An intelligent tour guide system based on AIGC

Abstract

The present invention relates to an intelligent guide system based on AIGC, belongs to the field of intelligent guide technology, and solves the problems of low accuracy and poor dynamic performance in the prior art. The system includes: a data collection module, which is used to obtain user data and exhibition hall data, wherein the user data includes reservation information, exhibition preferences and special needs, and the exhibition hall data includes space layout and exhibit information; a real-time perception module, which is used to perceive the user's line of sight, stay time and crowd density in real time; a first guide route generation module, which is used to generate a first guide route according to the user data and exhibition hall data; an interaction module, which is used to interact with the user and analyze the user's actual exhibition needs according to the user's feedback on the first guide route; a second guide route generation module, which is used to adjust the first guide route according to the special needs, the actual exhibition needs and the crowd density to generate a final guide route.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent navigation, and in particular relates to an intelligent navigation system based on AIGC. Background Art

[0002] With the development of artificial intelligence, many industries are moving towards intelligence. At present, some exhibition halls have begun to apply intelligent guide technology, distributing headphones to visitors, and automatically introducing corresponding exhibits to visitors by locating their positions. In addition, there is also a guided tour method in which QR codes are posted on the exhibits for visitors to scan and listen to the explanations.

[0003] However, in the existing intelligent guide system, the voice broadcast method based on the spatial positioning of the visitor ignores the visitor's real intention to visit the exhibition. Perhaps the visitor just mistakenly enters an area, or just passes through an area on the way to the target area. The visitor is actually not interested in these areas; and although the method of scanning the QR code to listen to the explanation avoids misoperation, the information contained in the QR code is preset in advance and cannot form instant information expansion at the exhibition site; therefore, the existing intelligent guide system has technical problems of low accuracy and poor dynamic performance, which affects the exhibition experience. Summary of the invention

[0004] In view of the above analysis, the embodiment of the present invention aims to provide an intelligent navigation system based on AIGC to solve the technical problems of low accuracy and poor dynamic performance in the prior art.

[0005] The object of the present invention is achieved in that:

[0006] An intelligent tour guide system based on AIGC, comprising:

[0007] A data collection module is used to obtain user data and exhibition hall data, wherein the user data includes reservation information, exhibition preferences and special needs, and the exhibition hall data includes space layout and exhibit information;

[0008] Real-time perception module, used to perceive the user's sight direction, dwell time and crowd density in real time;

[0009] A first guided tour generating module, electrically connected to the data collecting module, and configured to generate a first guided tour according to user data and exhibition hall data;

[0010] An interaction module, electrically connected to the first guided tour generating module, the interaction module being used to interact with the user and analyze the user's actual exhibition viewing needs according to the user's feedback on the first guided tour;

[0011] The second guided route generation module is electrically connected to the data collection module, the real-time perception module, the first guided route generation module and the interaction module. The second guided route generation module is used to adjust the first guided route according to the special needs, the actual exhibition viewing needs and the crowd density to generate a final guided route.

[0012] Furthermore, the data collection module includes: a user data collection unit, which is used to obtain the reservation information and special needs entered by the user in the front-end interface, and to analyze the exhibition preferences based on the user's reservation information and historical exhibition records; an exhibition hall data collection unit, which is used to obtain the spatial layout and exhibit information inside the exhibition hall; and a data storage unit, which is used to store the data obtained by the user data collection unit and the exhibition hall data collection unit.

[0013] Furthermore, the real-time perception module includes: a sight tracking unit for capturing the user's sight direction and the time the sight stays on a certain exhibit in real time; a density detection unit for detecting the crowd density of each exhibit area in the exhibition hall in real time through image recognition technology.

[0014] Furthermore, the first guided route generation module includes: a path planning unit, which is used to generate at least two guided routes according to the user's reservation information, and sort the priorities of the guided routes according to the exhibition viewing preferences, and set the guided route with the highest priority as the first guided route.

