Image Feature Prioritization for Accurate Self-Position Estimation
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing own-position estimating devices struggle to accurately determine the position of movable objects, such as cargoes, in logistics sites, leading to reduced accuracy in own-position estimation.
Innovation Solution
An own-position estimating device that matches features extracted from images with a database containing position information and features, where the device evaluates the matching eligibility of features and processes the database to prioritize features with high eligibility, thereby improving estimation accuracy.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Adaptability or versatility
If features of movable objects (e.g., cargo) are included in the database for own-position estimation, then the versatility of the estimation system is improved, but the measurement precision deteriorates due to low matching eligibility of movable object features
Solution Approach 1:
The patent segments features into two categories: movable object features (e.g., cargo) and stationary object features (e.g., shelves). By separating these feature types and applying different evaluation criteria, the system maintains versatility in tracking movable objects while ensuring high measurement precision through prioritization of stationary features with high matching eligibility.
Solution Approach 2:
The patent applies local quality by assigning different weights and evaluation standards to different feature types based on their characteristics. Stationary features receive higher priority for position estimation, while movable features are evaluated separately, allowing each feature type to contribute optimally to the overall estimation accuracy.
2Device complexity
If all features in the database are processed equally for matching, then the device complexity is reduced, but the measurement precision deteriorates due to inclusion of low-eligibility features
Solution Approach 1:
The patent performs preliminary evaluation of feature matching eligibility before the actual position estimation process. By pre-assessing and categorizing features based on their eligibility, the system prepares optimized data structures that enable efficient processing during runtime without sacrificing precision, thus balancing complexity and accuracy.
Solution Approach 2:
The patent changes the parameter of feature eligibility by introducing an evaluation mechanism that assigns different weights to features based on their matching reliability. This parameter transformation allows the system to prioritize high-eligibility features in position calculation, improving measurement precision while maintaining manageable device complexity through algorithmic optimization.
Data Source
AI summary
An own-position estimating device for estimating an own-position of a moving body by matching a feature extracted from an acquired image with a database in which position information and the feature are associated with each other in advance, includes an evaluation result acquiring unit acquiring an evaluation result obtained by evaluating matching eligibility of the feature in the database, and a processing unit processing the database on the basis of the evaluation result acquired by the evaluation result acquiring unit.