KR-102963898-B1 - OFFLINE TARGETED ADVERTISING SYSTEM AND METHOD USING ON-DEVICE ROBOT OBJECTS

Abstract

An offline targeted advertising system according to one embodiment is disclosed, comprising a mobile robot entity having display units arranged in multiple directions, a sensor unit for collecting user data and surrounding environment data, an on-device de-identification processing unit that processes the collected data locally without transmitting or storing it to an external server to generate de-identified user feature information and context information, an advertising control unit that controls the selection and display of locally stored advertising content by display unit based on the user feature information and context information, and a performance measurement unit that calculates the advertising performance of the displayed advertisement, thereby enabling low-latency offline targeted advertising execution without the risk of personal information leakage.

Inventors

• 이정음

Assignees

• 주식회사 피오유엘

Dates

Publication Date

20260512

Application Date

20250911

Claims (5)

1. In an offline targeted advertising system utilizing on-device-based robot entities, A mobile robot entity having display units arranged in multiple directions; A sensor unit mounted on the above-mentioned robot entity to collect user data and surrounding environment data; An on-device de-identification processing unit that generates de-identified user feature information and context information by processing the collected user data and surrounding environment data locally without transmitting or storing them to an external server; An ad control unit that selects an output target for each display unit from among locally stored ad content based on the above user feature information and context information, and displays the ad content selected as the output target on the plurality of displays; and A performance measurement unit that calculates the advertising performance of the advertising content displayed above; and The above-mentioned offline targeted advertising system further includes a movement control unit, and The above movement control unit is, The above robot entity moves autonomously based on a predefined movement path using position information, distance information, and image information acquired by the sensor unit, and Based on the obstacle detection result of the sensor unit, at least one of path replanning, deceleration, stopping, and detour is performed in real time, and When the battery status of the robot object drops below a predetermined threshold, the robot calculates a return path to a charging point and returns, and optimizes the movement path to improve the visibility of the display unit according to the display instructions of the advertising control unit. The above-mentioned advertising control unit is, User speech data, gestures, and interest signals obtained from the above sensor unit and de-identification processing unit are analyzed to extract user speech keywords, and Extracting environmental condition keywords by analyzing signboard character recognition results of a surrounding commercial area including a preset movement path, background music, surrounding advertising music, crowd density, illuminance, and time information from the video and audio inputs of the sensor unit, and Calculate the similarity between the above user utterance keyword and the above environment condition keyword, and if the similarity is greater than or equal to a preset standard, identify the ad content combination and display mode candidate for each display unit corresponding to the keyword combination as the first candidate ad provision combination. Among the first candidate ad offering combinations mentioned above, a combination in which the cumulative exposure overlap rate, acoustic collision risk, or visual congestion index does not exceed a preset allowable value is identified as a second candidate ad offering combination, and Among the above second candidate ad delivery combinations, the combination with the highest combination balance score calculated based on the interest prediction score, event participation probability, and conversion latency distribution calculated by the performance measurement unit is selected as the final ad delivery combination, and Extract advertising content corresponding to the final advertising delivery combination from among the advertising content stored locally above, and Calculate the advertising index by checking the number of the extracted advertising content and the length of each advertising content, and If the combination balance score of the extracted advertising content does not satisfy a preset standard, control is made to display the advertising content in a first display mode that outputs advertising content of different brands to multiple display units. When the combination balance score of the extracted advertising content satisfies a preset standard and the advertising index is below a preset standard, the advertising content is controlled to be displayed in a second display mode in which advertising content of the same brand is output to a plurality of display units. When the combination balance score of the extracted advertising content satisfies a preset standard and the advertising index is greater than or equal to the preset standard, control to display advertising content in a third display mode in which advertising content of the same brand and participatory advertising content are respectively output to a plurality of display units. The above second display mode is, A 2-1 display mode that renders the above-mentioned plurality of display units as a connected single screen to display advertising content of the same brand as a continuous screen, and It includes a 2-2 display mode that synchronizes and simultaneously displays advertising content of the same brand on the plurality of display units mentioned above, and The above-mentioned advertising control unit is characterized by controlling operation to one of the 2-1 display mode and the 2-2 display mode based on at least one of the gaze retention index, crowd density, robot speed, and difference in viewing angle. Offline targeted advertising system.
2. In paragraph 1, The above-mentioned de-identification processing unit is, The original image and voice data collected by the sensor unit are processed locally in real time without being transmitted to an external server or stored non-volatilely, and identifiable areas including face, body, and voice are converted into an unidentifiable state through masking, blurring, or low-resolution processing. From the above-mentioned transformed data, information including at least one of age group, gender, cluster size, gaze, dwell time, facial expression, and voice response is calculated as a de-identified feature vector, and Dynamically adjust the level of de-identification and output items according to at least one of location, time, crowd density and applicable regulations, and After processing is complete, personally identifiable original or intermediate data shall be immediately disposed of, and An offline targeted advertising system characterized by providing the above-described feature vector to the advertising control unit as user feature information and context information.
3. In paragraph 1, The above performance measurement unit is, For the advertising content displayed above, the number of exposures, gaze time, user group size, and response indicators are calculated based on the detection result of the sensor unit and the output of the anonymization processing unit, and The above calculation results are aggregated in a de-identified form based on age group, gender, cluster size, or time period to generate statistical data that does not include individual identification information, and An offline targeted advertising system characterized by formatting the statistical data generated above into insight data for advertisers and outputting it along with metadata to enable the generation of an insight report including at least one of preferred advertisements by age group, exposure effects by time of day, response rate by display direction, and interest index.
4. In an offline targeted advertising method utilizing an on-device-based robot entity, A step of controlling a mobile robot object to move based on preset movement path information; A step of collecting user data and surrounding environment data from a sensor mounted on the robot object; A step of generating user feature information and context information by de-identifying the collected data above; A step of selecting an output target from locally stored advertising content based on the above user feature information and context information; A step of displaying advertising content corresponding to the above output target on the display part of the robot object; A step of calculating advertising performance indicators based on user data collected from the sensor for the advertising content displayed above; and It includes the step of generating a performance report based on the advertising performance indicators calculated above and providing it to the advertiser, The step of controlling the above-mentioned mobile robot object is, A step of controlling autonomous movement along the movement path using position information, distance information, and image information obtained from the sensor; A step of controlling to replan the movement path or to perform at least one of deceleration, stopping, or detour based on obstacle information detected by the sensor; A step of controlling the mobile robot object to return by calculating a return path to a charging point when the battery status of the mobile robot object is below a preset threshold; and It includes a step of controlling to optimize the movement path so as to improve the visibility of the above-mentioned display unit, and The step of selecting the above advertising content is, A step of extracting user utterance keywords by analyzing user utterance data, gestures, and interest signals from the above user data; A step of extracting environmental condition keywords by analyzing signboard character recognition results, background music, ambient advertising sound, crowd density, illuminance, and time information from the above surrounding environment data; A step of calculating the similarity between the above user utterance keyword and the above environment condition keyword, and if the similarity is greater than or equal to a preset standard, identifying the ad content combination and display mode candidate as the first candidate ad provision combination; A step of identifying a combination among the first candidate ad offering combinations as a second candidate ad offering combination in which the cumulative exposure overlap rate, acoustic collision risk, or visual congestion index does not exceed a preset allowable value; A step of selecting the combination with the highest combination balance score calculated based on interest prediction scores, event participation probability, and conversion latency distribution among the above second candidate ad serving combinations as the final ad serving combination; A step of extracting advertising content corresponding to the above-mentioned final ad delivery combination from a local storage; A step of calculating an advertising index based on the number of advertising content and the length of each advertising content; and Based on the above combination balance score and advertising index, the method includes a step of controlling the display of advertising content by selecting one of a first display mode that outputs advertising content of different brands to a plurality of display units, a second display mode that outputs advertising content of the same brand to a plurality of display units, or a third display mode that outputs advertising content of the same brand and participatory advertising content to a plurality of display units, respectively. The above second display mode is, A 2-1 display mode that renders multiple display units as a single connected screen to display them as a continuous screen; and It includes a 2-2 display mode that synchronizes and simultaneously displays the same advertisement content on multiple display units, and The method further comprises the step of selecting either the 2-1 display mode and the 2-2 display mode based on at least one of a gaze retention index, crowd density, the movement speed of a mobile robot object, and the difference in viewing angle. Offline targeted advertising method utilizing on-device based robot entities.
5. delete