[0015] Furthermore, the interaction module includes: a first guide unit, used to guide the movement of visitors according to the first guide route and the position of the visitors; a voice broadcast unit, used to introduce the current exhibit to the user when the line of sight tracking unit detects that the time the user's line of sight stays on a certain exhibit exceeds a preset threshold; a natural language processing unit, used to parse the voice input of the user's feedback on the first guide route into text form; a demand analysis unit, used to extract keywords in the text form of the user's feedback on the first guide route, and parse the user's actual exhibition needs based on the keywords.

[0016] Furthermore, the second guided route generation module includes: a route adjustment unit, used to adjust the first guided route according to the user's actual exhibition viewing needs to obtain a second guided route; an optimization unit, used to optimize the second guided route according to the crowd density to obtain the final guided route.

[0017] Furthermore, the interaction module further comprises a feedback response unit, and the feedback response unit is used to feed back the final guided route to the user's mobile device in the form of voice or image.

[0018] Furthermore, the real-time perception module is obtained through pre-training, and the pre-training steps include: collecting a first data set of users visiting exhibits in the exhibition hall and a second data set of crowd density images in different areas of the exhibition hall; marking the user's line of sight direction and stay time for the first data set, and marking the level of crowd density for the second data set; using the labeled first data set to train a convolutional neural network to obtain the line of sight tracking unit; using the labeled second data set to train a convolutional neural network to obtain the density detection unit.

[0019] Furthermore, the specific working process of the path planning unit is: receiving the user data and exhibition hall data transmitted by the data collection module; using the path planning algorithm, based on the spatial layout, taking the exhibits in the reservation information as the target point, generating multiple guided routes; screening out guided routes whose total distance is less than a preset threshold; calculating the scores of the screened guided routes based on the exhibit information and the exhibition viewing preferences, and setting the guided route with the highest score as the first guided route.

[0020] Furthermore, the second guided tour generating module further includes a real-time adjustment unit, which is used to obtain temporary demands of the user detected by the interaction module, and adjust the final guided tour in real time according to the temporary demands.

[0021] Compared with the prior art, the present invention can achieve at least one of the following beneficial effects:

[0022] The AIGC-based intelligent tour guide system provided by the present invention generates a first tour guide route according to the user's reservation information and exhibition viewing preferences through a first tour guide route generating module, interacts with the visitors through an interactive module, and after obtaining their actual exhibition viewing needs, adjusts the first tour guide route according to the actual exhibition viewing needs through a second tour guide route module, while taking into account the special needs of the visitors and the density of crowds to generate a final tour guide route. The system can fully understand the true intentions of the visitors and adjust the tour guide route in real time according to the visitors' feedback and the dynamic environment, thereby improving the accuracy and dynamic performance of the intelligent tour guide system, and thus providing visitors with a better exhibition viewing experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the embodiments of this specification or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this specification. For ordinary technicians in this field, other drawings can also be obtained based on these drawings.

[0024] Figure 1 A schematic diagram of the architecture of the AIGC-based intelligent tour guide system provided by the present invention;

[0025] Figure 2 This is a schematic diagram of the overall architecture of the AIGC-based intelligent navigation system provided by the present invention.

[0026] Reference numerals:

[0027] 1-data collection module; 11-user data collection unit; 12-exhibition hall data collection unit; 13-data storage unit; 2-real-time perception module; 21-line of sight tracking unit; 22-density detection unit; 3-first guided route generation module; 31-path planning unit; 4-interaction module; 41-first guided route unit; 42-voice broadcast unit; 43-natural language processing unit; 44-demand analysis unit; 45-feedback response unit; 5-second guided route generation module; 51-route adjustment unit; 52-optimization unit; 53-real-time adjustment unit. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. It should be noted that, in the absence of conflict, the embodiments and features in the embodiments disclosed in this disclosure can be combined, separated, interchanged and / or rearranged with each other. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0029] Example 1

[0030] A specific embodiment of the present invention, as Figure 1 As shown, an AIGC-based intelligent navigation system is disclosed, including:

[0031] Data collection module 1, used to obtain user data and exhibition hall data, the user data includes reservation information, exhibition preferences and special needs, and the exhibition hall data includes space layout and exhibit information;

[0032] Real-time perception module 2, used to perceive the user's sight direction, dwell time and crowd density in real time;

