Personalized AI guide system and method for scenic spot

Abstract

The invention provides a scenic spot personalized AI guide system and method, and the method comprises the steps: collecting a movement factor, a displacement scale and an interaction mark of a tourist in a sightseeing process, and constructing a tourist behavior feature vector; generating a corresponding rhythm score based on the tourist behavior feature vector in combination with continuity constraint based on a historical displacement scale, and mapping the rhythm score into a discrete touring rhythm state; querying a preset strategy template set according to the rhythm score and the tour rhythm state, and generating a corresponding tour guide strategy instruction; and based on the guide strategy instruction, selecting a matched guide content version from a local content library through a preset mapping rule, and outputting the matched guide content version. According to the invention, a landing, configurable and extensible intelligent navigation control scheme is constructed, and a technical path which better accords with actual operation conditions is provided for personalized navigation service of scenic spots.

Description

Technical Field

[0001] This invention belongs to the field of personalized AI tour guides, and in particular relates to a personalized AI tour guide system and method for scenic spots. Background Technology

[0002] In existing scenic area guided tour services, the main methods still rely on human guides, fixed-content audio guides, or location-triggered mobile applications. These methods generally suffer from rigid responses, monotonous rhythms, and insufficient adaptability in practice. Human guides are constrained by human resources and time constraints, making it difficult to cover large numbers of tourists and provide continuous service. Traditional audio guides often use preset sequences or timed playback, failing to adjust the timing and depth of explanations based on actual tourist behavior. While location-based guided tour applications improve the convenience of information access, they typically use location as the trigger condition, ignoring the different tour states tourists may exhibit at the same location, such as quickly passing by, briefly stopping, or lingering, resulting in a mismatch between the guided tour content and the actual needs of tourists. Meanwhile, real scenic areas often have complex spatial structures, numerous obstructions, and dense crowds, making location data prone to short-term fluctuations, further amplifying the instability issues of guided tours driven by a single trigger condition. With the increasing diversity of tourists, the differences in information reception pace, focus, and tour style among different tourists are becoming more pronounced. Existing tour guide systems lack a comprehensive technical solution that can stably understand the current tour status of tourists without relying on complex sensing devices, and control the output of tour guide content accordingly.

[0003] Therefore, how to effectively match the pace of the tour with the state of tourists by utilizing the behavioral information available on the terminal under real scenic conditions remains a key problem that existing technologies have not yet solved. Summary of the Invention

[0004] The purpose of this invention is to propose a personalized AI-guided tour system and method for scenic spots to solve the above-mentioned problems.

[0005] To achieve the above objectives, a personalized AI-guided tour method for scenic spots is provided in a first aspect of the present invention, the method comprising the following steps: The system collects tourists' movement factors, displacement scales, and interaction markers during their visit to construct a tourist behavior feature vector. The movement factors represent the intensity of tourists' movement, the displacement scales represent the spatial variation of tourists, and the interaction markers represent whether tourists engage in active interaction. Based on the tourist behavior feature vector, combined with the continuity constraint based on historical displacement scale, a corresponding rhythm score is generated and mapped to a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of tourist attention in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy. Based on the rhythm score and tour rhythm status, a preset set of strategy templates is queried to generate corresponding tour guide strategy instructions; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factor; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status; Based on the navigation strategy instructions, a matching navigation content version is selected from the local content library and output according to preset mapping rules.

[0006] Furthermore, the displacement scale is calculated based on the current position coordinates obtained in two consecutive time windows and the spatial distance between the two position points; The movement factor is calculated based on the displacement scale, the corresponding time window, and a pre-set reference value. The interaction flag is generated by the event recording mechanism within the tour guide application. When a tourist performs any tour guide-related operation within the current time window, the interaction flag is assigned a valid value; if no tour guide interaction occurs, the interaction flag remains at its initial value.

[0007] Furthermore, the guided tour-related operations include clicking on an entry point for a specific attraction or triggering an audio guide.

