Autonomous Floor Robot Spot Cleaning via Color Blob Tracking
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Autonomous robotic vacuum cleaners face challenges in efficiently targeting and cleaning areas with a threshold level of dirt or debris, as they often rely on general navigation patterns rather than specific dirt detection and targeting.
Innovation Solution
A mobile floor cleaning robot equipped with a drive system, imaging sensor, and controller that segments images into color blobs, tracks their location, and issues drive commands to maneuver towards areas with significant dirt, allowing for spot cleaning of dirty regions before general cleaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the robot uses general navigation patterns to clean the floor, then it can cover the entire floor area, but it cannot efficiently target and clean areas with threshold level of dirt or debris
Solution Approach 1:
The robot performs preliminary dirt detection by capturing images and segmenting them into color blobs before cleaning. The controller identifies areas with threshold spatial distribution of colors indicating dirt, then navigates to these locations first for spot cleaning, rather than cleaning the entire floor uniformly. This preliminary detection and targeted approach resolves the contradiction by enabling both precise dirt detection and improved cleaning efficiency.
2Productivity
If the robot cleans the entire floor area uniformly, then it ensures complete coverage, but it wastes time cleaning already clean areas
Solution Approach 1:
The robot applies local quality by differentiating cleaning actions based on local dirt conditions. The image segmentation identifies specific regions with color blobs indicating dirt, and the robot performs spot cleaning only in those regions. After spot cleaning, it proceeds to general cleaning of the entire area. This ensures both efficiency (by skipping already clean areas) and completeness (by eventually cleaning everywhere).
3Measurement precision
If the robot uses image segmentation to identify dirty areas, then it can target specific dirt locations, but it increases processing complexity
Solution Approach 1:
The robot uses color changes as the basis for dirt detection. The controller segments images by identifying color blobs and determining their spatial distribution. Areas with threshold spatial distribution of certain colors are identified as dirty. This color-based approach provides accurate dirt location identification while maintaining relatively simple processing compared to more complex image analysis methods.
Data Source
Figure 1
Figure 2
Figure 3
AI summary
A mobile floor cleaning robot (100) includes a robot body (110) supported by a drive system (120) configured to maneuver the robot over a floor surface (10). The robot also includes a cleaning system (160) disposed on the robot body, an imaging sensor (510) disposed on the robot body, and a controller (150) in communication with the drive system and the imaging sensor. The controller receives a sequence of images (514b) of the floor surface (10), each image (514) having an array of pixels (516). The controller segments each image into color segments (23) by color quantizing pixels (514) of the image and determining a spatial distribution of at least one color of the image based on corresponding pixel locations (517). The controller then tracks a location (12a) of at least one color segment with respect to the imaging sensor across the sequence of images (514) for determining a drive command for issuance to the drive system (120) to maneuver the mobile floor cleaning robot (100).