Robot Cleaner Route Planning With Image-Based Obstacle Classification

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

Conventional robot cleaners fail to effectively recognize and avoid obstacles that do not obstruct their movement, leading to inefficient cleaning and potential loss of valuables, as they may suck up small items or move around contaminants without proper navigation.

Innovation Solution

An intelligent robot cleaner equipped with a travel driver, suction unit, image acquisition unit, and controller that analyzes images to classify objects on its path, setting a bypass path if necessary to avoid obstacles like immovable, fragile, or non-suction objects, while maintaining the path for movable objects and ignoring non-interfering items.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If the robot cleaner moves along a fixed predetermined travel path, then the cleaning coverage is ensured, but the robot cleaner cannot avoid obstacles that do not obstruct movement (such as contaminants, small valuables), leading to insufficient cleaning or loss of valuables

Engineering Contradiction:

Improvecleaning effectivenessVSAvoidobstacle recognition and avoidance capability

Core Design Contradiction:

ReliabilityVSAdaptability or versatility

Solution Approach 1:

The robot cleaner performs preliminary actions by acquiring images of the travel path before cleaning, analyzing the images to detect obstacles, and classifying obstacle types in advance. This allows the robot to identify contaminants, small valuables, and other obstacles before encountering them during cleaning operations, enabling proactive avoidance or special handling decisions

Inventive Principle:

Principle #10Preliminary action

Solution Approach 2:

An image acquisition unit and image analysis system serve as intermediaries between the robot cleaner and the environment. The image acquisition unit captures visual information about obstacles on the travel path, and the image analysis system processes this information to classify obstacle types, providing the robot with enhanced perception capabilities without requiring direct physical interaction with obstacles

Inventive Principle:

Principle #24Intermediary (Mediator)

2Reliability

If the robot cleaner avoids all detected objects, then the risk of losing valuables is reduced, but the cleaning becomes insufficient when objects are misclassified as avoidance objects

Engineering Contradiction:

Improvevaluables protectionVSAvoidcleaning efficiency

Core Design Contradiction:

ReliabilityVSProductivity

Solution Approach 1:

The robot cleaner applies different quality treatments to different objects based on their classification. Avoidance objects (small valuables, contaminants) receive special handling through path deviation, while non-avoidance objects (movable debris suitable for suction) are cleaned normally. This localized differentiation ensures valuables are protected while maintaining cleaning efficiency for appropriate targets

Inventive Principle:

Principle #3Local quality

Solution Approach 2:

The system changes the navigation parameter (avoidance behavior) based on the classification result of the image analysis. When an object is classified as an avoidance object, the robot changes its travel path to bypass it; when classified as a non-avoidance object, the robot maintains its original path and performs normal cleaning, dynamically adjusting behavior according to object characteristics

Inventive Principle:

Principle #35Parameter changes

3Device complexity

If the robot cleaner uses simple obstacle detection, then the device complexity is low, but the robot cannot distinguish between different types of obstacles (contaminants, valuables, movable objects)

Engineering Contradiction:

Improvedetection system simplicityVSAvoidobstacle classification accuracy

Core Design Contradiction:

Device complexityVSMeasurement precision

Solution Approach 1:

The patent replaces complex mechanical obstacle detection and classification systems with an optical-based image acquisition and analysis system. Instead of using multiple sensors, tactile feedback mechanisms, or complex mechanical probes to identify obstacle types, the system uses image capture and digital analysis to achieve accurate obstacle classification with simpler physical hardware

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS11565411B2Intelligent robot cleaner for setting travel route based on video learning and managing method thereof

Publication Date: 2023.01.31 LG ELECTRONICS INC

AI summary

An intelligent robot cleaner setting a travel path based on a video learning includes a travel driver, a suction unit, an image acquisition unit, and a controller. The travel driver moves to an area to be cleaned along the travel path. The suction unit sucks foreign substances on the travel path. The image acquisition unit acquires an image on the travel path. The controller analyzes the image, decides whether an object is present on the travel path, classifies a type of the object, and sets a bypass travel path that avoids the object if the object is an avoidance object.