[0008] Furthermore, the rhythm score is obtained by weighted summation of the movement factor term, displacement scale term, interaction flag term, and a continuity constraint for suppressing displacement abrupt changes, followed by compression using a logical function.

[0009] Furthermore, the rhythm score is mapped to discrete tour rhythm states, specifically: The system uses a tiered threshold to determine the rhythm state based on predefined rhythm state labels. When the interaction flag is actively interacted with, the threshold will be lowered accordingly, making the same rhythm score more likely to be judged as the state of outputting navigation content.

[0010] Furthermore, the movement factor item is calculated based on the movement factor, the displacement scale item is calculated based on the displacement scale, and the interaction flag item is calculated and generated based on the interaction flag.

[0011] Furthermore, the step of querying a preset set of strategy templates based on the rhythm score and tour rhythm status to generate corresponding tour guide strategy instructions specifically involves: Based on the corresponding tour rhythm state, determine the set of available tour guide strategy types; The strategy strength factor is calculated based on the offset of the rhythm score relative to the preset baseline threshold of the current tour rhythm state. The strategy strength factor is calculated by subtracting the baseline threshold of the current rhythm state from the rhythm score, dividing by the preset adjustment range width, and limiting the result to between 0 and 1.

[0012] The tour rhythm state is combined with the strategy strength factor to generate tour guide strategy instructions.

[0013] Furthermore, the step of selecting and outputting a matching version of the guide content from the local content library based on the guide strategy instruction and using preset mapping rules specifically involves: Based on the tour rhythm state in the tour guide strategy instructions, filter the set of applicable tour guide content versions in the content library; Based on the strategy intensity factor in the guide strategy instruction, select a specific content level from the guide content version set; Load the resources of the selected content level and output them in voice, text, or multimedia format, depending on the terminal's capabilities.

[0014] Furthermore, the step of selecting a specific content level from the set of guide content versions specifically involves: Multiply the strategy strength factor by the total number of available content levels under the corresponding tour rhythm state minus one, and then round down to obtain the corresponding content level index.

[0015] A second aspect of the present invention provides a personalized AI-guided tour system for scenic spots, the system comprising: The real-time tourist behavior feature construction module is used to collect tourists' movement factors, displacement scales, and interaction flags during the tour to construct tourist behavior feature vectors. The movement factors represent the intensity level of tourists' movement, the displacement scales represent the spatial change range of tourists, and the interaction flags represent whether tourists have engaged in active interaction behavior. The tour rhythm state construction module is used to generate a corresponding rhythm score based on the tourist behavior feature vector and combined with the continuity constraint based on the historical displacement scale, and map it into a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of attention of tourists in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy. The tour guide strategy instruction construction module is used to query a preset set of strategy templates and generate corresponding tour guide strategy instructions based on the rhythm score and tour rhythm status; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factors; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status; The navigation strategy instruction output module is used to select and output the matching navigation content version from the local content library based on the navigation strategy instruction and a preset mapping rule.

[0016] The beneficial technical effects of the present invention are at least as follows: This invention proposes an intelligent tour guide system and method for real-world scenic area applications. By structurally modeling real-time tourist behavior and introducing the intermediate control semantic of tour rhythm, it transforms continuous and variable behavioral characteristics into stable and controllable tour guide decision-making criteria, thereby achieving dynamic matching between the tour guide content output method and the tourist's current state. This invention does not simply rely on location or a single triggering condition, but rather uses a joint analysis of tourist movement intensity, spatial changes, and proactive interaction behavior to determine the tour rhythm. Based on this, it generates explicit tour guide strategy instructions to control whether the tour is triggered, the triggering content level, and the output intensity. By decoupling the rhythm state from the strategy intensity and managing it configurably, the system can maintain the stability and consistency of tour guide output under complex environments and behavioral fluctuations, while also subtly reflecting changes in tourist attention levels. Furthermore, this invention clearly separates strategy decision-making from content execution, creating a clear control chain in the engineering implementation of the tour guide system, avoiding redundant judgments and unnecessary interference outputs. This improves the adaptability and experience continuity of the tour guide service without increasing hardware dependence or system complexity. Overall, this invention addresses the core issues of "when to speak, how much to speak, and how to speak" during scenic area visits, and constructs a practical, configurable, and scalable intelligent tour guide control solution, providing a technical path that is more in line with actual operating conditions for personalized tour guide services in scenic areas. Attached Figure Description

