Tourism information management method based on service-oriented collaborative robot

Abstract

The invention discloses a tourism information management method based on a service-oriented collaborative robot, and relates to the technical field of federated learning, and the method comprises the steps: binding a tourist identity with a service-oriented collaborative robot sandbox, and generating a tourist mirror image set; inputting the tourist mirror image set into a geo-fence detector, comparing an electronic grid map by combining real-time geographic coordinates, and outputting a buffer zone operation signal and a border crossing signal through two channels of physical isolation; triggering the service-oriented collaborative robot to generate a privacy gradient parameter set based on the buffer zone operation signal; the federal encryption box receives the privacy gradient parameter set, converts the privacy gradient parameter set into a gear coding sequence and outputs the gear coding sequence; and if the geo-fence detector outputs a border crossing signal, physical fusing is triggered to destroy the tourist mirror image set. According to the method, model training iteration is performed on the initial service label in the local privacy sandbox of the service-oriented collaborative robot, so that personalized service parameter dynamic generation based on tourist identity characteristics is realized.

Description

Technical Field

[0001] The present invention relates to the field of federated learning technology, and in particular to a tourism information management method based on a service-type collaborative robot. Background Art

[0002] In the application scenarios of deep integration of service robots and intelligent information management, data privacy protection and dynamic service response mechanisms in a multi-agent collaborative environment are facing systemic challenges. With the widespread penetration of distributed computing architecture in the public service industry, how to achieve accurate scheduling and trusted interaction of cross-domain service resources while ensuring that user identity information cannot be tracked has become a core issue for improving the level of intelligent services. To this end, building a collaborative management mechanism with physical isolation capabilities, support for dynamic policy generation, and compliance with privacy compliance constraints has become a key technical path to achieve autonomous and trusted decision-making of service robots.

[0003] Existing tourism information management systems often rely on location-based services (LBS) architectures that interact with central servers. Mobile terminals report coordinates to enable attraction recommendations or route planning. This often relies on users continuously authorizing location data, posing privacy risks and lacking effective support for service robots' local decision-making capabilities. To mitigate privacy concerns, some solutions incorporate differential privacy or anonymization techniques, but these can lead to reduced service accuracy or increased response latency, making it difficult to balance security and efficiency. Summary of the Invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a tourism information management method based on a service-type collaborative robot to solve the problem of privacy exposure risks.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a tourism information management method based on a service-type collaborative robot, which comprises: Bind the visitor identity to the service collaborative robot sandbox to generate a visitor image set; The visitor mirror set is input into the geofence detector, which combines the real-time geographic coordinates with the electronic grid map to output the buffer zone operation signal and the boundary crossing signal through physically isolated dual channels. The buffer zone operation signal triggers the service collaborative robot to generate a privacy gradient parameter set. The federated encryption box receives the privacy gradient parameter set and converts it into a gear coding sequence output; if the geo-fence detector outputs an out-of-bounds signal, it triggers a physical fuse to destroy the visitor mirror set; The gear coding sequence is input into the federation aggregation core, and the federation global parameters are generated through weighted aggregation; after the tourist service center receives the federation global parameters, it parses and processes them to generate strategy execution instructions.

[0007] As a preferred solution of the tourism information management method based on the service collaborative robot of the present invention, the tourist identity is bound to the service collaborative robot sandbox as follows: Scan the visitor's passport information page using the optical character recognition component built into the service collaborative robot; Extract the tourist's name, nationality, passport number, and date of birth from the passport information page; The service-oriented collaborative robot activates the privacy sandbox and performs a one-way hash process on the extracted visitor's name, nationality, passport number, and date of birth to generate a unique identification code; The generated unique identity code is written into the isolated storage area of ​​the privacy sandbox to establish a binding relationship between the visitor's identity and the service collaborative robot sandbox.

[0008] As a preferred solution of the tourism information management method based on the service-type collaborative robot of the present invention, wherein: generating a tourist mirror set is as follows: Call the preset service preference template, load the default language, common payment methods and basic navigation type according to the extracted nationality, and form the initial service label; The initial service tag and the unique identification code are associated and stored in the isolated storage area of ​​the privacy sandbox to generate a visitor image set.

[0009] As a preferred solution of the tourism information management method based on the service-type collaborative robot described in the present invention, the tourist mirror set is input into the geo-fence detector, combined with the real-time geographic coordinate comparison electronic grid map, and the buffer zone operation signal and the boundary crossing signal are output through the physically isolated dual channels, specifically as follows: The geo-fence detector compares the associated real-time geographic coordinates with the coordinate range of the national boundary line in the electronic grid map pre-stored in the geo-fence detector; If the real-time geographic coordinates are within five kilometers outside the coordinate range of the national boundary line, a buffer zone operation signal is generated; If the real-time geographic coordinates fall into the area on the opposite side of the national boundary coordinate range, a cross-border signal is generated; The buffer zone operation signal and the out-of-bounds signal are output through physically isolated dual channels.

[0010] As a preferred solution of the tourism information management method based on the service collaborative robot of the present invention, wherein: the buffer zone operation signal triggers the service collaborative robot to generate a privacy gradient parameter set, which is specifically as follows: The service collaborative robot receives the buffer zone operation signal from the geofence detector and activates the service preference model in the local privacy sandbox; Input the initial service labels in the tourist image set into the service preference model and perform local training iterations; After training is completed, the service preference model parameters are stored in the local privacy sandbox built into the service collaborative robot; The service collaborative robot reads the service preference model parameters before and after local training iterations from the local privacy sandbox, extracts the updated content of the language preference layer weights, payment method preference layer weights, and tour type preference layer weights, and generates a privacy gradient parameter set.

