Method of detecting a difference in level of a surface in front of a robotic cleaning device
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Robotic vacuum cleaners often get stuck on obstacles like doorsteps or thick rugs, and fail to detect ledges, which can result in damage when they fall down stairs.
Innovation Solution
A method and device for detecting differences in level using illumination and image capture, where a robotic cleaning device projects light onto a surface, captures images, identifies segments, and determines the positional relationship between them to assess the level difference, allowing for timely navigation and avoidance of obstacles.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Extent of automation
If the robotic cleaner uses autonomous navigation to move freely around a space, then it can clean surfaces without human intervention, but it may get stuck on obstacles like doorsteps or thick rugs and fail to detect ledges
Solution Approach 1:
The patent introduces an intermediary detection system consisting of light sources and cameras between the robotic cleaner and the obstacles/ledges. This intermediary system captures images of the surface, detects luminous sections, and processes positional relationships to identify level differences, enabling the autonomous cleaner to reliably detect obstacles and ledges without direct contact
Solution Approach 2:
The patent implements preliminary detection of ledges and obstacles by analyzing images captured before the robotic cleaner reaches them. The system processes the captured images to detect luminous sections and determine positional relationships, allowing the cleaner to plan its movement in advance and avoid falling off ledges or getting stuck on obstacles
2Productivity
If the robotic cleaner approaches obstacles like doorsteps or thick rugs, then it can clean these areas, but it tends to get stuck on them
Solution Approach 1:
The patent applies preliminary detection by capturing images and processing them to identify obstacles like doorsteps and thick rugs before the robotic cleaner reaches them. The system determines the height and type of obstacles in advance, allowing the cleaner to plan its approach and movement strategy to avoid getting stuck while still maintaining cleaning coverage
3Measurement precision
If the robotic cleaner detects ledges by physical contact or proximity sensors, then it can identify obstacles, but it may not detect ledges in time to prevent falling
Solution Approach 1:
The patent implements preliminary detection by using light sources and cameras to capture images of the surface in front of the robotic cleaner. The system processes these images to detect luminous sections and determine positional relationships, identifying ledges at a distance before the cleaner reaches them, providing sufficient time for response and avoidance
Solution Approach 2:
The patent replaces mechanical or proximity-based detection systems with an optical detection system using light sources and cameras. This substitution enables earlier and more accurate detection of ledges by analyzing the positional relationship of luminous sections in captured images, providing both high measurement precision and adequate response time
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
Enables the robotic device to plan its movement effectively, avoid traversing obstacles, and efficiently manage brush usage by determining the height of surfaces, preventing damage and ensuring efficient cleaning.
Implementation Method 1
illuminating the surface with light
Implementation Method 2
detecting a luminous section in the captured image caused by the light
Data Source
AI summary
A method for a robotic cleaning device of detecting a difference in level of a surface in front of the robotic cleaning device moves. The method includes illuminating the surface with light, capturing an image of the surface, detecting a luminous section in the captured image caused by the light, identifying at least a first segment and a second segment representing the detected luminous section, and detecting, from a positional relationship between the identified first and second segment, the difference in level of the surface.