Mobile Robot Boundary-Line Mapping for Low-Feature Environments
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing moving robots face challenges in accurately generating maps of environments with few features, leading to inefficient searching and difficulty in distinguishing between searched and unsearched regions, particularly when using grid maps and feature point uncertainty.
Innovation Solution
A moving robot system that employs a combination of LiDAR sensors and image processing to create a cell-based grid map, selects an optimal boundary line using a cost function, and updates the map by planning a path to the selected boundary, thereby efficiently distinguishing between searched and unsearched regions and reducing unnecessary searching.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If feature points are extracted from images to create search paths, then self-location recognition accuracy is improved, but in environments with few features, uncertainty remains high leading to continuous searching of the same region
Solution Approach 1:
The patent introduces a boundary line extraction mechanism as an intermediary between feature point extraction and path planning. By extracting boundary lines from the grid map and using them as search targets, the system overcomes the limitation of feature point methods in low-feature environments. The boundary lines serve as reliable search targets that can be identified even when traditional feature points are scarce or uncertain
Solution Approach 2:
The patent changes the search target parameter from feature points (which have high uncertainty in low-feature environments) to boundary lines (which provide clear spatial delimiters). This parameter change transforms the search problem from finding uncertain feature points to following deterministic boundary lines, thereby improving both location recognition accuracy and search efficiency
2Device complexity
If grid maps are used to represent the environment, then the map structure is simple and easy to process, but the boundary between searched and unsearched regions cannot be accurately extracted
Solution Approach 1:
The patent extracts boundary lines from the grid map by identifying transitions between searched and unsearched regions. This extraction process isolates the critical boundary information from the complete grid map, creating a simplified search target that maintains accuracy while reducing complexity. The boundary line extraction selectively takes out only the necessary spatial delimiters needed for efficient searching
3Measurement precision
If the robot continuously searches for feature points to update its location, then location accuracy may be maintained, but time is wasted searching in already searched regions
Solution Approach 1:
The patent performs preliminary boundary line extraction and path planning before executing the search. By pre-identifying the optimal search path along boundary lines and determining the next best view in advance, the robot avoids time-wasting random searches. The preliminary path planning ensures that the robot only searches unsearched regions along efficient trajectories
Solution Approach 2:
The patent implements feedback by continuously updating the grid map with searched regions and using this updated information to recalculate boundary lines and paths. This feedback loop ensures that the robot adapts its search strategy based on current map knowledge, preventing redundant searches in already explored areas while maintaining accurate location tracking
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
The system enables accurate and efficient map creation by effectively identifying and navigating to the optimal boundary lines, reducing redundant searching and improving the robot's ability to recognize its environment and perform tasks.
Implementation Method 1
a sensor unit configured to create LiDAR data regarding an external geometry
Data Source
AI summary
A method for creating a map by a moving robot includes receiving sensor data regarding a distance to an external object through a distance measurement sensor, and creating a cell-based grid map based on the sensor data. Image processing is performed to distinguish between regions in the grid map and to create a boundary line between the regions. An optimal boundary line is selected from the one or more boundary lines, and a path to the optimal boundary line is planned. The grid map is updated while the moving robot is moving along the path such that the map may be automatically created.