[0011] As a preferred solution of the tourism information management method based on the service-type collaborative robot of the present invention, the federated encryption box receives the privacy gradient parameter set and converts it into a gear coding sequence output, as follows: The federated encryption box receives the privacy gradient parameter set from the service collaborative robot; Arrange the language preference layer weights, payment method preference layer weights, and tour type preference layer weights in the privacy gradient parameter set in the order of their respective preference layer weights to form a gradient parameter sequence; According to the value of each preference layer weight in the gradient parameter sequence, the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight are mapped to the preset scale interval of the mechanical encoding disk; The federal encryption box drives the mechanical encoding disk to rotate to the corresponding scale position, and outputs the corresponding gear code sequence through the physical gear set of the mechanical encoding disk.

[0012] As a preferred solution of the tourism information management method based on the service-type collaborative robot of the present invention, if the geo-fence detector outputs an out-of-bounds signal, a physical fuse is triggered to destroy the tourist mirror set, as follows: Transmitting the out-of-bounds signal to the storage center of the visitor's mirror set through a physically isolated dedicated channel; When the storage center of the guest mirror set receives the out-of-bounds signal, it immediately activates the built-in physical fuse; The physical fuse performs a mechanical deformation action, destroying the magnetic recording layer in the storage center of the visitor mirror set, causing the visitor mirror set to be irreversibly destroyed.

[0013] As a preferred solution of the tourism information management method based on the service collaborative robot of the present invention, the gear code sequence is input into the federation aggregation core, and the federation global parameters are generated by weighted aggregation, as follows: The federated aggregation core obtains the gradient parameter sequences from multiple service collaborative robots and determines the weighting coefficient according to the frequency of tourist services provided by each service collaborative robot. The updated weight values ​​of the language preference layer, the payment method preference layer, and the tour type preference layer are cumulatively calculated according to the weighted coefficients to generate updated layer parameters; The updated layer parameters of each preference layer are merged to form the federated global parameters.

[0014] As a preferred solution of the tourism information management method based on service-type collaborative robots of the present invention, after receiving the federated global parameters, the tourist service center parses and processes them to generate strategy execution instructions, as follows: The Tourist Service Center receives the federation global parameters and reads the weight values ​​of the language preference layer, payment method preference layer, and tour type preference layer in the federation global parameters through the internal parsing structure; Match the weight values ​​of each preference layer with the preset judgment condition table to determine the corresponding service policy configuration items; The service policy configuration items are converted into specific policy execution instructions and sent to the control end of the service collaborative robot.

[0015] As a preferred solution of the tourism information management method based on the service-type collaborative robot described in the present invention, the service policy configuration item is a specific service setting defined by the service-type collaborative robot, including configuration options for language preference, payment method preference and tour type preference, guiding the service-type collaborative robot to perform corresponding service operations.

[0016] The beneficial effects of the present invention are as follows: by performing model training iterations on the initial service tag in the local privacy sandbox of the service collaborative robot, dynamic generation of personalized service parameters based on the identity characteristics of the visitor is achieved. The pre-trained service preference model is used to adjust the weights of the default language, common payment methods, and basic tour types in the initial service tag to complete the update of the service preference model parameters; after the service preference model is trained locally, the weight update content between the service preference model parameters before the local training iteration and the service preference model parameters after the local training iteration is extracted to generate a privacy gradient parameter set, ensuring that the original identity information of the visitor and the initial service tag do not leave the privacy sandbox, ensuring data privacy and security. At the same time, it supports subsequent federated aggregation to perform weighted aggregation of privacy gradient parameter sets from multiple service collaborative robots, thereby improving the personalization, security, and collaborative efficiency of service strategy generation in tourism information management. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0018] Figure 1 Flowchart of the tourism information management method based on service collaborative robots.

[0019] Figure 2 Flowchart of geo-fencing and dual-channel branching.

[0020] Figure 3 Flowchart of the service preference model and privacy gradient parameter set.

[0021] Figure 4 Flowchart of the parsing and policy execution instructions for the Visitor Service Center. DETAILED DESCRIPTION

[0022] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0023] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0024] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0025] Reference Figures 1 to 4 , is an embodiment of the present invention, which provides a tourism information management method based on a service-type collaborative robot, comprising the following steps: S1. Bind the visitor identity to the service collaborative robot sandbox to generate a visitor image set.

[0026] S1.1. Scan the visitor's passport information page using the optical character recognition component built into the collaborative service robot to extract the visitor's name, nationality, passport number, and date of birth from the passport information page; The collaborative service robot activates a local privacy sandbox and performs a one-way hash on the extracted visitor's name, nationality, passport number, and date of birth to generate a unique identification code. The local privacy sandbox is an isolated data processing space within the collaborative service robot that is used to securely store and process visitor data. It is based on existing privacy protection technologies (such as sandbox isolation). Specifically: In the privacy sandbox, start the SHA-256 hash function and perform one-way hash processing on the extracted tourist name, nationality, passport number and date of birth: First, concatenate the tourist name, nationality, passport number and date of birth into a string in order (such as the field arrangement order of the passport information page), for example, concatenate the tourist name, nationality, passport number and date of birth in sequence, and use the separator "|" in the middle to form a complete input string; then, input the complete input string into the SHA-256 hash function, and the SHA-256 hash function performs a one-way conversion on the input string to generate a fixed-length 256-bit hash value; the 256-bit hash value is used as a unique identity code and stored in the isolated storage area of ​​the local privacy sandbox for subsequent binding of the tourist identity to the service collaborative robot sandbox.

