Robotic Cleaner Image Control for Targeted Sub-Area Sweeping

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Existing robotic cleaners lack efficient methods to identify and sweep specific sub-areas within a larger area, leading to unnecessary sweeping and reduced efficiency.

Innovation Solution

The method involves acquiring an environment image of the area, determining the sub-area that needs to be swept based on the image, and controlling the robotic cleaner to reach and perform a sweeping task in that sub-area, thereby reducing computation complexity and improving efficiency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If the robotic cleaner sweeps the whole area using automatic sweeping mode, then the entire area is cleaned, but time and energy are wasted on areas that do not need cleaning

Engineering Contradiction:
Improvearea identification accuracyVSAvoidsweeping time
Core Design Contradiction:
Measurement precisionVSLoss of time

Solution Approach 1:

The system performs preliminary image acquisition and sub-area determination before the sweeping operation. By capturing an environment image and identifying dirty sub-areas in advance, the robotic cleaner can navigate directly to areas requiring cleaning rather than sweeping the entire area, thus reducing time consumption while maintaining cleaning effectiveness.

Inventive Principle:
Principle #10Preliminary action

2Productivity

If the robotic cleaner performs active recognition to identify dirty areas, then sweeping efficiency is improved, but computation complexity increases

Engineering Contradiction:
Improvesweeping efficiencyVSAvoidcomputation complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

The environment image is divided into multiple blocks for independent analysis. Each block is processed to determine whether it represents a dirty sub-area, allowing the system to identify cleaning targets without analyzing the entire image as a single complex unit. This segmentation approach improves detection accuracy while managing computational complexity through divide-and-conquer processing.

Inventive Principle:
Principle #1Segmentation

Data Source

PatentEP3327533B1Method and device for controlling robotic cleaner and electronic device
Publication Date: 2025.05.21 BEIJING XIAOMI MOBILE SOFTWARE CO LTD
  • EP3327533B1 patent drawingFigure 1A~1B
  • EP3327533B1 patent drawingFigure 1C~1D
  • EP3327533B1 patent drawingFigure 2

AI summary

A method and a device are provided for controlling a robotic cleaner. According to an example of the method, an environment image of an area may be acquired, and a sub-area to be swept in the area may be determined based on the environment image. Then, the robotic cleaner may be controlled to reach the sub-area to be swept and perform a sweeping task in the sub-area to be swept.