Mobile Robot Obstacle Detection Using Single-Camera Image Segmentation

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robot cleaners struggle to perform autonomous traveling and detect obstacles using only a single camera, limiting their ability to navigate effectively in various environments.

Innovation Solution

A robot cleaner equipped with a main body, a cleaning unit, a driving unit, a camera sensor, and an operation sensor, where the controller processes images from the camera sensor to detect obstacles and control the driving unit accordingly, using techniques such as image segmentation, feature point extraction, and weight value setting.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Ease of manufacture

If a single camera is used for obstacle detection, then manufacturing costs are reduced and device complexity is simplified, but measurement precision and detection reliability deteriorate

Engineering Contradiction:
Improvemanufacturing costVSAvoidobstacle detection precision
Core Design Contradiction:
Ease of manufactureVSMeasurement precision

Solution Approach 1:

The patent segments the captured image into multiple regions (e.g., foreground, background, sky, ground) and processes each region separately to detect obstacles. This segmentation approach enables a single camera to achieve reliable obstacle detection by analyzing different image regions with appropriate detection algorithms, thereby resolving the contradiction between using a simple single-camera system and achieving precise obstacle detection.

Inventive Principle:
Principle #1Segmentation

2Device complexity

If a single camera is used for obstacle detection, then device complexity is reduced, but the ability to accurately detect obstacles in various environments deteriorates

Engineering Contradiction:
Improvesensor system complexityVSAvoidenvironmental adaptability
Core Design Contradiction:
Device complexityVSAdaptability or versatility

Solution Approach 1:

The patent changes detection parameters (such as detection thresholds, region definitions, and processing algorithms) based on the captured image content and environmental conditions. By dynamically adjusting these parameters, the single-camera system can adapt to various environments (indoor, outdoor, different lighting conditions) while maintaining simple device architecture, thus resolving the contradiction between device simplicity and environmental adaptability.

Inventive Principle:
Principle #35Parameter changes

3Measurement precision

If image processing techniques are applied to enhance obstacle detection, then measurement precision improves, but processing time and computational complexity increase

Engineering Contradiction:
Improveobstacle detection precisionVSAvoidprocessing time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The patent performs preliminary actions by pre-defining image regions (such as separating sky and ground regions) and pre-setting detection parameters before actual obstacle detection. This preliminary processing organizes the image data in advance, allowing the main detection algorithm to work more efficiently on structured regions, thereby reducing overall processing time while maintaining high detection precision.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3328597B1Mobile robot and control method thereof
Publication Date: 2025.03.12 LG ELECTRONICS INC
  • EP3328597B1 patent drawingFigure 1a~1b
  • EP3328597B1 patent drawingFigure 2~3
  • EP3328597B1 patent drawingFigure 4~5a

AI summary

A cleaner performing autonomous traveling includes a main body having a suction opening, a cleaning unit provided within the main body and sucking a cleaning target through the suction opening, a driving unit moving the main body, a camera sensor attached to the main body and capturing a first image, an operation sensor sensing information related to movement of the main body, and a controller detecting information related to an obstacle on the basis of at least one of the captured image and the information related to movement and controlling the driving unit on the basis of the detected information related to the obstacle.