[0027] S1.2. Write the generated unique identification code into the isolated storage area of ​​the privacy sandbox to establish a binding relationship between the visitor's identity and the service collaborative robot sandbox; Specifically, first, a unique record entry is created in the isolated storage area to store the unique identity code, and the record entry uses the unique identity code as the primary key; then, the unique identity code is associated with the internal identifier of the service collaborative robot sandbox. The internal identifier is a fixed string of the service collaborative robot sandbox and is used to identify the current service collaborative robot sandbox entity; then, a key-value pair is added to the record entry in the isolated storage area to map the unique identity code to the internal identifier of the service collaborative robot sandbox to form a binding relationship; finally, the record entry is saved to the isolated storage area to ensure that the association between the unique identity code and the internal identifier of the service collaborative robot sandbox is persistent, which is used for the subsequent generation of visitor mirror sets; Call the preset service preference template, load the default language, common payment methods and basic navigation type according to the extracted nationality, and form the initial service label; Specifically: first, read the nationality information extracted from the passport information page from the isolated storage area; then, access the service preference template pre-installed in the service collaborative robot sandbox. The service preference template is a preset mapping table based on statistical data on nationality and common tourism service needs, which includes the correspondence between nationality and default language, common payment methods and basic tour types. For example, the nationality of "China" corresponds to the default language of "Chinese", the common payment method of "mobile payment" and the basic tour type of "historical and cultural tour"; according to the nationality information, search for the corresponding default language, common payment method and basic tour type in the service preference template and extract matching entries; then, combine the extracted default language, common payment method and basic tour type into a structured data set, named the initial service tag.

[0028] S1.3. Associate the initial service tag with the generated unique identification code in the isolated storage area of ​​the privacy sandbox to generate a visitor image set.

[0029] S2. Input the visitor mirror set into the geo-fence detector, combine it with the real-time geographic coordinate comparison electronic grid map, and output the buffer zone operation signal and the boundary crossing signal through the physically isolated dual channels; based on the buffer zone operation signal, the service collaborative robot is triggered to generate a privacy gradient parameter set.

[0030] S2.1. The service collaborative robot activates the built-in positioning component to collect real-time geographic coordinates, which include latitude and longitude data.

[0031] S2.2. The geo-fence detector receives the visitor mirror set and the real-time geographic coordinates from the service collaborative robot, and establishes a timestamp association between the real-time geographic coordinates and the unique identification code in the visitor mirror set; The geofence detector compares the associated real-time geographic coordinates with the coordinate range of the national boundary line in an electronic grid map pre-stored in the geofence detector; wherein the electronic grid map is a digitized map pre-stored in the geofence detector and is used to store the coordinate range of the national boundary line. The geofence detector determines the area where the real-time geographic coordinates are located by comparing the real-time geographic coordinates with the coordinate range of the national boundary line. If the real-time geographic coordinates are within five kilometers outside the coordinate range of the national boundary line, a buffer zone operation signal is generated; If the real-time geographic coordinates fall into the area on the opposite side of the national border coordinate range (i.e., the geographical area on the other side of the border line, for example, if the border line is a national border, the area on the opposite side of the national border coordinate range is the geographical range of the neighboring country or region on the other side of the border line), a border crossing signal is generated; The buffer zone operation signal and the out-of-bounds signal are output through physically isolated dual channels respectively; Specifically: First, the geofence detector confirms the generated buffer zone operation signal and cross-border signal; then, the geofence detector transmits the buffer zone operation signal to the physically isolated first channel. The first channel is an independent signal transmission line, which is only used to transmit the buffer zone operation signal to the service collaborative robot to trigger subsequent operations; then, the geofence detector transmits the cross-border signal to the physically isolated second channel. The second channel is another independent signal transmission line, which is only used to transmit the cross-border signal to the visitor mirror set storage center to trigger the physical fuse; finally, the buffer zone operation signal is transmitted to the service collaborative robot through the first channel, and the cross-border signal is transmitted to the visitor mirror set storage center through the second channel, ensuring that the output processes of each signal (buffer zone operation signal and cross-border signal) are isolated and do not interfere with each other; It should be noted that by comparing the real-time geographic coordinates with the coordinate range of the national boundary line in the electronic grid map, the buffer zone operation signal and the cross-border signal are generated and output respectively through the physically isolated dual channels; compared with the lack of privacy protection and dynamic response capabilities in location monitoring in the existing technology, this step uses the physically isolated dual channels to output the buffer zone operation signal and the cross-border signal to ensure that the signal transmission is isolated and does not interfere with each other, triggering the local training of the service collaborative robot or the destruction of the visitor mirror set, thereby improving data security and real-time performance, and solving the defects of location data leakage and delayed response.