[0033] A first guided tour generating module 3, electrically connected to the data collecting module 1, and configured to generate a first guided tour according to user data and exhibition hall data;

[0034] An interactive module 4, electrically connected to the first guided tour generating module 3, the interactive module 4 is used to interact with the user and analyze the user's actual exhibition viewing needs according to the user's feedback on the first guided tour;

[0035] The second guided route generating module 5 is electrically connected to the data collecting module 1, the real-time sensing module 2, the first guided route generating module 3 and the interactive module 4. The second guided route generating module 5 is used to adjust the first guided route according to the special needs, the actual exhibition viewing needs and the crowd density to generate a final guided route.

[0036] Compared with the prior art, the AIGC-based intelligent tour guide system provided in this embodiment generates a first tour guide route according to the user's reservation information and exhibition viewing preferences through a first tour guide route generation module, interacts with visitors through an interaction module, and obtains their actual exhibition viewing needs. Then, the first tour guide route is adjusted according to the actual exhibition viewing needs through a second tour guide route module, while taking into account the special needs of visitors and the density of crowds to generate a final tour guide route. The system can fully understand the true intentions of visitors and adjust the tour guide route in real time according to the visitors' feedback and the dynamic environment, thereby improving the accuracy and dynamic performance of the intelligent tour guide system, and thus providing visitors with a better exhibition viewing experience.

[0037] In this embodiment, the data collection module 1 includes:

[0038] The user data collection unit 11 is used to obtain the reservation information and special needs entered by the user in the front-end interface, and analyze the user's reservation information and historical exhibition records to obtain the exhibition viewing preference;

[0039] The exhibition hall data collection unit 12 is used to obtain the spatial layout and exhibit information inside the exhibition hall;

[0040] The data storage unit 13 is used to store the data acquired by the user data acquisition unit 11 and the exhibition hall data acquisition unit 12 .

[0041] For example, users can enter reservation information such as name, age, ID number, contact number, facial information, exhibits to be visited, and special needs such as accessibility requirements and dining requirements through front-end interfaces such as WeChat mini-programs; the spatial layout of exhibition hall data includes but is not limited to the location and type labels of each exhibition area, and exhibit information includes but is not limited to the specific location of the exhibits.

[0042] In some embodiments, the exhibit information also includes introductory information and non-introductory information, where the introductory information is the language information that needs to be output when AIGC initiates the exhibit introduction, while the non-introductory information is all the exhibit-related information that is publicly available and retrievable on the Internet and is entered based on AIGC's powerful learning ability. The amount of information in this part may be thousands or even tens of thousands of times that of the "introductory information". Entering non-introductory information is not the purpose. The real purpose is to extract all feature tags related to the exhibits based on limited introductory information and unlimited non-introductory information, so as to form matching, combination and output of exhibit information for different feature tags in thousands of possible interactive scenarios, thereby achieving the ultimate navigation goal.

[0043] Take the "Qinggang Sword" as an example. The labels that can be given to the Qinggang Sword include "Zhao Yun", "Three Kingdoms", "Sword", "Cold Ice Weapon", "Famous Weapon", etc. There may be hundreds of them, and the interests of visitors will inevitably focus on this series of labels. When the interactive module interacts with the visitors, if the visitors continue to ask: "The Qinggang Sword is so powerful, then are there any other weapons in the Three Kingdoms period that can compete with it?" At this time, the interactive module will capture the label keywords "Three Kingdoms", "Weapons", "Comparable" in the conversation. Through these keywords, all data with relevant labels in the data storage unit are retrieved and matched, and finally the "intersection" is taken and the answer is given. For example, after searching, AIGC answers: "Of course, the weapons that can compete with the Qinggang Sword in the Three Kingdoms period include Guan Yu's Qinglong Yanyue Sword, Zhang Fei's Zhangba Snake Spear, Lu Bu's Fangtian Painted Halberd, etc. Among them, the Qinglong Yanyue Sword is also exhibited in this museum. Do you want to go and see it?