[0017] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0018] Figure 1 This is a flowchart of a personalized AI-guided tour method for scenic spots according to the present invention. Detailed Implementation

[0019] Embodiments of the present invention are described in detail below. Examples of these embodiments are shown in the accompanying drawings, wherein the same or similar reference numerals denote the same or similar elements or elements having the same or similar functions throughout. The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as limiting the present invention.

[0020] like Figure 1As shown in the figure, an embodiment of the present invention provides a personalized AI-guided tour method for scenic spots, the method comprising: S1. Collect tourists' movement factors, displacement scales, and interaction markers during the tour to construct a tourist behavior feature vector; wherein, the movement factors are used to represent the tourists' movement intensity level, the displacement scales are used to represent the tourists' spatial change range, and the interaction markers are used to indicate whether the tourists have engaged in active interaction behavior.

[0021] Specifically, the data used in this step all comes from the tour guide terminal devices carried by tourists. This data can be directly obtained from regular smartphones or wearable devices through the system interface. It includes the location coordinates continuously output by the terminal's positioning module, data from the terminal's internal motion sensing module reflecting changes in device displacement, and user interaction events recorded by the tour guide application itself. The system samples the above data at fixed time windows on the terminal side, with each time window corresponding to a specific moment. This is used to construct the behavioral characteristics at that moment.

[0022] Furthermore, in the specific implementation process, the system first processes the location data based on the fundamental behavioral characteristic of "whether the tourist has moved spatially." The terminal operates within two consecutive time windows. and The system obtains the current location coordinates and calculates the spatial distance between these two locations to determine the tourist's displacement scale within the given time window. : ; in, and Indicates the terminal is in the current time window The coordinate values ​​obtained by the positioning module. and This represents the corresponding coordinate value obtained within the previous time window. The preset time window length for the system. This value is used to depict the spatial variation of tourists within adjacent time windows. When tourists are standing still and observing, this value remains at a low level, while when tourists walk along the scenic route, this value continues to increase.

[0023] Furthermore, after obtaining the displacement scale, the system calculates the movement intensity of tourists by combining it with the time window length and performs uniform scale processing. To avoid excessive differences in numerical distribution due to different spatial scales of scenic areas or individual walking habits, the system introduces a reference quantity to normalize the movement intensity, constructing a movement factor. : ; in, Indicates that within the current time window, by and The calculated movement intensity These are reference values ​​pre-set during the system deployment phase based on the spatial characteristics of the scenic area, used to uniformly map the movement behaviors of different tourists. Through this process, the movement behaviors of different tourists under different spatial conditions can be compressed into a consistent numerical range, making it convenient for the subsequent state determination module to use directly.

[0024] Furthermore, in addition to spatial and mobility features, this step also introduces an interactive flag to describe proactive attention behavior. This indicator is generated by the event logging mechanism within the tour guide application. It appears when a visitor performs any tour-related action within the current time window (such as clicking on an attraction's information portal or triggering an audio guide). It is assigned a valid value; if no guided tour interaction occurs, then Retain the initial value. This flag does not distinguish between specific interaction types; it only reflects whether the visitor is actively paying attention to the guided tour content at the current moment.

[0025] Furthermore, after completing the above processing, the system will move the factor. Displacement scale and interactive logos Combined in a fixed order, the final tourist behavior feature vector is formed: ; in, For a moment The corresponding behavioral feature vector has three components: the tourist's movement intensity level, spatial variation range, and whether or not active interaction behavior occurs.

