CN-121999773-A - Scenic spot explanation machine system with AI interaction and data feedback

CN121999773ACN 121999773 ACN121999773 ACN 121999773ACN-121999773-A

Abstract

The invention relates to the field of explanation machines, in particular to a scenic spot explanation machine system with AI interaction and data feedback, which comprises a device initialization module, a content matching and template display module, an AI interaction response module, a scenic spot operation background and a background data analysis module, wherein the device initialization module is started and then loads a customized navigation interface, a GPS and Bluetooth Beacon dual mode is started to position and output position data, the content matching and template display module matches the position data with a built-in scenic spot database, the AI interaction response module receives tourist voice/touch questions, an ASR, LLM and NLP are processed and then retrieves a knowledge base to generate answers, the answers are synchronously displayed through TTS voice broadcasting and a screen, the tourist images and a route adjustment module generate personalized routes and push the personalized routes, the data encryption feedback module collects tourist question records and behavior data, the encrypted data are wirelessly transmitted to the scenic spot operation background, and the background data analysis module receives the feedback data, generates interest heat points and high-frequency question and answer statistics reports, and provides data support for scenic spot optimization.

Inventors

HE XIAOFENG
WANG LU
XING YUHUI
HE JUNYAO

Assignees

中领科瀚有限公司

Dates

Publication Date: 20260508
Application Date: 20260127

Claims (10)

