Mobile robot control method
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current methods for mobile robots to recognize their location in a traveling zone face accuracy issues due to environmental changes such as lighting conditions and object movements, especially in low-illuminance environments, which affect the reliability of location recognition and map creation.
Innovation Solution
The implementation of block-based feature extraction and matching technology, where images are divided into blocks, and feature points are extracted and matched within these blocks, allowing for robust location recognition and map creation even in low-illuminance conditions by using descriptors based on brightness gradient distribution characteristics.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If point-based feature point extraction and matching is used for location recognition, then the method is simple and fast, but accuracy deteriorates in low-illuminance environments and under lighting changes
Solution Approach 1:
The patent divides the image into multiple blocks and performs feature extraction and matching within each block separately. This segmentation approach allows the system to maintain computational efficiency while improving robustness to lighting changes, as each block can be processed independently with localized feature descriptors that are less sensitive to global illuminance variations.
Solution Approach 2:
The patent transitions from point-based feature representation to block-based feature representation, adding a spatial dimension to the feature extraction process. By considering features within block regions rather than isolated points, the system gains additional contextual information that improves location recognition accuracy under varying lighting conditions while maintaining computational feasibility.
2Measurement precision
If block-based feature extraction is used to improve location recognition accuracy in low-illuminance environments, then measurement precision improves, but device complexity increases
Solution Approach 1:
By segmenting the image into blocks and performing feature extraction within each block, the patent reduces the complexity of finding reliable point-based features in low-illuminance environments. The segmentation creates localized regions with more consistent lighting conditions, making feature extraction simpler and more reliable while maintaining overall system accuracy.
3Productivity
If traditional feature point matching is used, then the system operates quickly, but reliability deteriorates when environmental conditions change such as lighting and object location
Solution Approach 1:
The patent segments the image into blocks and performs feature matching within each block, which improves reliability under environmental changes. This segmentation creates localized feature descriptors that are more robust to global lighting changes and object movements, while the block-based approach maintains processing efficiency by limiting the search space for feature matching.
Solution Approach 2:
The patent changes the fundamental parameter of feature representation from point-based to block-based descriptors. This parameter change makes the feature extraction more sensitive to local structural patterns rather than global lighting conditions, thereby improving reliability under varying environmental conditions while maintaining computational efficiency through the structured block approach.
Data Source
AI summary
A mobile robot and a method of controlling the mobile robot are disclosed. The method includes acquiring an image of an inside of a traveling zone. The method further includes performing a point-based feature point extraction by extracting a first feature point from the acquired image. The method also includes performing a block-based feature point extraction by dividing the acquired image into blocks having a predetermined size and extracting a second feature point from each of the divided block-unit images. The method also includes determining the current location by performing a point-based feature point matching using the first feature point and performing a block-based feature point using the second feature point. The method also includes storing the determined current location in association with the first feature point and the second feature point in a map.