[0026] in, For a moment The corresponding behavioral feature vector has three components: the tourist's movement intensity level, spatial variation range, and whether or not active interaction behavior occurs.

[0027] S2. Based on the tourist behavior feature vector and combined with the continuity constraint based on the historical displacement scale, a corresponding rhythm score is generated and mapped to a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of tourist attention in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy.

[0028] Specifically, this step will use the tourist behavior feature vector obtained in the previous step. Transition to rhythm state This allows the navigation control to shift from continuous numerical values ​​to a few semantically clear states.

[0029] In scenic areas, typical tourist behaviors include "briefly pausing to take photos," "minor positional shifts caused by crowd movement," and "clicking on the narration while walking." If instantaneous values ​​are used directly to determine the state, frequent state changes can easily occur, leading to unstable narration timing. This step focuses on two points: "stable determination" and "interaction priority." On the one hand, a temporal continuity constraint is introduced into the rhythm scoring to suppress false triggers caused by minor positional shifts. On the other hand, interaction flags are used to apply a slight bias at the scoring threshold, enabling the act of actively selecting / activating the narration to be quickly identified as a state more suitable for outputting a guided tour.

[0030] Furthermore, the input for this step is the tourist behavior feature vector output from the previous step. .in Based on the normalization results of the movement intensity from the previous step, Distance calculation results from adjacent time window locations, Interaction events recorded by the navigation app within the current time window.

[0031] In addition, to describe "state continuity", the system retains the displacement scale calculated in the previous time window at the terminal side. This value is temporarily stored in the local runtime context after the terminal completes the displacement calculation in the previous time window, and is directly read and used in the current time window.

[0032] Furthermore, the system first calculates the rhythm score. This rating is used to... The three components are compressed into a controllable scalar that is more robust to scenic area noise. The score consists of three parts: the movement intensity term (from...) Spatial variation term (from) ), interactive enhancement items (from A continuity constraint term oriented towards the micro-drift of scenic area positioning is introduced to suppress unexpected abrupt changes in the displacement scale of adjacent windows. The rhythm score is calculated as follows: ; in, For logical functions This is used to compress the score into a stable, finite range, which facilitates subsequent threshold division; , , , The parameters configured on the terminal side during deployment are written into the application configuration file or local parameter table, and the scenic area operator sets them all at once based on experience such as route density and indoor-outdoor ratio. , , Taken from the input vector ; Cached by the terminal in the previous time window It is obtained by reading its displacement components; Corresponding to the "amplitude of abrupt changes in displacement scale," this item will significantly increase when a tourist takes a photo in place but experiences a brief jump in positioning, thereby reducing... This prevents the state from being affected by jitter. For example, when a tourist stops at a viewing platform and takes photos continuously, the actual behavior corresponds to low movement and low displacement. However, in crowded or obstructed conditions, the positioning point may experience slight drift, causing... If a certain window suddenly rises, the continuity constraint will pull this sudden change back to a scoring range that is more consistent with the dwelling behavior, thereby maintaining state stability.

[0033] Furthermore, to obtain The system then maps it to discrete rhythm states. The mapping uses an "interaction-first" threshold bias approach: when This means that when a visitor actively triggers a guided tour interaction within the current window, the threshold will generate a slight bias, making the same... It is more likely to be judged as a state where guided tour content can be output, in order to match the typical operating habits of tourists in scenic areas who "click to open the guide while walking" and "immediately activate the guide after stopping." The state judgment rules are as follows: ; in, , The two thresholds configured during deployment are written to the terminal's local parameter table; The interaction bias strength parameter is also configured during deployment. From When tourists click on the guided tour entry for a specific attraction or activate the audio guide within the window, To obtain a valid value, the threshold is shifted downwards accordingly, making To get into a state more suitable for guided tour output more quickly. , , Predefined rhythm state labels can correspond to semantics such as "low participation / walk and see", "normal tour / moderate explanation can be given" and "stop and pay attention / explanation can be strengthened". The specific semantics are fixed in the system configuration and correspond one by one in the subsequent strategy library.