[0032] S2.3, the geo-fence detector forwards the initial service tag in the visitor mirror set to the service collaborative robot; The service collaborative robot receives the buffer zone operation signal from the geofence detector and activates the service preference model in the local privacy sandbox.

[0033] S2.4. Input the initial service labels in the tourist image set into the service preference model and perform one local training iteration. Specifically, the initial service tag in the visitor image set is read from the isolated storage area. The initial service tag contains preference data for the default language, common payment methods, and basic tour types. Then, the initial service tag is used as input data for the service preference model, which is a pre-trained neural network consisting of a language preference layer, a payment method preference layer, and a tour type preference layer. Next, the service preference model adjusts the weights of the language preference layer, the payment method preference layer, and the tour type preference layer based on the preference data of the initial service tag, completing a weight update process to form the updated service preference model parameters. Finally, the updated service preference model parameters are stored in the local privacy sandbox for subsequent generation of the privacy gradient parameter set. Service preference model training process: The service preference model is a pre-trained neural network, and the initial training process is completed before the service collaborative robot is deployed. The training data is derived from historical tourism service data, which contains the language preferences, payment method preferences, and tour type preferences of tourists of different nationalities. The weights of the language preference layer, payment method preference layer, and tour type preference layer are adjusted multiple times using the gradient descent method: First, the tourist service preference dataset from the historical tourism service data is input into the service preference model. The tourist service preference dataset contains the default language, common payment method, and basic tour type data of tourists of different nationalities. Then, the service preference model adjusts the weights of the language preference layer, payment method preference layer, and tour type preference layer layer by layer based on the tourist service preference dataset. Each adjustment gradually updates the weight value by comparing the difference between the predicted output and the actual preference data. Then, this adjustment process is repeated multiple times until the weight value tends to stabilize, forming the initial service preference model parameters. After training is completed, the service preference model parameters are stored in the local privacy sandbox built into the service collaborative robot, supporting subsequent local training iterations based on the initial service labels; The collaborative service robot reads the service preference model parameters before and after local training iterations from the local privacy sandbox, extracts the updated language preference layer weights, payment method preference layer weights, and tour type preference layer weights, generates a privacy gradient parameter set, and transmits it to the built-in federated encryption box of the collaborative service robot. Specifically: First, the service preference model parameters before and after the local training iteration are read from the isolated storage area of ​​the local privacy sandbox. The service preference model parameters include the values ​​of the language preference layer weight, the payment method preference layer weight, and the tour type preference layer weight; then, the service preference model parameters before and after the local training iteration are compared, and the change values ​​of the language preference layer weight before and after the iteration, the change values ​​of the payment method preference layer weight before and after the iteration, and the change values ​​of the tour type preference layer weight before and after the iteration are extracted to form a weight update content set; then, the language preference layer weight change values, the payment method preference layer weight change values, and the tour type preference layer weight change values ​​in the weight update content set are arranged in the order of the language preference layer, the payment method preference layer, and the tour type preference layer to generate a privacy gradient parameter set; finally, the privacy gradient parameter set is transmitted to the federated encryption box built into the service collaborative robot through the internal data interface built into the service collaborative robot for subsequent processing.

[0034] S2.5. It should be noted that this step uses machine learning technology to perform local training iterations of the service preference model to generate a privacy gradient parameter set. Compared with the existing technology, the application of machine learning in tourism information management often faces the shortcomings of data privacy leakage, delayed personalized service response, and lack of dynamic scene adaptation. This step is innovative in combining geo-fences with federated learning mechanisms to protect tourist data security through a local privacy sandbox environment and avoid original data transmission. It uses real-time location monitoring of geo-fences to dynamically trigger local training to ensure that the service preference model is quickly adjusted according to the real-time location of tourists. It generates a privacy gradient parameter set through federated learning to achieve multi-robot collaborative optimization of service strategies. It solves the defects of existing technologies such as easy leakage of tourist data, inability of personalized services to adapt to dynamic scenes in real time, and low efficiency of cross-device collaboration, and improves the privacy protection capability, real-time response capability and collaboration efficiency of tourism information management.

[0035] S3. The federated encryption box receives the privacy gradient parameter set and converts it into a gear coding sequence output; if the geo-fence detector outputs an out-of-bounds signal, a physical fuse is triggered to destroy the visitor mirror set.