1. Scenic spot explanation machine system with AI interaction and data feedback, which is characterized in that the system comprises: the device initialization module is used for automatically loading a customized guide interface after the interpreter is started, synchronously starting a GPS and Bluetooth Beacon dual-mode positioning function, and acquiring and outputting the current position data of the tourist in real time; The content matching and template displaying module is used for receiving the position data, matching corresponding information with the built-in scenic spot database, automatically calling image-text cards, digital persons, VR, video, map navigation or time axis event line templates according to the scenic spot types, and displaying explanation content on the touch screen; The AI interaction response module is used for receiving the tourist voice or the touch question, starting ASR voice recognition to convert text, analyzing semantics through LLM and NLP, searching a scenic spot knowledge base to generate an answer, triggering TTS voice broadcasting and synchronously displaying auxiliary information on a screen; The tourist portrait and route adjustment module is used for collecting the age, the stay time and the clicking behavior of the tourist to construct a dynamic portrait, generating an adjusted personalized route through a path planning algorithm by combining the real-time crowding degree of a scenic spot, and pushing the personalized route to the touch screen; The data encryption return module is used for collecting tourist question records, encrypting behavior data and transmitting the encrypted behavior data to the scenic spot operation background through wireless communication; The background data analysis module is used for receiving the feedback data, generating tourist interest heat point diagrams and high-frequency question-answer statistical reports, and providing data support for scenic spot optimization content and service.
2. The scenic spot explanation machine system with AI interaction and data return according to claim 1, wherein the image-text card, the digital person, the VR, the video, the map navigation or the time axis event line template is automatically called according to the scenic spot type, specifically, the real-time position data of the explanation machine end is received and checked by the processor, and compared and matched with an offline scenic spot database containing scenic spot coordinate range, type and core explanation content, and the corresponding display template is automatically called according to the natural scenery, the historical site scenic spot type or the cross-scenic spot moving scene.
3. The scenic spot explaining machine system with AI interaction and data feedback according to claim 1, wherein the said explanation content is presented on the touch screen, concretely, the said explanation machine presents the explanation content through the touch screen, the picture and text card template presents scenic spot main map and core points, the time axis template presents event nodes horizontally and can be expanded with details, the map navigation template displays surrounding scenic spots and recommended route, support sliding switch, zoom operation and click to view details, the explanation content synchronizes speaker audio and screen text subtitle, screen brightness is automatically adjusted along with ambient light, the content adopts pre-buffer mechanism to avoid jamming, tourists manually switch templates through shortcut buttons at the bottom of the screen.
4. The scenic spot lecturer system with AI interaction and data feedback according to claim 1, wherein the system is characterized in that the system receives guest voices or touch questions, starts ASR voice recognition to convert texts, specifically, a lecturer microphone monitors in real time, supports wake-up of wake-up words and touch voice key triggering, wake-up word configuration or self definition, a microphone has a noise reduction function, collects voices and transmits the voices to an ASR submodule, adopts local and cloud dual-mode processing, automatically switches the local area by a weak network, and the touch questions support text input by an input method or click on a high-frequency problem shortcut card for submitting, and the contents of voice conversion texts or touch input are all output to a subsequent semantic analysis link as text data.
5. The scenic spot explaining machine system with AI interaction and data feedback according to claim 1, wherein the text data input through ASR conversion or touch is received through LLM and NLP, redundant information is removed through preprocessing, word segmentation labeling and core keyword extraction are carried out through NLP, knowledge base classification labels are matched, LLM is called to realize deep semantic understanding so as to identify real intention of questioning, distinguish requirement types and disambiguation, a structured query instruction defining a knowledge base search range is generated, local and cloud dual-mode operation is supported, cloud precision is improved when a network is good, a local lightweight model guarantee basic function is switched when a weak network or no network is provided, and an analysis result is output to a knowledge base search module.
6. The scenic spot interpretation system with AI interaction and data feedback as claimed in claim 1, wherein the scenic spot knowledge base generates an answer, triggers TTS voice broadcast and synchronously displays auxiliary information on a screen, specifically, the scenic spot knowledge base includes a local buffer library and a cloud complete library, the local buffer library is matched with the cloud complete library when searching, the cloud is searched without corresponding answer, the LLM generates a concise and natural answer based on knowledge base core information when no item is completely matched, the TTS voice broadcast is triggered after generation, the TTS supports local and cloud dual-mode operation, the voice speed volume can be manually adjusted and can automatically adapt to environmental noise, the screen synchronously displays the auxiliary information such as text answer, related pictures and functional icons, supports clicking to view details, and the auxiliary information is stored for a period of time after broadcasting is finished, and returns to the original interpretation interface.
7. The scenic spot explanation machine system with AI interaction and data feedback according to claim 1, wherein the system is characterized in that the system collects age, stay time and clicking behavior of tourists to construct dynamic images, specifically, after the explanation machine is started, an age selection entry is popped up, a tourist touch selection or default mark is unknown, whether the tourists are in a scenic spot range is judged based on dual-mode positioning, stay time and operation data are recorded, touch screen operation data are collected in real time, interest labels are generated by associating the age, stay time and operation data, the images are dynamically updated, label weights are adjusted, and the image data are encrypted and stored locally and then returned to an operation background along with other behavior data.
8. The scenic spot lecturer system with AI interaction and data return according to claim 1, wherein the scenic spot real-time crowding degree is combined, an adjusted personalized route is generated through a path planning algorithm and is pushed to a touch screen, specifically, an operation background receives an encrypted data packet returned by the lecturer, the data such as tourist behaviors, question-answer interactions, images and the like are classified and sorted and stored in a correlated mode after decryption and verification, a visual interest heat point diagram marking the tourist crowding degree and interest labels is generated, and a high-frequency question-answer statistical report marking question-answer classification and inaccurate answer questions is counted and displayed through a visual interface, so that data support is provided for scenic spot optimizing explanation content, a knowledge base, scenic spot identification, service configuration and AI semantic analysis and route recommendation algorithm.
9. The scenic spot explanation machine system with AI interaction and data feedback according to claim 1, wherein the system is characterized in that the system collects and encrypts tourist question records and behavior data, and transmits the data to a scenic spot operation background through wireless communication, specifically, the explanation machine locally collects specified data such as tourist question related data, touch operation records, scenic spot stay time, age information, interest tags, AI question and answer logs and the like in real time, and adds unique equipment identification and time stamp after encryption processing, and the system preferentially uses scenic spot Wi-Fi transmission through multimode wireless communication, temporarily stores the data to be recovered for the network compensation when no network exists, adopts a transmission mode combining real time with batch for data verification, and the operation background receives the encrypted data packet and then analyzes and stores the encrypted data packet, and associates the equipment identification and the time stamp to facilitate retrospective analysis.
10. The scenic spot interpreter system with AI interaction and data feedback according to claim 1, wherein the receiving feedback data generates a tourist interest heat point map and a high-frequency question and answer statistics report, specifically, the operating background receives an encrypted data packet transmitted back by the interpreter, decrypts and checks, classifies and sorts tourist behaviors, question and answer interactions and portrait association data, generates a visual interest heat point map marking the aggregation degree of tourists and interest labels, and statistically classifies questions and answers and marks the high-frequency question and answer statistics report of inaccurate questions and answers.

