Mobile Robot Area Classification for Predictable Sequential Coverage

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Autonomous robots lack predictability and user control in their navigation, making it difficult for users to understand which areas they will cover next and how missions will be executed, especially in environments with clutter and varying flooring types.

Innovation Solution

A computing device generates a segmentation map based on occupancy data collected by a mobile robot, classifying areas as non-clutter and clutter, and computes a coverage pattern for navigation, allowing the robot to sequentially traverse these areas in a predetermined sequence, which can be modified by user input for specific cleaning levels and boundaries.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of information

If the robot navigates autonomously without area classification, then the navigation is flexible and adaptive to environment changes, but the user cannot predict which areas the robot will cover next and lacks understanding of mission execution

Engineering Contradiction:
Improveuser understanding of mission executionVSAvoidarea classification system
Core Design Contradiction:
Loss of informationVSDevice complexity

Solution Approach 1:

The patent divides the environment into distinct area types (cluttered areas, non-cluttered areas, boundaries) and creates separate coverage patterns for each type. This segmentation allows the system to provide predictable, area-specific navigation while maintaining overall autonomous flexibility. The segmentation map organizes environmental information into manageable categories that improve user understanding without requiring complete system redesign.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If the robot uses a simple coverage pattern, then the navigation is fast and computationally efficient, but it cannot adapt to different area types such as cluttered and non-cluttered regions

Engineering Contradiction:
Improveadaptation to different area typesVSAvoidnavigation efficiency
Core Design Contradiction:
Adaptability or versatilityVSProductivity

Solution Approach 1:

The patent applies different coverage patterns to different area types: systematic patterns for non-cluttered areas and random patterns for cluttered areas. This local quality approach allows the robot to optimize navigation efficiency for each specific area type while maintaining overall adaptability. The system achieves both versatility and productivity by matching the complexity of the coverage pattern to the complexity of the environment in each region.

Inventive Principle:
Principle #3Local quality

3Reliability

If the robot sequentially navigates classified areas according to a predetermined sequence, then the predictability and efficiency of navigation is enhanced, but the system requires more complex computation to generate and follow the coverage pattern

Engineering Contradiction:
Improvepredictability of navigationVSAvoidcoverage pattern computation
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent generates the segmentation map and coverage pattern in advance, before the robot begins navigation. This preliminary action separates the complex computational tasks from the execution phase, allowing the robot to follow predetermined sequences during actual navigation. The heavy computation is performed offline to create the navigation plan, while online execution simply follows the established pattern, thereby achieving predictability without real-time computational burden.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentEP3512668B1Systems and methods for configurable operation of a robot based on area classification
Publication Date: 2021.07.21 IROBOT CORP
  • EP3512668B1 patent drawingFigure 1~2A
  • EP3512668B1 patent drawingFigure 2B~2C
  • EP3512668B1 patent drawingFigure 3~4

AI summary

A method of operating a mobile robot includes generating a segmentation map defining respective regions of a surface based on occupancy data that is collected by a mobile robot responsive to navigation of the surface, identifying sub-regions of at least one of the respective regions as non-clutter and clutter areas, and computing a coverage pattern based on identification of the sub-regions. The coverage pattern indicates a sequence for navigation of the non-clutter and clutter areas, and is provided to the mobile robot. Responsive to the coverage pattern, the mobile robot sequentially navigates the non-clutter and clutter areas of the at least one of the respective regions of the surface in the sequence indicated by the coverage pattern. Related methods, computing devices, and computer program products are also discussed.