[0036] S3.1. The federated encryption box receives the privacy gradient parameter set from the service collaborative robot; Arrange the language preference layer weights, payment method preference layer weights, and tour type preference layer weights in the privacy gradient parameter set in the order of their respective preference layer weights to form a gradient parameter sequence; According to the value of each preference layer weight in the gradient parameter sequence, the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight are mapped to the preset scale interval of the mechanical encoding disk; Specifically: first, read the language preference layer weight, payment method preference layer weight and navigation type preference layer weight in the gradient parameter sequence, and the weight value is in the form of a floating point number; then, compare the values ​​of the language preference layer weight, payment method preference layer weight and navigation type preference layer weight with the preset scale interval of the mechanical encoding disk. The preset scale interval of the mechanical encoding disk is based on the weight value range of the language preference layer, payment method preference layer and navigation type preference layer, and is a fixed value range list. For example, the language preference layer weight corresponds to the scale interval 0 to 100, the payment method preference layer weight corresponds to the scale interval 101 to 200, and the navigation type preference layer weight corresponds to the scale interval 101 to 200. The scale interval is 201 to 300. Then, according to the size of the weight values ​​of each preference layer, the scale interval corresponding to the language preference layer weight value is determined to be 0 to 100, the scale interval corresponding to the payment method preference layer weight value is determined to be 101 to 200, and the scale interval corresponding to the navigation type preference layer weight value is determined to be 201 to 300. For example, if the language preference layer weight value falls within the range of 0 to 100, it is mapped to a specific scale position within the scale interval of 0 to 100 corresponding to the language preference layer weight value. Finally, the language preference layer weight, payment method preference layer weight, and navigation type preference layer weight are respectively mapped to the corresponding scale intervals of the mechanical encoder disk, completing the mapping process. The Federal Encryption Box drives the mechanical encoding disk to rotate to the corresponding scale position, and outputs the corresponding gear code sequence through the physical gear set of the mechanical encoding disk; Specifically, first, the federal encryption box receives the scale interval position of the mechanical encoding disk after the language preference layer weight, the payment method preference layer weight and the navigation type preference layer weight are mapped. For example, the language preference layer weight is mapped to a specific position within the scale interval of 0 to 100, the payment method preference layer weight is mapped to a specific position within the scale interval of 101 to 200, and the navigation type preference layer weight is mapped to a specific position within the scale interval of 201 to 300; then, the federal encryption box starts the internal motor and drives the mechanical encoding disk to rotate to the corresponding scale positions according to the scale positions of the language preference layer weight, the payment method preference layer weight and the navigation type preference layer weight. The mechanical encoder disk's rotation is then transmitted through a physical gear set, which consists of multiple gears that are linked in sequence according to a preset gear tooth ratio, converting the rotation into a sequence of gear rotations. Finally, the rotation sequence of the physical gear set is recorded as a gear code sequence, which represents the combination of scale positions in binary form for subsequent transmission to the central aggregation node. The gear tooth ratio is set based on the range of the mechanical encoder disk's scale interval. For example, the language preference layer weight corresponds to a gear tooth ratio of 1:1, the payment method preference layer weight is 1:2, and the navigation type preference layer weight is 1:3. Through the dedicated communication interface of the federal encryption box, the gear coding sequence is transmitted to the central aggregation node via a shielded cable in the form of a differential signal; differential signaling is a communication method that transmits opposite electrical signals through two signal lines to reduce interference; shielded cable is a cable wrapped with a metal shielding layer to prevent external electromagnetic interference from affecting signal transmission.

[0037] S3.2. The geo-fence detector continuously monitors the real-time geographic coordinates of the visitor and generates a border crossing signal if the real-time geographic coordinates fall into the area on the opposite side of the national boundary coordinate range; Transmitting the out-of-bounds signal to the storage center of the visitor's mirror set through a physically isolated dedicated channel; When the storage center of the guest mirror set receives the out-of-bounds signal, it immediately activates the built-in physical fuse; The physical fuse performs a mechanical deformation action, destroying the magnetic recording layer in the storage center of the visitor mirror set, causing the visitor mirror set to be irreversibly destroyed; Specifically: first, the physical fuse receives an out-of-bounds signal from the geo-fence detector, triggering the activation of an internal mechanical switch; then, the mechanical switch activates the spring-driven mechanism in the physical fuse, which releases the pre-stored mechanical force to push a set of metal cutting blades to move; then, the metal cutting blades act directly on the magnetic recording layer of the visitor mirror set storage center. The magnetic recording layer is the physical medium for storing visitor mirror sets, such as the magnetic coating of a disk; the metal cutting blades destroy the structure of the magnetic recording layer through physical cutting or squeezing actions, making the stored visitor mirror set data completely unreadable or unrecoverable; finally, after the physical fuse completes its action, the magnetic recording layer is permanently damaged, ensuring that the visitor mirror set is irreversibly destroyed.

[0038] S4. Input the gear coding sequence into the federation aggregation core and generate federation global parameters through weighted aggregation. After receiving the federation global parameters, the tourist service center parses and processes them to generate strategy execution instructions.

[0039] S4.1. The central aggregation node receives the gear code sequence transmitted via the shielded cable; The photoelectric sensor array in the central aggregation node is used to analyze the tooth position arrangement state of the gear coding sequence and restore it to the corresponding gradient parameter sequence; Specifically, first, the central aggregation node receives a gear code sequence transmitted via a shielded cable. The gear code sequence represents the scale position combination of the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight in binary form. Then, the photoelectric sensor array scans the tooth position arrangement state of the gear code sequence. The photoelectric sensor array is composed of multiple photoelectric sensors and can identify the high and low tooth position states of each binary bit in the gear code sequence. For example, a high tooth position represents 1 and a low tooth position represents 0. Next, the photoelectric sensor array converts the scanned tooth position arrangement state into a corresponding binary sequence, restoring the original numerical form of the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight. Finally, the restored language preference layer weight, payment method preference layer weight, and navigation type preference layer weight are arranged in the order of language preference layer weight, payment method preference layer weight, and navigation type preference layer weight to form a gradient parameter sequence for subsequent allocation to the update fields of the language preference layer, payment method preference layer, and navigation type preference layer.

