IoT Device Control via Server-Side Intent Mapping

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Current voice-assisted IoT devices are limited to predefined commands, leading to increased cognitive load for users when they fail to recognize non-predefined inputs, as users must reutter the command, which can be cumbersome and inefficient.

Innovation Solution

A method and system that allow IoT devices to interpret user inputs by transmitting unrecognized commands to a server for processing, which identifies activities, objects, and contexts, and maps them to corresponding actions based on user interactions and probability information, enabling the device to perform actions even if the initial command is not predefined.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If voice-assisted devices support only predefined commands, then device complexity is reduced and ease of operation is improved for common tasks, but adaptability deteriorates when users provide non-predefined inputs

Engineering Contradiction:

Improveability to recognize and process non-predefined user inputsVSAvoidcomplexity of command processing system

Core Design Contradiction:

Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent introduces a server as an intermediary component that receives unrecognized voice inputs from the electronic device, processes them using machine learning models, and returns appropriate actions. This mediator handles the complexity of natural language understanding externally, allowing the electronic device to maintain simplicity while gaining enhanced adaptability through the server's processing capabilities.

Inventive Principle:

Principle #24Intermediary (Mediator)

2Ease of operation

If the system requires users to reutter predefined commands when unrecognized, then measurement precision of command recognition is maintained, but ease of operation deteriorates due to increased cognitive load

Engineering Contradiction:

Improveuser interaction convenienceVSAvoidloss of user intent when command not recognized

Core Design Contradiction:

Ease of operationVSLoss of information

Solution Approach 1:

The system implements a feedback mechanism where the server analyzes unrecognized voice inputs, identifies user intent through machine learning, and provides feedback in the form of suggested actions or corrections to the user. This feedback loop allows the system to learn from user interactions and progressively improve command recognition accuracy while maintaining ease of operation by guiding users rather than simply rejecting their inputs.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The system performs preliminary analysis of unrecognized voice inputs by transmitting them to the server for processing before final action execution. The server pre-processes these inputs using trained machine learning models to identify potential user intents and map them to appropriate actions, preparing the system to handle diverse user inputs without requiring exact predefined command matches.

Inventive Principle:

Principle #10Preliminary action

3Adaptability or versatility

If the system transmits unrecognized inputs to a server for processing, then adaptability is improved, but loss of time increases due to network communication overhead

Engineering Contradiction:

Improveability to handle diverse user inputsVSAvoidtime for network communication and processing

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system applies partial action by transmitting only unrecognized voice inputs to the server for processing, while recognized predefined commands are executed locally without network communication. This selective transmission approach minimizes network overhead and time loss by limiting server communication to only those cases where additional processing is necessary, rather than transmitting all user inputs.

Inventive Principle:

Principle #16Partial or excessive action

Data Source

PatentUS20230385377A1Device, method, and computer program for performing actions on IoT devices

Publication Date: 2023.11.30 SAMSUNG ELECTRONICS CO LTD

AI summary

A method of controlling an Internet of Things (IoT) device to perform an action, is performed by an electronic device and includes obtaining at least one input from a user, based on a predefined action corresponding to the obtained at least one input being not identifiable, transmitting the obtained at least one input to a server, receiving, from the server, information regarding at least one activity that is mapped to at least one IoT device that is identified based on the transmitted at least one input, and information regarding at least one object and/or at least one context, and identifying an action to be performed on an IoT device among the at least one IoT device, based on a correlation between the obtained at least one input and the at least one activity mapped to the at least one IoT device.