[0034] S3. Based on the rhythm score and tour rhythm status, query the preset strategy template set and generate corresponding tour guide strategy instructions; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factor; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status.

[0035] With the rhythm already clearly defined, this step completes the most crucial "action decision" stage of the tour guide system, which involves applying the tour rhythm obtained in the previous stage. With rhythm score This is converted into a single, directly executable navigation strategy command on the terminal. In actual scenic area operation, the reproducibility of the navigation strategy is not reflected in algorithmic complexity, but rather in "what the system will do under what conditions." Therefore, this step focuses on the clarity and configurability of the strategy formation path, enabling engineers to clearly define each judgment and output action.

[0036] Specifically, the input for this step is the rhythm state output from the previous step. and rhythm score . The tour rhythm state is used to define the basic scope of the tour guiding strategy; A rhythm score is assigned to reflect subtle differences in the level of tourist attention in the current state. These two quantities are passed as a set of state parameters to the strategy decision module on the terminal side, and the system has fully cached them in the current time window context before proceeding to this step.

[0037] In its implementation, the system first determines the rhythm state. The system locates the corresponding set of strategy templates. This set is stored locally on the terminal as a configuration table, for example, using a key-value structure. The key is the rhythm state label, and the value is the set of tour guide strategy types allowed to be triggered in that state. Each strategy type corresponds to a specific system behavior, such as "do not trigger tour guide," "trigger brief explanation," or "trigger full explanation." In this way, the system constrains the strategy space at the state level, preventing tour guide behavior from being triggered in unsuitable rhythm states.

[0038] Furthermore, after determining the available strategy types, the system utilizes rhythm scoring. The intensity of strategy execution is continuously adjusted. To ensure consistency in the scale of intensity adjustment across different rhythm states, the system provides each... Pre-configure a benchmark scoring threshold And an adjustment range width These two parameters are set based on scenic area visit experience during the system deployment phase and written as constants into the terminal configuration file. During operation, the system uses the following formula to... Mapped to policy strength factor : ; in, The rhythm score output from the previous step; This indicates the minimum score required to trigger guided tour behavior under the current rhythm state; This represents the score range from the lowest trigger level to the highest execution intensity. Through this calculation, Constrained within a stable range, this approach reflects differences in ratings without causing sudden shifts in strategy strength due to short-term fluctuations. For example, under the same "pause and attention" state, when a visitor's stay is brief... Approaching the threshold, A lower value corresponds to a shorter prompt; this applies when visitors linger and repeatedly view the guide content. Increase Approaching the upper limit, a more complete explanation version is provided.

[0039] Furthermore, after obtaining the strategy strength factor, the system will determine the rhythm state. With intensity factor Combine to generate the final navigation strategy instructions. This instruction is represented in a structured format, explicitly containing two fields: strategy category and execution intensity. Its generation method is as follows: ; in, This indicates the basic state category of the current tour guide behavior. This indicates the strength of the navigation execution in this state. This structured instruction does not directly trigger content playback; instead, it is passed as a control parameter to the content invocation module in the next step, which then determines the appropriate function. Select the matching tutorial version from the content library and complete the terminal output.

[0040] S4. Based on the navigation strategy instructions, select the matching navigation content version from the local content library through preset mapping rules and output it.

[0041] Specifically, assuming the tour guidance strategy is already defined, this step completes the most practical application step in the entire system: translating the tour guidance strategy instructions into a real-world tour guidance content output action occurring on the visitor's terminal. The input for this step is the tour guidance strategy instruction output in the previous stage. When performing this step, the system first... The analysis yields the rhythmic states contained within. With strategy strength factor ,in Indicates the currently allowed navigation semantic categories. This indicates the relative strength of the navigation output within that semantic category. In the specific implementation, the system first determines the relative strength of the navigation output based on... rhythm state The local guided tour content library is filtered. The library was structured during system deployment, with each attraction's guided tour content split into multiple versions, and their appropriate pace clearly indicated in the configuration file. For example, the same attraction might have both a brief, informative version for quick visits and a detailed version for more focused viewing. During system runtime, only a limited number of versions are needed. Using this as an index, the set of content that can be output under the current rhythm state can be located in the content library.