Description

Scenic spot explanation machine system with AI interaction and data feedback Technical Field The invention belongs to the field of explanation machines, and particularly relates to a scenic spot explanation machine system with AI interaction and data feedback. Background The related technical scheme of the prior scenic spot explanation has obvious short boards, the core problems are concentrated in the following four types, and the tourist guiding experience and the scenic spot service optimizing efficiency are seriously affected: The intelligent interaction capability is lacking, the existing explanation equipment (fixed point trigger type and GPS positioning type) is mostly in a unidirectional information playing mode, audio/image-text contents can be output only according to preset logic, tourists cannot respond to real-time questions (such as scenic spot background details, service facility inquiry and the like) of the tourists, information acquisition is passive and the flexibility is low, and mobile phone App navigation supports partial search, but is complex in operation and cannot realize natural language interaction. The personalized service is lacking, the existing system does not establish a tourist image acquisition and analysis mechanism, can not adjust service by combining personalized information such as the age, interest preference, tourist rhythm and the like of the tourist, i.e. the recommended route is single and solidified, the explanation content is uniform, and the core requirements of different tourists (such as parent-child families, historical lovers and leisure tourists) are difficult to match. The content display form is solidified, the explanation interface mostly adopts a single template (pure audio and fixed graphics) and cannot be dynamically adapted according to scenes, such as development venation which is presented on a time axis required by historical sites, highlighting effect of graphics and text cards required by natural landscapes, clear path required by map modes required by walking navigation and the like, so that the information transmission efficiency is low and the immersion is insufficient. Most explanation devices do not design the functions of collecting and returning the behavior data of tourists, the scenic spot cannot acquire key data such as a tourist question hot spot, stay time, route preference and the like, and therefore, the explanation content is difficult to optimize in a targeted manner, the scenic spot layout is difficult to adjust or the service configuration is improved, and the service iteration lacks data support. The environment adaptability and the operation experience are insufficient, the indoor positioning of the GPS positioning type equipment is inaccurate, the mobile phone App navigation is limited by the screen size and the network environment (poor experience under weak network and operation cutting), the coverage range of the fixed point triggering type equipment is limited, and the scene adaptation short plates exist. Disclosure of Invention The invention aims to provide a scenic spot explaining machine system with AI interaction and data return, which solves the problems in the background technology. In order to solve the technical problems, the invention provides the following technical scheme: A scenic spot explaining machine system with AI interaction and data return, comprising: the device initialization module is used for automatically loading a customized guide interface after the interpreter is started, synchronously starting a GPS and Bluetooth Beacon dual-mode positioning function, and acquiring and outputting the current position data of the tourist in real time; The content matching and template displaying module is used for receiving the position data, matching corresponding information with the built-in scenic spot database, automatically calling image-text cards, digital persons, VR, video, map navigation or time axis event line templates according to the scenic spot types, and displaying explanation content on the touch screen; The AI interaction response module is used for receiving the tourist voice or the touch question, starting ASR voice recognition to convert text, analyzing semantics through LLM and NLP, searching a scenic spot knowledge base to generate an answer, triggering TTS voice broadcasting and synchronously displaying auxiliary information on a screen; the tourist portrait and route adjustment module is used for collecting the age, the stay time and the clicking behavior of the tourist to construct a dynamic portrait, generating an adjusted personalized route through a path planning algorithm in combination with the real-time crowding degree of a scenic spot, and pushing the personalized route to a touch screen; The background data analysis module is used for receiving the feedback data, generating tourist interest heat point diagrams and high-frequency question-answer statistical reports, and providin