[0044] In this embodiment, the real-time perception module 2 includes:

[0045] The sight tracking unit 21 is used to capture the user's sight direction and the time the sight stays on a certain exhibit in real time;

[0046] The density detection unit 22 is used to detect the density of people flow in each exhibit area in the exhibition hall in real time through image recognition technology.

[0047] In this embodiment, the real-time perception module 2 is obtained through pre-training, and the pre-training steps include:

[0048] A first data set of users visiting exhibits in an exhibition hall and a second data set of crowd density images in different areas of the exhibition hall are collected; the user's line of sight direction and dwell time are marked on the first data set, and the level of crowd density is marked on the second data set; a convolutional neural network is trained using the marked first data set to obtain the line of sight tracking unit 21; a convolutional neural network is trained using the marked second data set to obtain the density detection unit.

[0049] Specifically, the gaze tracking unit includes a gaze tracking model and a dwell time prediction model, both of which need to be pre-trained;

[0050] The pre-training steps of the gaze tracking model include:

[0051] (I) Data collection: Use professional equipment, such as cameras with eye-tracking functions, to collect behavioral data on visitors’ gaze directions in front of different exhibits in the exhibition hall;

[0052] (II) Data annotation: annotating the user’s gaze points on the collected behavioral data;

[0053] (III) Model training: Use the labeled behavioral data to train a convolutional neural network to obtain a gaze tracking model that can predict the direction of visitors’ gaze in front of the exhibits;

[0054] The pre-training steps of the residence time prediction model include:

[0055] (I) Data collection: Record the gaze direction output by the gaze tracking model, i.e., the gaze point coordinate data, in chronological order to form a gaze direction data sequence;

[0056] (II) Model training: Use the gaze direction data sequence to train a time series analysis model, such as a long short-term memory network, to obtain a model that can predict how long a user's gaze stays on a certain exhibit.

[0057] It should be noted that the density detection unit needs to pre-train a convolutional neural network to obtain the crowd density detection model. The pre-training steps are similar to the pre-training of the tracking model and will not be repeated here.

[0058] In this embodiment, the first navigation route generating module 3 includes:

[0059] The route planning unit 31 is used to generate at least two guided routes according to the user's reservation information, and sort the priorities of the guided routes according to the exhibition viewing preferences, and set the guided route with the highest priority as the first guided route.

[0060] Specifically, the specific working process of the path planning unit 31 is:

[0061] Receiving the user data and exhibition hall data of the exhibition hall transmitted by the data collection module 1;

[0062] Using a path planning algorithm, based on the spatial layout and taking the exhibits in the reservation information as target points, a plurality of guided tour routes are generated;

[0063] Filter out the guided routes whose total distance is less than a preset threshold value; calculate the scores of the filtered guided routes according to the exhibit information and the exhibition viewing preference, and set the guided route with the highest score as the first guided route.

[0064] Specifically, the path planning unit 31 needs to be implemented through pre-training of a reinforcement learning network, such as a deep Q network, etc., which is not limited in the present invention. The training process can be achieved by setting a certain reward mechanism. For example, when a visitor reaches the exhibits in the reservation information by the shortest path, he can get a higher reward score, and the exhibits that cover the visitor's preference on the guided route can also get a certain reward score. Through multiple iterative learning, the guided route with the highest score is output as the first guided route.

[0065] Preferably, when generating multiple guided tours, the special needs of visitors must also be fully considered. For example, if a visitor raises accessibility requirements when making a reservation, when generating a guided tour with the exhibits in the reservation information as the target point, the guided tour should cover accessible elevators, ramps and mother-and-child rooms as much as possible. If the visitor has made a dining reservation, dining areas should be provided near the exhibits that the visitor visits during meal times as much as possible.

[0066] For example, if the visitor raises an accessibility requirement when making a reservation, the path planning unit should give the highest priority to whether the generated first guided route covers accessible facilities; if the visitor raises a dining requirement when making a reservation, then AIGC will, after 11:30, conduct an evaluation of "arrival at the restaurant + waiting time < 15 minutes" according to the following algorithm. When the condition that dining is available within 15 minutes is met, AIGC will proactively initiate a dining invitation to the visitor. If the visitor agrees to go to the restaurant, the interaction ends; if the visitor hopes to dine later, AIGC will start the evaluation of "arrival at the restaurant + waiting time < 15 minutes" 10 minutes in advance based on the expected dining time proposed by the visitor. When the condition that dining is available within 15 minutes is met, AIGC will proactively initiate a dining invitation to the visitor again.

