Back to results
US 11,637,904 B2Granted
Electronic device for effect-driven dynamic media selection for visual IoT service using reinforcement learning and operating method thereof
AssigneeKorea Advanced Institute of Science and Technology
InventorsIn-Young Ko, Kyeongdeok Baek
Life of this patent
Estimating the term…
Abstract
An electronic device and an operating method thereof relate to effect-driven dynamic media selection for visual Internet of things (IoT) service using reinforcement learning, and may be configured to monitor a user in an Internet of things (IoT) service environment, predict a visual service effect of at least one service medium related to the user in the IoT service environment, select one of the at least one service medium based on the visual service effect, and provide service for the user through the selected service medium.
Claims8
An operating method of an electronic device, comprising:
- monitoring a user in an Internet of things (IoT) service environment;
- predicting a visual service effect of at least one service medium related to the user in the IoT service environment;
- selecting one of the at least one service medium based on the visual service effect;
- and
- providing service for the user through the selected service medium;
- wherein the predicting of the visual service effect comprises predicting the visual service effect of the at least one service medium based on at least one of a distance between a location of the user and a location of each service medium, whether a visible area of the user is included in the location of each service medium, or an angle between an orientation of the user and an orientation of each service medium;
- and
- wherein the selecting of one of the at least one service medium comprises:
- predicting effectiveness of the at least one service medium based on the visual service effect using a predetermined reinforcement learning (RL) algorithm;
- and
- selecting one of the at least one service medium based on the effectiveness;
- wherein the monitoring of the user comprises monitoring the user while providing the service for the user through a previously selected service medium,
- the operating method further comprising:
- comparing an identifier of the selected service medium and an identifier of the previously selected service medium;
- and
- performing handover from the previously selected service medium to the selected service medium based on a result of the comparison,
- wherein the performing of the handover comprises:
- reducing the predicted visual service effect of the selected service medium when the selected service medium is different from the previously selected service medium;
- determining whether to perform handover to the selected service medium based on the reduced visual service effect of the selected service medium;
- and
- performing the handover if it is determined that the handover to the selected service medium has to be performed.
18 more elements
An electronic device comprising:
- a memory;
- and
- a processor connected to the memory and configured to execute at least one instruction stored in the memory,
- wherein the processor is configured to:
- monitor a user in an Internet of things (IoT) service environment,
- predict a visual service effect of at least one service medium related to the user in the IoT service environment,
- select one of the at least one service medium based on the visual service effect, and
- provide service for the user through the selected service medium;
- wherein the processor is configured to predict the visual service effect of the at least one service medium based on at least one of a distance between a location of the user and a location of each service medium, whether a visible area of the user is included in the location of each service medium, or an angle between an orientation of the user and an orientation of each service medium;
- and
- wherein the processor is configured to:
- predict effectiveness of the at least one service medium based on the visual service effect using a predetermined reinforcement learning (RL) algorithm, and
- select one of the at least one service medium based on the effectiveness;
- wherein the monitoring of the user comprises monitoring the user while providing the service for the user through a previously selected service medium,
- the operating method further comprising:
- comparing an identifier of the selected service medium and an identifier of the previously selected service medium;
- and
- performing handover from the previously selected service medium to the selected service medium based on a result of the comparison,
- wherein the performing of the handover comprises:
- reducing the predicted visual service effect of the selected service medium when the selected service medium is different from the previously selected service medium;
- determining whether to perform handover to the selected service medium based on the reduced visual service effect of the selected service medium;
- and
- performing the handover if it is determined that the handover to the selected service medium has to be performed.
20 more elements
Dependencies are inferred from the claim text.