Robot cleaner and method for controlling the same
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot cleaners face challenges in distinguishing between light patterns projected for navigation and those generated by light reflections from floors or objects, which can interfere with accurate area mapping and obstacle detection.
Innovation Solution
The robot cleaner employs a pattern projection unit and an image acquisition unit to project light patterns and capture images, using a controller with a validity determination module to geometrically analyze the patterns and differentiate between valid projected patterns and those caused by reflections, allowing for accurate obstacle detection and navigation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If light is projected toward the floor to recognize cleaning area state, then the robot can map the cleaning area and detect obstacles, but reflected light patterns are generated that interfere with accurate pattern recognition
Solution Approach 1:
The patent applies geometric analysis in a multi-dimensional space by considering the spatial relationships between the pattern projection unit, image acquisition unit, and detected patterns. By analyzing patterns from multiple geometric perspectives and using coordinate transformations, the system can distinguish reflected patterns from direct patterns based on their positional relationships in the imaging space.
Solution Approach 2:
The patent introduces geometric analysis as an intermediary processing step between pattern detection and interpretation. The geometric analysis module acts as a mediator that uses spatial relationships and geometric constraints to filter out reflected patterns before they interfere with the final cleaning area mapping, thereby resolving the contradiction between utilizing reflected light for illumination and eliminating its interference.
2Loss of information
If a camera photographs the projected light pattern to analyze cleaning area state, then the robot can recognize floor position and obstacles, but it becomes difficult to discriminate necessary patterns from unnecessary reflected patterns
Solution Approach 1:
The patent performs preliminary geometric analysis on detected patterns before using them for cleaning area mapping. By pre-filtering patterns based on geometric constraints and spatial relationships, the system eliminates reflected patterns early in the processing chain, preventing them from contaminating the final cleaning area information and reducing the complexity of subsequent interpretation steps.
Solution Approach 2:
The patent implements a feedback mechanism where the geometric analysis results are used to validate and correct the detected patterns. The system continuously compares detected pattern positions with expected geometric relationships, and uses this feedback to identify and exclude reflected patterns, thereby maintaining high information accuracy without requiring overly complex discrimination mechanisms.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
This solution enables the robot cleaner to effectively discriminate between projected and reflected light patterns, enhancing its ability to map the cleaning area and detect obstacles, thereby improving its navigation and cleaning efficiency.
Implementation Method 1
since some materials of floors or objects to which light is projected reflect the light, a new light pattern due to reflection can be detected in addition to the original light pattern
Data Source
Figure 1
Figure 2~3
Figure 4
AI summary
A method for controlling a robot cleaner configured to project light of a predetermined pattern to a floor of an area in front thereof in a specific direction and to acquire an image of the area to which the light is projected, according to the present invention, includes: (a) acquiring a front view image of the robot cleaner; (b) detecting the pattern from the image acquired in (a); and (c) discriminating a pattern formed by reflected light from a pattern formed by light directly projected from the robot cleaner, from among patterns indicated in the acquired image, on the basis of geometrical characteristics defined by a projection direction of the light and the direction of an optical axis in which the image is acquired when two or more patterns are detected from the image.