[0067] The specific algorithm is as follows: AIGC forms a corresponding relationship with the current dining waiting time based on the number of people in the current queue for ordering food A, the number of meals prepared by the system B, and the number of available tables C. Based on the waiting time, it forms dining guide suggestions for visitors, as follows:

[0068] (I) When the value of A+BC is larger, it means that the current number of orders and food preparation is large, while the number of available tables is small. This situation is not conducive to dining and often requires longer waiting time;

[0069] (II) When the value of A+BC is smaller, it means that the current number of orders and food preparation is small, while the number of available tables is large. This situation is conducive to dining and there is often no need to wait;

[0070] (III) Therefore, according to the analysis in (I) and (II), a corresponding time value can be preset in the backend database module. Assuming A+BC=D, the estimated time value is as follows:

[0071] When D=-10, the waiting time = 0min, and when D increases by 1, the waiting time increases by 2min. For example, when D=10, the waiting time is 40min.

[0072] (IV) Because the waiting time in (III) is an estimate, this time value needs to be constantly revised through big data. Through the accumulation of a large amount of data, the estimated "waiting time" will eventually approach the actual waiting time. Eventually, the system will obtain the waiting time corresponding to various A, B, C, and D values, thereby forming an enumeration of waiting times corresponding to various dining situations.

[0073] Compared with the prior art, the intelligent guide system provided in this embodiment fully considers the special needs of visitors when visiting the exhibition when generating the guide route, which is more humane and further enhances the visitors' exhibition experience.

[0074] In this embodiment, the interaction module 4 includes:

[0075] A first guiding unit 41, used for guiding the movement of the visitor according to the first guiding route and the location of the visitor;

[0076] The voice announcement unit 42 is used to introduce the current exhibit to the user when the sight tracking unit 21 detects that the user's sight stays on a certain exhibit for more than a preset threshold;

[0077] A natural language processing unit 43, configured to parse the voice input of the user's feedback on the first guided route into text form;

[0078] The demand analysis unit 44 is used to extract keywords from the text form of the user's feedback on the first guided route, and analyze the user's actual exhibition viewing needs based on the keywords.

[0079] For example, when visitors arrive at the target exhibition hall, they can start to visit the exhibition. Video surveillance equipment including an eye tracking model will be installed in the display device of each exhibit, which will record the direction and duration of each visitor's sight in real time. When the eye tracking model detects that the visitor stays at a display device for more than 1 minute and the direction of sight is always looking at the exhibits there, AIGC will initiate a voice dialogue. The following is an example of the dialogue content:

[0080] AIGC: "I noticed that you seem to be interested in exhibit XXX. Do you need me to introduce it to you?" At this time, the visitor can give an affirmative or negative answer. If the answer is negative, the interaction ends; if the answer is affirmative, AIGC will start to introduce this exhibit or exhibit series; it should be noted here that the interactive dialogue between AIGC and the visitor is completely based on natural language, that is, it is the same as chatting between people; if the content of the dialogue is related to other exhibits in this exhibition hall, it means that the visitor may be interested in the related exhibits of this exhibit. If the related exhibits are in the current exhibition area, then AIGC will inform the visitor of the location of the exhibit, guide the visitor to watch it, and introduce the related exhibits to the visitor. If the related exhibit C is not in the current exhibition area, then AIGC will call the first guided route generation module to include the related exhibits into the target point and regenerate the first guided route.

[0081] Specifically, if visitors are dissatisfied with the generated first guided route, they can promptly reflect the dissatisfaction or real needs to the interactive module. After the interactive module receives the visitors' feedback, the natural language processing unit processes the feedback into a text form that can be understood by the computer, and then the demand analysis unit analyzes the visitors' actual viewing needs, and the subsequent corresponding adjustment module adjusts the first guided route.