[0042] Furthermore, after obtaining the candidate content set, the system further considers the strategy strength factor. Select a specific content level. To do this, the system assigns a sequential level number to each candidate content version in the content library, representing its relative position in terms of information density. The total number of level numbers... The hierarchy varies depending on the rhythm state; for example, in some rhythm states only two levels of content are allowed, while in others more granular levels are permitted. At runtime, the system uses the following rules to... Mapped to specific content level indexes : ; in, From navigation strategy instructions This is used to reflect the intensity of the current navigation output; In rhythm state The number of available content levels, a value provided by the content library configuration file; This is the index of the final selected content level. Using this mapping method, when… When the threshold is low, the system selects the version with less information; when... When the threshold is high, the system selects the version with more complete information, thus achieving a smooth change in the guide content based on the visitor's level of attention without increasing decision-making complexity. After determining the specific content instance, the system loads the content into the terminal output module. The loading process includes reading the corresponding text, audio, or multimedia resources from local storage and selecting an appropriate presentation path based on the terminal's capabilities. For example, when the terminal supports voice playback, the system sends the text content to the speech synthesis module to generate and play the speech; when the terminal only supports graphic display, it is presented in text or graphic form. This selection process is based on the terminal capability identifier and is unrelated to visitor behavior or strategy calculation. During the actual content output process, the system also utilizes the strategy strength factor. The output rhythm can be parameterized and controlled, for example, by adjusting the speed of voice playback or the switching speed of text and image displays, so that the guided tour presentation is consistent with the current tour rhythm of the tourists.

[0043] This invention also provides a personalized AI-guided tour system for scenic spots, the system comprising: The real-time tourist behavior feature construction module is used to collect tourists' movement factors, displacement scales, and interaction flags during the tour to construct tourist behavior feature vectors. The movement factors represent the intensity level of tourists' movement, the displacement scales represent the spatial change range of tourists, and the interaction flags represent whether tourists have engaged in active interaction behavior. The tour rhythm state construction module is used to generate a corresponding rhythm score based on the tourist behavior feature vector and combined with the continuity constraint based on the historical displacement scale, and map it into a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of attention of tourists in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy. The tour guide strategy instruction construction module is used to query a preset set of strategy templates and generate corresponding tour guide strategy instructions based on the rhythm score and tour rhythm status; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factors; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status; The navigation strategy instruction output module is used to select and output the matching navigation content version from the local content library based on the navigation strategy instruction and a preset mapping rule.

[0044] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0045] In the embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection of apparatuses or units may be electrical, mechanical, or other forms.

[0046] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0047] Although embodiments of the invention have been shown and described, those skilled in the art will understand that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the claims and their equivalents.

Claims

1. A personalized AI-guided tour method for scenic spots, characterized in that, The method includes: The system collects tourists' movement factors, displacement scales, and interaction markers during their visit to construct a tourist behavior feature vector. The movement factors represent the intensity of tourists' movement, the displacement scales represent the spatial variation of tourists, and the interaction markers represent whether tourists engage in active interaction. Based on the tourist behavior feature vector, combined with the continuity constraint based on historical displacement scale, a corresponding rhythm score is generated and mapped to a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of tourist attention in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy. Based on the rhythm score and tour rhythm status, a preset set of strategy templates is queried to generate corresponding tour guide strategy instructions; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factor; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status; Based on the navigation strategy instructions, a matching navigation content version is selected from the local content library and output according to preset mapping rules.

2. The personalized AI-guided tour method for scenic spots according to claim 1, characterized in that, The displacement scale is calculated based on the current position coordinates obtained in two consecutive time windows and the spatial distance between the two position points. The movement factor is calculated based on the displacement scale, the corresponding time window, and a pre-set reference value. The interaction flag is generated by the event recording mechanism within the tour guide application. When a tourist performs any tour guide-related operation within the current time window, the interaction flag is assigned a valid value; if no tour guide interaction occurs, the interaction flag remains at its initial value.

