3D Point-Cloud Self-Localization Using Geometric Feature Sets
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing self-localization methods for mobile bodies in environments with large amounts of three-dimensional point group data, such as outdoor settings, face significant processing time challenges due to the high number of data points.
Innovation Solution
A self-localization device that generates and utilizes geometric features like planes, straight lines, and spheres to estimate location, reducing processing time by selecting and evaluating sets of geometric features rather than individual three-dimensional points, and employing sphere groups to enhance coincidence evaluation.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If the mobile body estimates self-location by matching a plurality of three-dimensional points, then the estimation accuracy is improved, but the processing time becomes excessively long in environments with large amounts of three-dimensional point data
Solution Approach 1:
The patent segments the large set of three-dimensional points into smaller subsets based on distance ranges (short-distance points and long-distance points). This segmentation reduces the computational burden by processing smaller point sets separately, thereby decreasing processing time while maintaining estimation accuracy through multi-range feature matching.
Solution Approach 2:
The patent extracts feature points from the three-dimensional point cloud data based on specific feature amounts (distance ranges). By extracting only the most relevant feature points rather than processing all points, the system reduces the number of points to be matched, significantly decreasing processing time while preserving the essential geometric features needed for accurate self-location estimation.
2Measurement precision
If the mobile body processes all three-dimensional points to estimate self-location, then the estimation accuracy is improved, but the computational complexity increases significantly
Solution Approach 1:
The patent divides the three-dimensional point cloud into multiple distance ranges, creating separate point sets for short-distance and long-distance features. This segmentation simplifies the computational complexity by reducing the size of each point set that needs to be processed, while the combination of multiple segmented results maintains overall estimation accuracy.
Solution Approach 2:
The patent extracts only the necessary feature points from the complete three-dimensional data based on predetermined feature amount thresholds. This extraction process eliminates unnecessary data processing, significantly reducing computational complexity while retaining the critical geometric features required for accurate self-location estimation.
Data Source
AI summary
A self-localization device that can reduce the processing time even when estimating the self-location in an environment where the number of pieces of three-dimensional point group data is large wherein the self-localization device includes an estimation unit that includes: a map generation unit that generates a map of surroundings of the mobile robot based on three-dimensional point group data detected by a detection unit; a geometric feature extraction unit that extracts geometric features from a current map and extracts geometric features from a past map; a self-location calculation unit that selects the geometric features extracted by the geometric feature extraction unit as sets of geometric features and calculates self-locations in the past map; and a self-location evaluation unit that evaluates a degree of coincidence between the current map and the past map for each set of geometric features.