[0082] In this embodiment, the second guided route generating module 5 includes:

[0083] A route adjustment unit 51, configured to adjust the first guided tour route according to the actual exhibition viewing needs of the user to obtain a second guided tour route;

[0084] The optimization unit 52 is used to optimize the second guided route according to the crowd density to obtain the final guided route.

[0085] In some embodiments, the second guided tour generating module 5 further includes a real-time adjusting unit 53 for acquiring temporary demands of the user detected by the interactive module 4 and adjusting the final guided tour in real time according to the temporary demands.

[0086] For example, if the visitor is not interested in some exhibits in the first guided tour route, the visitor can interact with the interactive module in time, and the interactive module analyzes the visitor's actual viewing needs and sends them to the route adjustment unit. The route adjustment unit removes the exhibits that the visitor is not interested in from the first guided tour route, and adjusts the first guided tour route with the remaining exhibits as target points to obtain a second guided tour route.

[0087] Taking into account that some exhibit areas have a high density of people and require a long queue time, which may cause discomfort to visitors, a scoring mechanism is built into the optimization unit to calculate the score of each exhibit, and the route with the highest score is selected as the final guided route. The specific scoring mechanism is as follows: score = arrival time score * 0.6 + crowd density score * 0.4, where the arrival time scoring rule can be: 1 point for all arrivals within 3 minutes, 0.1 point for each minute exceeding, no lower limit, and negative points are allowed; the crowd density scoring rule can be: 1 point for all 50% crowd intensity, 0.1 point for each 5% exceeding, and the minimum is 0 point. Both scoring rules can be adjusted according to actual needs.

[0088] Taking into account that visitors may have temporary needs due to emergencies during the exhibition, such as the need to rest or have a meal, it is necessary to handle the temporary needs of visitors in a timely manner. Visitors express their temporary needs by interacting with the interactive module, and then the real-time adjustment unit adjusts the final guided route in real time to meet the visitors' temporary needs such as rest or meal.

[0089] In some embodiments, if the visitor raises a need for a rest when making an appointment, the interactive module 4 sends a rest reminder to the visitor at a preset time according to the visitor's age segment, and the second guided route module adjusts the guided route in real time.

[0090] For example, different rest reminder times can be set for people of different age groups through the following rules:

[0091] (I) Children under 6 years old need their parents to help them make reservations, and they will be reminded every 0.5 hours;

[0092] (II) 6-30 years old, prompt every two hours;

[0093] (III) 30-40 years old, prompt every 1.5 hours;

[0094] (IV) 40-60 years old, prompt once every hour;

[0095] (V) For those over 60 years old, prompt every 0.5 hours.

[0096] In some embodiments, the interactive module collects the actual rest time of each age group in real time, and obtains the most reasonable rest reminder time through big data.

[0097] In this embodiment, the interaction module 4 further includes a feedback response unit 45, and the feedback response unit 45 is used to feed back the final guided route to the user's mobile device in the form of voice or image.

[0098] Specifically, when the visitor expresses his or her actual viewing needs to the interactive module, the route adjustment unit in the second guided route adjusts the first guided route according to the user's actual viewing needs to obtain the second guided route. The optimization unit optimizes the second guided route according to the crowd density. After obtaining the final guided route, the final guided route is fed back to the visitor's mobile devices such as mobile phones and headphones through the feedback response unit. If the visitor has temporary needs, the guided route adjusted in real time according to the temporary needs will be fed back to the visitor's mobile device.

[0099] Professionals should also be further aware that the operation of the module can be realized by the protocol or program of the prior art, without relying on the new computer program itself. The units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented with electronic hardware, computer software or a combination of the two. In order to clearly illustrate the interchangeability of hardware and software, the composition and steps of each example have been generally described in the above description according to the function. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to exceed the scope of the present invention.

[0100] The steps of the method or algorithm described in conjunction with the embodiments disclosed herein may be implemented using hardware, a software module executed by a processor, or a combination of the two. The software module may be placed in a random access memory (RAM), a memory, a read-only memory (ROM), an electrically programmable ROM, an electrically erasable programmable ROM, a register, a hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.

