Mobile Robot Home Appliance Mapping Using Image-Based Deep Learning

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Mobile robots face challenges in determining the position and type of home appliances within a traveling area, which hinders their ability to effectively communicate and remotely control these appliances for user convenience and situational control.

Innovation Solution

A mobile robot equipped with an image acquisition unit and a controller that performs deep learning to identify home appliances based on images, generating a map that indicates the type, position, and controllability of appliances, allowing for remote control based on user input and acquired information.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If the mobile robot uses traditional obstacle detection methods to navigate the traveling area, then the robot can avoid obstacles and move autonomously, but the robot cannot identify home appliances and their positions for remote control functionality

Engineering Contradiction:
Improvehome appliance identification capabilityVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The image acquisition unit and image processing algorithms serve multiple functions: they detect obstacles for navigation, identify home appliance types through pattern recognition, and determine appliance positions for mapping. This multi-functional approach enables the robot to perform both navigation and home appliance identification using the same hardware and software resources, thereby improving adaptability without proportionally increasing system complexity

Inventive Principle:
Principle #6Universality (Multi-functionality)

2Loss of information

If the mobile robot equips an image acquisition unit and deep learning system to identify home appliances, then the robot can determine appliance types and positions, but the device complexity and computational requirements increase

Engineering Contradiction:
Improveappliance information acquisitionVSAvoidprocessing system complexity
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The system pre-processes images to extract key features and pre-trains deep learning models with appliance data before deployment. By performing preliminary actions such as feature extraction and model training in advance, the robot reduces the computational burden during actual operation, enabling accurate appliance identification while managing processing system complexity

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The robot creates a digital map that copies and represents the physical traveling area, including the positions and types of home appliances. This virtual representation allows the system to store and process appliance information efficiently without requiring continuous complex image analysis, thereby reducing ongoing computational requirements while maintaining complete appliance information

Inventive Principle:
Principle #26Copying

3Loss of information

If the mobile robot generates a detailed map with home appliance information, then the robot can provide intuitive appliance arrangement visualization, but the data processing and storage requirements increase

Engineering Contradiction:
Improveappliance arrangement informationVSAvoiddata volume
Core Design Contradiction:
Loss of informationVSQuantity of substance

Solution Approach 1:

The system extracts only the essential information needed for appliance identification and positioning from the full images, such as appliance type labels and coordinate positions. By extracting only the necessary data elements and discarding redundant image details, the robot creates a compact appliance map that provides complete appliance arrangement information while minimizing data volume for storage and processing

Inventive Principle:
Principle #2Taking out (Extraction)

Data Source

PatentUS11317778B2Mobile robot
Publication Date: 2022.05.03 LG ELECTRONICS INC
  • US11317778B2 patent drawing
  • US11317778B2 patent drawing
  • US11317778B2 patent drawing

AI summary

Disclosed is a mobile robot for determining a type and position of a home appliance present within in a traveling area and controlling the home appliance according to a control command or a situation. The mobile robot includes a controller for determining the type and position of the home appliance positioned within the traveling area based on the image acquired through an image acquisition unit, and for generating and outputting a home appliance map that is a traveling area map on which at least one of the type or position of the home appliance is indicated, wherein the home appliance is remotely controlled according to a control command of a user, a situation is determined according to the acquired information without a control command, and a home appliance appropriate for a specific situation is controlled.