[0040] S4.2. Allocate the gradient parameter sequence to the update fields of the language preference layer, the payment method preference layer, and the navigation type preference layer according to the preset rules; Specifically, first, a gradient parameter sequence is obtained from the central aggregation node. The gradient parameter sequence includes the values ​​of the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight, and the arrangement order is language preference layer weight, payment method preference layer weight, and navigation type preference layer weight; then, the allocation rule table preset in the central aggregation node is accessed. The allocation rule table is set based on the mapping relationship between the arrangement order of the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight in the gradient parameter sequence and the corresponding update fields. The allocation rule table defines the corresponding relationship between each weight in the gradient parameter sequence and the language preference layer, payment method preference layer, and navigation type preference layer update fields. For example, the language preference layer weight corresponds to the language preference layer update field, the payment method preference layer weight corresponds to the payment method preference layer update field, and the navigation type preference layer weight corresponds to the navigation type preference layer update field; then, according to the allocation rule table, the language preference layer weight value in the gradient parameter sequence is stored in the language preference layer update field, the payment method preference layer weight value is stored in the payment method preference layer update field, and the navigation type preference layer weight value is stored in the navigation type preference layer update field; finally, the storage of the language preference layer update field, the payment method preference layer update field, and the navigation type preference layer update field is completed to form an update field set for subsequent processing by the federated aggregation core.

[0041] S4.3, the central aggregation node submits the gradient parameter sequences from multiple service collaborative robots to the internal federated aggregation core; The federated aggregation core obtains the gradient parameter sequences from multiple service collaborative robots and determines the weighting coefficient according to the frequency of tourist services provided by each service collaborative robot. Specifically, first, the federated aggregation core receives the gradient parameter sequences sent by multiple service collaborative robots submitted by the central aggregation node; then, the tourist service frequency data of each service collaborative robot is extracted from the service record database in the central aggregation node. The tourist service frequency data records the number of times each service collaborative robot handles tourist requests within a specific time period. For example, a service collaborative robot handles 100 tourist requests; then, the tourist service frequency data of each service collaborative robot is compared with the preset frequency-coefficient mapping table. The frequency-coefficient mapping table is based on the relationship between tourist service frequency and service collaborative robot service frequency. The statistical relationship of the robot's contribution is set, and the frequency-coefficient mapping table defines the correspondence between the frequency of tourist service and the weighted coefficient. For example, 100 requests correspond to a weighted coefficient of 0.5, and 200 requests correspond to a weighted coefficient of 0.8. Then, according to the frequency-coefficient mapping table, the gradient parameter sequence of each service collaborative robot is assigned a corresponding weighted coefficient. For example, a service collaborative robot that processes 100 requests is assigned a weighted coefficient of 0.5. Finally, the gradient parameter sequence of each service collaborative robot is associated with the corresponding weighted coefficient and stored to form a weighted coefficient allocation set, which is used to subsequently generate updated layer parameters.

[0042] S4.4. The weight update values ​​of the same preference layer (language preference layer, payment method preference layer, and tour type preference layer) are cumulatively calculated according to the weighted coefficients to generate the updated layer parameters, which are expressed as follows: ; Where, is a layer parameter, representing the final updated weight value of the language preference layer, payment method preference layer, or navigation type preference layer. is the total number of service collaborative robots participating in the federated aggregation core processing. For example, 10 service collaborative robots =10, is the index of the service collaborative robot, which is used to identify one of the multiple service collaborative robots participating in the federated aggregation core processing, for example, =1 indicates the first service collaborative robot, =2 indicates the second service collaborative robot, For the The weighted coefficient of each service collaborative robot is obtained from the frequency-coefficient mapping table based on the frequency of tourist services. For example, 100 requests correspond to a weighted coefficient of 0.5. For the The weight update value of the corresponding preference layer in the gradient parameter sequence of the service collaborative robot, for example, the weight update value of the language preference layer; The updated layer parameters of each preference layer are merged to form the federated global parameters and transmitted to the tourist service center.

