Vehicle Positioning Using Marker-Corrected Visual SLAM
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In dynamic environments like factories and warehouses, existing positioning systems using Visual-SLAM struggle with cumulative errors in determining the current position of moving bodies due to varying backgrounds, leading to inaccurate position measurement over time.
Innovation Solution
A positioning apparatus that utilizes an image capturing system to extract feature points and markers, correcting relative positions and attitudes based on absolute positions and attitudes, and adjusts the relative attitude calculation by referencing the direction of passageways instead of absolute attitudes when the vehicle is traveling along them, thereby reducing error.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Device complexity
If Visual-SLAM is used to measure position of a moving body, then the position can be measured without requiring a large number of wireless transmitters, but cumulative errors increase with lapse of time leading to reduced measurement precision
Solution Approach 1:
The patent applies feedback by using detected markers as reference points to correct and reset cumulative position errors. The system continuously monitors position deviations relative to known marker locations and adjusts the position calculation to eliminate accumulated errors, thereby maintaining measurement precision over time without requiring complex infrastructure.
Solution Approach 2:
Markers serve as intermediary reference objects that bridge the visual SLAM system and the actual position space. These markers provide known reference points that allow the system to calibrate and correct its position estimates, acting as mediators between the visual data and the true spatial coordinates to maintain accuracy.
2Measurement precision
If markers are disposed at predetermined positions to correct position errors, then measurement precision is improved, but device complexity increases due to additional marker management requirements
Solution Approach 1:
The system employs self-service by enabling the moving body to autonomously detect and utilize markers in its environment without external intervention. The onboard imaging device automatically captures images, identifies markers, and performs position corrections, eliminating the need for complex external marker management infrastructure while maintaining measurement precision.
3Measurement precision
If absolute position and attitude are used for correction, then position measurement accuracy is improved, but calculation complexity increases
Solution Approach 1:
The patent extracts and utilizes only the essential correction information from marker detections - specifically the position and attitude relative to known marker locations. By extracting only the necessary correction parameters rather than processing complete absolute position data, the system reduces calculation complexity while maintaining the precision benefits of absolute reference points.
Data Source
Figure 1
Figure 2~3
Figure 4~6
AI summary
A first calculator calculates a relative position and a relative attitude of a vehicle based on a plurality of images captured by an image capturing apparatus mounted on the vehicle. A second calculator extracts one marker from an image captured by the image capturing apparatus, and calculates an absolute position and an absolute attitude of the vehicle based on a position and an attitude of the one extracted marker. A corrector corrects the relative position and the relative attitude based on the absolute position and the absolute attitude to generate a corrected position and a corrected attitude of the vehicle. The corrector determines whether or not the vehicle is turning, and generates the corrected attitude not using the absolute attitude calculated during a period in which the vehicle is turning, but using only the absolute attitude calculated during a period in which the vehicle is not turning.