Description

Offline Targeted Advertising System and Method Using On-Device Robot Objects The following embodiments relate to a system and method in which a mobile robot entity equipped with on-device multimodal artificial intelligence performs display control, performance measurement, and provision of insights for targeted advertisements in an offline environment, based on the premise of anonymization of personal information. Offline advertising has traditionally been executed primarily through fixed billboards or simple electronic displays; consequently, precise targeting reflecting contextual factors such as the age, gender, and crowd size of the audience has been difficult, and quantitatively measuring interest and conversion relative to exposure has been limited. While some digital signage attempts camera-based response analysis, the server-dependent structure for transmitting and storing original video and audio poses risks of personal information leakage, network latency, and regulatory compliance burdens. Furthermore, it presents limitations in that real-time adaptation and fault response are difficult in the event of connection instability. The combination of robotic technology and lightweight multimodal artificial intelligence has the potential to overcome the aforementioned limitations by enabling context awareness considering movement paths and surrounding environments, on-device anonymization, multi-faceted display synchronization, display quality control based on energy conditions, and performance reporting based on aggregated statistics. However, to achieve this, a system is required that integrates a processing structure that does not store or transmit personally identifiable information, ad matching combining viewer characteristics and environmental conditions, mode switching and synchronization control for front and rear displays, interlocking of movement and display controls, and low-power inference and performance guarantee technologies. Accordingly, there is a need to develop offline targeted advertising systems and methods utilizing on-device-based robot entities. FIG. 1 is a diagram illustrating the configuration of an offline targeted advertising system utilizing an on-device-based robot entity according to one embodiment. FIG. 2 is a flowchart illustrating an offline targeted advertising method using an on-device-based robot entity according to one embodiment. FIG. 3 is a diagram illustrating the autonomous movement process of a robot object according to one embodiment. Hereinafter, embodiments are described in detail with reference to the attached drawings. However, various modifications may be made to the embodiments, and thus the scope of the patent application is not limited or restricted by these embodiments. It should be understood that all modifications, equivalents, and substitutions to the embodiments are included within the scope of the rights. Specific structural or functional descriptions of the embodiments are disclosed for illustrative purposes only and may be modified and implemented in various forms. Accordingly, the embodiments are not limited to the specific disclosed forms, and the scope of this specification includes modifications, equivalents, or substitutions that fall within the technical concept. Terms such as "first" or "second" may be used to describe various components, but these terms should be interpreted solely for the purpose of distinguishing one component from another. For example, the first component may be named the second component, and similarly, the second component may be named the first component. When it is stated that a component is "connected" to another component, it should be understood that it may be directly connected to or joined to that other component, or that there may be other components in between. The terms used in the embodiments are for illustrative purposes only and should not be interpreted as intended to be limiting. Singular expressions include plural expressions unless the context clearly indicates otherwise. In this specification, terms such as "comprising" or "having" are intended to indicate the existence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof. Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the embodiments pertain. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application. In addition, when describing with reference to the attached drawings, iden