[0043] S4.5. The Visitor Service Center receives the federation global parameters and reads the weight values ​​of the language preference layer, payment method preference layer, and tour type preference layer in the federation global parameters through an internal parsing structure. Specifically, first, the tourist service center receives the federated global parameters from the central aggregation node. The federated global parameters include the updated layer parameters of the language preference layer, the updated layer parameters of the payment method preference layer, and the updated layer parameters of the tour type preference layer. Then, the tourist service center activates the internal parsing structure, which is a predefined data processing program for parsing the structured data of the federated global parameters. Next, the parsing structure extracts the language preference layer weight value, the payment method preference layer weight value, and the tour type preference layer weight value from the federated global parameters in the order of the language preference layer, the payment method preference layer weight value, and the tour type preference layer weight value. For example, the language preference layer weight value indicates the language preference priority, the payment method preference layer weight value indicates the payment method priority, and the tour type preference layer weight value indicates the tour type priority. Finally, the extracted language preference layer weight value, payment method preference layer weight value, and tour type preference layer weight value are stored in the local storage area of ​​the tourist service center. The predefined data processing program is written based on the structured data format and parsing logic of the federated global parameters, and its specific implementation depends on the programming language or scripting language selected by the developer. Match the weight values ​​of each preference layer with the preset judgment condition table to determine the corresponding service policy configuration items; Specifically: first, read the language preference layer weight value, payment method preference layer weight value and tour guide type preference layer weight value from the local storage area of ​​the tourist service center; then, access the preset judgment condition table in the tourist service center. The judgment condition table is set based on the statistical relationship between the range of the language preference layer weight value, payment method preference layer weight value and tour guide type preference layer weight value and the priority of tourism service demand. The judgment condition table is a predefined mapping table, which contains the correspondence between the weight value range and the service strategy configuration item. For example, the language preference layer weight value in the range of 0 to 50 corresponds to the "English tour guide" configuration item, and the range of 51 to 100 corresponds to the "Chinese tour guide" configuration item; the payment method preference layer weight value in the range of 0 to 50 corresponds to the "credit card payment" configuration item, and the range of 51 to 100 corresponds to the "credit card payment" configuration item. The value within the range corresponds to the "mobile payment" configuration item; the weight value of the tour type preference layer in the range of 0 to 50 corresponds to the "historical and cultural tour" configuration item, and the value within the range of 51 to 100 corresponds to the "natural scenery tour" configuration item; then, the language preference layer weight value, the payment method preference layer weight value and the tour type preference layer weight value are compared with the corresponding ranges in the judgment condition table respectively to determine the range to which each weight value belongs. For example, the language preference layer weight value of 45 corresponds to the "English tour" configuration item; finally, according to the comparison results, the service policy configuration items corresponding to the language preference layer weight value, the service policy configuration items corresponding to the payment method preference layer weight value and the service policy configuration items corresponding to the tour type preference layer weight value are extracted to form a service policy configuration item set for subsequent generation of policy execution instructions.

[0044] S4.6. Convert the service policy configuration items into specific policy execution instructions and send them to the control terminal of the service collaborative robot; Specifically: First, read the service policy configuration item set from the local storage area of ​​the tourist service center. The service policy configuration item set includes the service policy configuration item corresponding to the language preference layer weight value, the service policy configuration item corresponding to the payment method preference layer weight value, and the service policy configuration item corresponding to the tour type preference layer weight value; then, access the preset instruction conversion table in the tourist service center. The instruction conversion table is set based on the mapping relationship between the service policy configuration items and the executable operations of the service-type collaborative robot. The instruction conversion table is a predefined mapping table that contains the correspondence between the service policy configuration items and the policy execution instructions. For example, the "English tour" configuration item corresponds to the instruction "Set voice output to English", and the "Mobile payment" configuration item corresponds to the instruction "Prioritize activation". Use the mobile payment interface, and the "historical and cultural tour" configuration item corresponds to the instruction "load historical and cultural tour content"; then, according to the instruction conversion table, the service policy configuration item corresponding to the language preference layer weight value in the service policy configuration item set is converted into the corresponding language policy execution instruction, the service policy configuration item corresponding to the payment method preference layer weight value is converted into the corresponding payment policy execution instruction, and the service policy configuration item corresponding to the tour type preference layer weight value is converted into the corresponding tour policy execution instruction, forming a policy execution instruction set; finally, through the communication interface of the tourist service center, the policy execution instruction set is transmitted to the control end of the service collaborative robot in the form of an encrypted data packet to guide the service collaborative robot to perform corresponding service operations.

[0045] This embodiment also provides a computer device, which is suitable for the tourism information management method based on a service-type collaborative robot, including: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute computer-executable instructions to implement the tourism information management method based on a service-type collaborative robot proposed in the above embodiment.

[0046] The computer device may be a terminal, comprising a processor, memory, a communication interface, a display, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores an operating system and computer programs. The internal memory provides an environment for the operating system and computer programs stored in the non-volatile storage media. The communication interface of the computer device is used to communicate with external terminals via wired or wireless communication. Wireless communication may be achieved via Wi-Fi, a carrier network, NFC (near-field communication), or other technologies. The display of the computer device may be a liquid crystal display or an electronic ink display. The input device may be a touchscreen overlay on the display, buttons, a trackball, or a touchpad on the computer device housing, or an external keyboard, touchpad, or mouse.

[0047] This embodiment also provides a storage medium having a computer program stored thereon, which, when executed by a processor, implements the tourism information management method based on a service-type collaborative robot as proposed in the above embodiment; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, disk or optical disk.

[0048] In summary, the present invention achieves dynamic generation of personalized service parameters based on tourist identity characteristics by: performing model training iterations on the initial service tag in the local privacy sandbox of the service collaborative robot. A pre-trained service preference model is used to adjust the weights of the default language, common payment methods, and basic tour type in the initial service tag to update the service preference model parameters. After the service preference model is trained locally, the weight update content between the service preference model parameters before the local training iteration and the service preference model parameters after the local training iteration is extracted to generate a privacy gradient parameter set. This ensures that the original identity information of the tourist and the initial service tag do not leave the privacy sandbox, thereby protecting data privacy and security. The invention also supports subsequent federated aggregation to perform weighted aggregation of privacy gradient parameter sets from multiple service collaborative robots, thereby improving the personalization, security, and collaborative efficiency of service strategy generation in tourism information management.

