area intelligent agent performs intent recognition and keyword extraction on the scenic area interaction data in the interactive information to obtain multiple scenic area-related keywords corresponding to each intent category. Each scenic spot-related keyword is input into the candidate sub-agent corresponding to the intent category to process the scenic spot data, and the processing results and the service data generated by the real-scene question-and-answer sub-agent are obtained. The service result is obtained by aggregating the service data, processing results, and configuration information of the agent interface of the candidate sub-agents.

7. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, After determining the target sub-agent operation to be executed corresponding to the interaction information among the configured multiple sub-agents, the process further includes: If the target sub-agent is a candidate sub-agent, the service result is obtained by collaboratively processing scenic area data through the candidate sub-agent and the real-scene question-and-answer sub-agent.

8. The data processing method based on scenic area intelligent agents according to claim 7, characterized in that, The process of obtaining service results through collaborative processing of scenic area data by the candidate sub-agent and the real-scene question-answering sub-agent includes: The scenic area agent calls the candidate sub-agent to process scenic area data, obtains the access address of the scenic area subroutine corresponding to the candidate sub-agent, and queries the scenic area candidate interactive text preset by the candidate sub-agent in the scenic area knowledge base. Generate the agent interface of the real-scene question-and-answer sub-agent corresponding to the candidate interactive text of the scenic area, input the agent information of the candidate sub-agent into the real-scene question-and-answer sub-agent and combine it with the real-time data of the scenic area to generate service text, and generate service results according to the configuration information of the agent interface corresponding to the candidate interactive text of the scenic area and the service text.

9. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, Before the step of receiving the interactive information submitted after the intelligent agent interface configured on the scenic area service page by the user terminal is triggered is executed, the following steps are also included: The encrypted sensing access data submitted by the user terminal calling the data transmission interface of the scenic area service is obtained; the data transmission interface obtains the encrypted sensing access data transmitted by the scenic area IoT sensing node by parsing the data and combining it with the geolocation verification result. The encrypted access data is decrypted to obtain the access instruction for the scenic area service page, and the scenic area service page is sent to the user terminal in response to the access instruction.

10. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, The step of generating service response data based on the service result includes: The scenic area data included in the service results are classified and processed according to the data type to obtain the scenic area service data corresponding to each data type. The categorized scenic area service data is written into the scenic area data template to obtain the service response data.

11. The data processing method based on scenic area intelligent agents according to claim 1, characterized in that, After determining the target sub-agent operation corresponding to the interaction information among the configured multiple sub-agents and executing the operation, and before executing the operation of obtaining service results through scenic area data collaborative processing by the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information, the method further includes: Search the scenic area knowledge base to see if there is a preset scenic area response text that matches the scenic area interaction data in the interactive information; If it exists, obtain the preset scenic spot response text and write the preset scenic spot response text into the service result.

12. The data processing method based on scenic area intelligent agents according to claim 3, characterized in that, The process of obtaining the service result by performing scenic area data collaborative processing according to the scenic area data collaboration method includes: The scenic area intelligent agent extracts keywords from the interactive information to obtain scenic area ticketing keywords, and inputs the scenic area ticketing keywords into the ticketing verification sub-intelligent agent to query and verify scenic area ticketing information to obtain ticketing verification information; the ticketing verification information is obtained by calling the data query and verification interface of the ticketing verification subroutine using the scenic area ticketing keywords as input data. The ticket verification information and the interaction information are input into the real-world question-and-answer sub-agent to generate a service text, and the ticket verification information, the service text, and the access address of the ticket verification sub-program are used as the service result.

13. The data processing method based on scenic area intelligent agents according to claim 3, characterized in that, The step of generating service response data based on the service result includes: Extract the configuration information of the agent interface of the real-world question-and-answer sub-agent corresponding to the response text and the candidate interactive text of the scenic area contained in the service result, and generate a service response message based on the response text and the preset scenic area service elements; Based on the configuration information, the intelligent agent interface corresponding to the candidate interactive text of the scenic area is configured in the service response message, and the configured service response message is used as the service response data.

14. The data processing method based on scenic area intelligent agents according to claim 6, characterized in that, The step of generating service response data based on the service result includes: Extract the access address of the scenic area subroutine corresponding to the candidate sub-agent and the response text generated by the real-world question-and-answer sub-agent from the service results; Rich text is generated based on the access address and the response text, combined with real-life images / videos of the scenic area, and used as the service response data.

15. A data processing device based on a scenic area intelligent agent, characterized in that, include: The receiving module is used to receive the interactive information submitted by the user terminal after the intelligent agent interface configured in the scenic area service is triggered. The scenic area service is entered after the user terminal completes the geolocation matching and performs near-field perception interaction with the scenic area IoT sensing node. The invocation module is used to invoke the scenic area intelligent agent for the scenic area service based on the interaction information, so that the scenic area intelligent agent can determine the target sub-intelligent agent corresponding to the interaction information among the multiple configured sub-intelligent agents; The processing module is used to obtain service results by performing scenic area data collaborative processing through the real-scene question-and-answer sub-agent and the sub-agent adapted to the interaction information if the target sub-agent is a real-scene question-and-answer sub-agent. The delivery module is used to generate service response data based on the service result and deliver it to the user terminal.

16. An electronic device, characterized in that, include: A memory, a processor, and a processor program stored in the memory and executable on the processor, wherein the processor executes the program as the steps of the data processing method based on scenic area intelligent agents as described in any one of claims 1 to 14.

17. A storage medium, characterized in that, The computer processing program is stored and can be loaded by a processor and executed as described in any one of claims 1 to 14.

Images