Robot Cleaner Depth Image Supplementation for Thin Obstacle Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing robot cleaners struggle to accurately sense and avoid thin obstacles like wires due to unclear portions in depth images, leading to potential damage.
Innovation Solution
Supplement depth images with color or brightness information from IR or RGB images to enhance obstacle recognition, using a camera to combine and correct discontinuities in distance values.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If a depth sensor is used to detect obstacles, then the robot cleaner can sense the distance to objects, but thin obstacles like wires produce unclear depth images due to noise and diffused reflection/absorption
Solution Approach 1:
The patent combines depth information from the depth sensor with color/brightness information from an IR or RGB camera to create a composite image. This merging of multiple information sources allows thin obstacles that produce unclear depth images to be identified through their color or brightness characteristics, thereby resolving the contradiction between measurement precision and information loss.
Solution Approach 2:
The patent introduces an intermediary processing step that identifies singularities (unclear portions) in the depth image and supplements them with corresponding regions from the color or brightness image. This intermediary process acts as a bridge between the depth sensor and the final obstacle detection, allowing thin obstacles to be recognized through multiple modalities.
2Reliability
If the robot cleaner uses only depth sensing, then the device complexity is low, but it cannot accurately identify thin obstacles
Solution Approach 1:
The patent merges the depth sensing function with color/brightness sensing functions by integrating a camera (IR or RGB) with the existing depth sensor system. This combination enhances reliability for identifying thin obstacles while maintaining relatively simple device architecture through shared processing infrastructure.
Solution Approach 2:
The patent makes the imaging system multi-functional by using the camera to serve both as a color/brightness sensor and as a supplement to the depth sensor. The same camera hardware performs multiple roles: capturing color information for thin obstacle detection and providing visual data that complements depth information, thereby improving reliability without proportionally increasing device complexity.
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
Improves obstacle sensing accuracy, enabling the robot cleaner to effectively avoid thin obstacles like wires, preventing damage.
Implementation Method 1
irradiating, by a light source, light toward a location the same as a location where the acquired image is captured, receiving, by a sensor, the light irradiated from the light source and reflected on an object
Data Source
AI summary
Disclosed is a method for controlling a robot cleaner including acquiring, by a camera, an image, irradiating, by a light source, light toward a location the same as a location where the acquired image is captured, receiving, by a sensor, the light irradiated from the light source and reflected on an object, processing an image received from the sensor to contain a distance value of an individual location, and supplementing the image received from the sensor with the image captured by the camera when a singularity is found, wherein distance values calculated in adjacent portions are discontinuous at the singularity.