[0049] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A tourism information management method based on a service-type collaborative robot, characterized by: include, Bind the visitor identity to the service collaborative robot sandbox to generate a visitor image set; The visitor mirror set is input into the geofence detector, which combines the real-time geographic coordinates with the electronic grid map to output the buffer zone operation signal and the boundary crossing signal through physically isolated dual channels. The buffer zone operation signal triggers the service collaborative robot to generate a privacy gradient parameter set. The federated encryption box receives the privacy gradient parameter set and converts it into a gear coding sequence output; if the geo-fence detector outputs an out-of-bounds signal, it triggers a physical fuse to destroy the visitor mirror set; The gear coding sequence is input into the federation aggregation core, and the federation global parameters are generated through weighted aggregation; after the tourist service center receives the federation global parameters, it parses and processes them to generate strategy execution instructions.

2. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: The visitor identity is bound to the service collaborative robot sandbox as follows: Scan the visitor's passport information page using the optical character recognition component built into the service collaborative robot; Extract the tourist's name, nationality, passport number, and date of birth from the passport information page; The service-oriented collaborative robot activates the privacy sandbox and performs a one-way hash process on the extracted visitor's name, nationality, passport number, and date of birth to generate a unique identification code; The generated unique identity code is written into the isolated storage area of ​​the privacy sandbox to establish a binding relationship between the visitor's identity and the service collaborative robot sandbox.

3. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: Generate a guest mirror set, as follows: Call the preset service preference template, load the default language, common payment methods and basic navigation type according to the extracted nationality, and form the initial service label; The initial service tag and the unique identification code are associated and stored in the isolated storage area of ​​the privacy sandbox to generate a visitor image set.

4. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: The tourist mirror set is input into the geo-fence detector, combined with the real-time geographic coordinate comparison electronic grid map, and the buffer zone operation signal and the boundary crossing signal are output through the physically isolated dual channels, as follows: The geo-fence detector compares the associated real-time geographic coordinates with the coordinate range of the national boundary line in the electronic grid map pre-stored in the geo-fence detector; If the real-time geographic coordinates are within five kilometers outside the coordinate range of the national boundary line, a buffer zone operation signal is generated; If the real-time geographic coordinates fall into the area on the opposite side of the national boundary coordinate range, a cross-border signal is generated; The buffer zone operation signal and the out-of-bounds signal are output through physically isolated dual channels.

5. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: The buffer zone operation signal is used to trigger the service collaborative robot to generate a privacy gradient parameter set, specifically as follows: The service collaborative robot receives the buffer zone operation signal from the geofence detector and activates the service preference model in the local privacy sandbox; Input the initial service labels in the tourist image set into the service preference model and perform local training iterations; After training is completed, the service preference model parameters are stored in the local privacy sandbox built into the service collaborative robot; The service collaborative robot reads the service preference model parameters before and after local training iterations from the local privacy sandbox, extracts the updated content of the language preference layer weights, payment method preference layer weights, and tour type preference layer weights, and generates a privacy gradient parameter set.

6. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: The federated encryption box receives the privacy gradient parameter set and converts it into a gear coding sequence output, as follows: The federated encryption box receives the privacy gradient parameter set from the service collaborative robot; Arrange the language preference layer weights, payment method preference layer weights, and tour type preference layer weights in the privacy gradient parameter set in the order of their respective preference layer weights to form a gradient parameter sequence; According to the value of each preference layer weight in the gradient parameter sequence, the language preference layer weight, the payment method preference layer weight, and the navigation type preference layer weight are mapped to the preset scale interval of the mechanical encoding disk; The federal encryption box drives the mechanical encoding disk to rotate to the corresponding scale position, and outputs the corresponding gear code sequence through the physical gear set of the mechanical encoding disk.

7. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: If the geo-fence detector outputs an out-of-bounds signal, a physical fuse is triggered to destroy the visitor mirror set, as follows: Transmitting the out-of-bounds signal to the storage center of the visitor's mirror set through a physically isolated dedicated channel; When the storage center of the guest mirror set receives the out-of-bounds signal, it immediately activates the built-in physical fuse; The physical fuse performs a mechanical deformation action, destroying the magnetic recording layer in the storage center of the visitor mirror set, causing the visitor mirror set to be irreversibly destroyed.

8. The tourism information management method based on a service-type collaborative robot according to claim 1, wherein: The gear coding sequence is input into the federated aggregation core, and the federated global parameters are generated by weighted aggregation, as follows: The federated aggregation core obtains the gradient parameter sequences from multiple service collaborative robots and determines the weighting coefficient according to the frequency of tourist services provided by each service collaborative robot. The updated weight values ​​of the language preference layer, the payment method preference layer, and the tour type preference layer are cumulatively calculated according to the weighted coefficients to generate updated layer parameters; The updated layer parameters of each preference layer are merged to form the federated global parameters.

9. The tourism information management method based on a service-type collaborative robot according to claim 1, characterized in that: After receiving the federation global parameters, the tourist service center parses and processes to generate policy execution instructions, as follows: The Tourist Service Center receives the federation global parameters and reads the weight values ​​of the language preference layer, payment method preference layer, and tour type preference layer in the federation global parameters through the internal parsing structure; Match the weight values ​​of each preference layer with the preset judgment condition table to determine the corresponding service policy configuration items; The service policy configuration items are converted into specific policy execution instructions and sent to the control end of the service collaborative robot.

10. The tourism information management method based on a service-type collaborative robot according to claim 9, characterized in that: The service policy configuration item is a specific service setting defined by the service collaborative robot, including configuration options for language preference, payment method preference and navigation type preference, which guide the service collaborative robot to perform corresponding service operations.

Images