3. The personalized AI-guided tour method for scenic spots according to claim 2, characterized in that, The guided tour-related operations include clicking on the entry point for a specific attraction or triggering an audio guide.

4. The personalized AI-guided tour method for scenic spots according to claim 1, characterized in that, The rhythm score is obtained by weighted summation of the movement factor, displacement scale, interaction flag, and a continuity constraint to suppress abrupt displacement changes, followed by compression using a logical function.

5. The personalized AI-guided tour method for scenic spots according to claim 4, characterized in that, The rhythm score is mapped to discrete tour rhythm states, specifically: The system uses a tiered threshold to determine the rhythm state based on predefined rhythm state labels. When the interaction flag is actively interacted with, the threshold will be lowered accordingly, making the same rhythm score more likely to be judged as the state of outputting navigation content.

6. The personalized AI-guided tour method for scenic spots according to claim 4, characterized in that, The movement factor item is calculated based on the movement factor, the displacement scale item is calculated based on the displacement scale, and the interaction flag item is generated based on the interaction flag.

7. The personalized AI-guided tour method for scenic spots according to claim 1, characterized in that, The step involves querying a preset set of strategy templates based on the rhythm score and tour rhythm status to generate corresponding tour guide strategy instructions, specifically: Based on the corresponding tour rhythm state, determine the set of available tour guide strategy types; The strategy strength factor is calculated based on the offset of the rhythm score relative to the preset baseline threshold of the current tour rhythm state. The strategy strength factor is calculated as follows: after subtracting the baseline threshold of the current rhythm state from the rhythm score, divide by the preset adjustment range width, and limit the result to between 0 and 1. The tour rhythm state is combined with the strategy strength factor to generate tour guide strategy instructions.

8. The personalized AI-guided tour method for scenic spots according to claim 1, characterized in that, The step of selecting and outputting a matching version of the guide content from the local content library based on the guide strategy instructions and using preset mapping rules is as follows: Based on the tour rhythm state in the tour guide strategy instructions, filter the set of applicable tour guide content versions in the content library; Based on the strategy intensity factor in the guide strategy instruction, select a specific content level from the guide content version set; Load the resources of the selected content level and output them in voice, text, or multimedia format, depending on the terminal's capabilities.

9. A personalized AI-guided tour method for scenic spots according to claim 8, characterized in that, The specific step of selecting a content level from the set of guide content versions is as follows: Multiply the strategy strength factor by the total number of available content levels under the corresponding tour rhythm state minus one, and then round down to obtain the corresponding content level index.

10. A personalized AI-guided tour system for scenic spots, characterized in that, The system includes: The real-time tourist behavior feature construction module is used to collect tourists' movement factors, displacement scales, and interaction flags during the tour to construct tourist behavior feature vectors. The movement factors represent the intensity level of tourists' movement, the displacement scales represent the spatial change range of tourists, and the interaction flags represent whether tourists have engaged in active interaction behavior. The tour rhythm state construction module is used to generate a corresponding rhythm score based on the tourist behavior feature vector and combined with the continuity constraint based on the historical displacement scale, and map it into a discrete tour rhythm state; the rhythm score is used to reflect the subtle differences in the degree of attention of tourists in the current state; the tour rhythm state is used to limit the basic scope of the tour guide strategy. The tour guide strategy instruction construction module is used to query a preset set of strategy templates and generate corresponding tour guide strategy instructions based on the rhythm score and tour rhythm status; wherein, the tour guide strategy instructions include tour rhythm status and strategy intensity factors; the strategy intensity factor group is used to represent the strength of tour guide execution under the corresponding tour rhythm status; The navigation strategy instruction output module is used to select and output the matching navigation content version from the local content library based on the navigation strategy instruction and a preset mapping rule.

Images