Mobile Robot Distance Map Area Division for Complex Indoor Cleaning
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing mobile robot navigation systems struggle to accurately map and clean complex indoor environments, such as offices, due to difficulties in segmenting areas and navigating around obstacles, especially in spaces with many narrow areas and varying structures.
Innovation Solution
The mobile robot employs a combination of vision-based and lidar-based location recognition technologies, using cameras and laser sensors to generate robust maps and recognize obstacles, while also utilizing artificial neural networks for improved obstacle detection and path planning, allowing for efficient area division and cleaning.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If area segmentation is performed using conventional methods (feature points, erosion-expansion), then the cleaning area can be divided into regions, but the segmentation points do not match suitable locations in complex environments with many narrow areas, leading to poor cleaning performance
Solution Approach 1:
The patent changes the parameter basis for segmentation from geometric features (feature points, erosion-expansion results) to depth information from distance maps. By using depth values at pixel centers and comparing them against threshold levels, the system identifies suitable segmentation points that correspond to actual spatial boundaries, ensuring accurate area division in complex environments with narrow passages
Solution Approach 2:
The patent replaces the mechanical/geometric segmentation approach (based on pixel width and structural features) with a depth-based information processing method. Instead of relying on fixed geometric rules, the system uses depth map analysis to dynamically identify segmentation points that reflect the actual three-dimensional structure of the environment, improving adaptability to various spatial configurations
2Extent of automation
If the robot uses a grid map for navigation, then the map can be generated and processed, but the grid map created by the robot is significantly different from the actual area drawing, making it difficult for users to intuitively grasp area information
Solution Approach 1:
The patent introduces distance maps as an intermediary representation between the raw grid map and the final area segmentation. The distance map, which encodes depth information from obstacles to each pixel, serves as a mediator that preserves spatial relationships and structural information, enabling both automatic processing and intuitive visualization of area boundaries
Solution Approach 2:
The patent adds a depth dimension to the traditional two-dimensional grid map by generating distance maps that encode three-dimensional spatial information. This dimensional enhancement allows the system to capture vertical distance relationships and structural features, transforming flat grid data into rich spatial representations that maintain fidelity to the actual environment
3Productivity
If conventional area division methods are used in complex environments with many narrow areas, then the cleaning area can be segmented, but the segmentation points will not be matched where it is suitable, and there are many obstacles in one area making it not suitable for pattern driving with straightness
Solution Approach 1:
The patent changes the segmentation criterion from geometric features to depth parameter analysis. By examining depth values at pixel centers in the distance map and comparing them against threshold levels, the system identifies segmentation points that accurately correspond to spatial boundaries, ensuring reliable matching even in complex environments with narrow passages and multiple obstacles
Solution Approach 2:
The patent implements a dynamic segmentation approach where segmentation points are identified based on real-time depth map analysis rather than fixed geometric rules. The system adaptively determines segmentation locations by evaluating depth variations and threshold conditions, allowing flexible adjustment to different environmental configurations and obstacle arrangements
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 approach enables the mobile robot to effectively navigate and clean complex environments by accurately mapping and segmenting areas, minimizing uncleaned spaces and reducing the likelihood of collisions, thereby optimizing cleaning paths and improving user experience.
Implementation Method 1
an image acquiring unit configured to acquire an image outside the main body
Implementation Method 2
a sensing unit configured to sense a surrounding environment
Data Source
AI summary
A mobile robot of the present disclosure includes: a traveling unit configured to move a main body; a cleaning unit configured to perform a cleaning function; a sensing unit configured to sense a surrounding environment; an image acquiring unit configured to acquire an image outside the main body; and a controller configured to generate a distance map indicating distance information from an obstacle for a cleaning area based on information detected and the image through the sensing unit and the image acquiring unit, divide the cleaning area into a plurality of detailed areas according to the distance information of the distance map and control to perform cleaning independently for each of the detailed areas. Therefore, the area division is optimized for the mobile robot traveling in a straight line by dividing the area in a map showing a cleaning area.