[0101] The specific implementation methods described above further illustrate the objectives, technical solutions and beneficial effects of the present invention in detail. It should be understood that the above description is only a specific implementation method of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc. made within the spirit and principles of the present invention should be included in the scope of protection of the present invention.

Claims

1. An intelligent tour guide system based on AIGC, characterized in that: include: A data collection module is used to obtain user data and exhibition hall data, wherein the user data includes reservation information, exhibition preferences and special needs, and the exhibition hall data includes space layout and exhibit information; Real-time perception module, used to perceive the user's sight direction, dwell time and crowd density in real time; A first guided tour generating module, electrically connected to the data collecting module, and configured to generate a first guided tour according to user data and exhibition hall data; The first guided route generation module includes a path planning unit, and the specific working process of the path planning unit is: receiving the user data and exhibition hall data transmitted by the data collection module; using the path planning algorithm, based on the spatial layout, taking the exhibits in the reservation information as the target point, generating multiple guided routes; screening out guided routes whose total distance is less than a preset threshold among the guided routes; calculating the scores of the screened guided routes according to the exhibit information and the exhibition viewing preferences, and setting the guided route with the highest score as the first guided route; An interaction module, electrically connected to the first guided tour generating module, the interaction module being used to interact with the user and analyze the user's actual exhibition viewing needs according to the user's feedback on the first guided tour; The interaction module includes: a first guide unit, used to guide the movement of the visitor according to the first guide route and the position of the visitor; a voice broadcast unit, used to introduce the current exhibit to the user when the sight tracking unit detects that the time the user's sight stays on a certain exhibit exceeds a preset threshold; a natural language processing unit, used to parse the voice input of the user's feedback on the first guide route into text form; a demand analysis unit, used to extract keywords in the text form of the user's feedback on the first guide route, and parse the user's actual viewing needs according to the keywords; a second guided route generation module, electrically connected to the data collection module, the real-time perception module, the first guided route generation module and the interaction module, the second guided route generation module being used to adjust the first guided route according to the special needs, the actual exhibition viewing needs and the crowd density to generate a final guided route; The second guided route generation module includes: a route adjustment unit, used to adjust the first guided route according to the user's actual exhibition viewing needs to obtain a second guided route; an optimization unit, used to optimize the second guided route according to the crowd density to obtain the final guided route.

2. The AIGC-based intelligent tour guide system according to claim 1, characterized in that: The data collection module includes: A user data collection unit is used to obtain the reservation information and special needs entered by the user in the front-end interface, and analyze the user's reservation information and historical exhibition records to obtain the exhibition viewing preference; The exhibition hall data collection unit is used to obtain the spatial layout and exhibit information inside the exhibition hall; The data storage unit is used to store the data acquired by the user data acquisition unit and the exhibition hall data acquisition unit.

3. The AIGC-based intelligent tour guide system according to claim 1, characterized in that: The real-time perception module includes: The eye tracking unit is used to capture the user's eye direction and the time the eye stays on a certain exhibit in real time; The density detection unit is used to detect the density of people flow in each exhibit area in the exhibition hall in real time through image recognition technology.

4. The AIGC-based intelligent tour guide system according to claim 3, characterized in that: The interaction module further includes a feedback response unit, which is used to feed back the final guided route to the user's mobile device in the form of voice or image.

5. The AIGC-based intelligent tour guide system according to claim 3, characterized in that: The real-time perception module is obtained through pre-training, and the pre-training steps include: Collecting a first data set of users visiting exhibits in the exhibition hall and a second data set of crowd density images in different areas of the exhibition hall; The user's sight direction and dwell time are marked for the first data set, and the level of crowd density is marked for the second data set; Using the annotated first data set to train a convolutional neural network to obtain the gaze tracking unit; The convolutional neural network is trained using the labeled second data set to obtain the density detection unit.

6. The AIGC-based intelligent tour guide system according to claim 3, characterized in that: The second guide route generation module further includes a real-time adjustment unit, which is used to obtain the temporary needs of the user detected by the interaction module, and adjust the final guide route in real time according to the temporary needs.

Images