Centroid Follower for AGV Ellipsoidal Shape Detection
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing methods for detecting ellipsoidal shapes in real images, such as those used in Automated Guide Vehicles (AGVs), face challenges with perspective distortion, blurriness, and noise, leading to inaccurate geometric estimation and positioning.
Innovation Solution
The proposed solution involves a centroid follower method that evaluates the exact center of a circular finder pattern by analyzing grey level profiles in four directions, rotating the axes by small increments, and calculating distances to determine the major axis, allowing for precise geometric estimation even in distorted or blurred images.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Difficulty of detecting and measuring
If edge-based methods are used to detect ellipsoidal shapes, then shape detection capability is provided, but measurement precision deteriorates due to weak edge acquisition, noise, and computational intensity
Solution Approach 1:
The patent segments the ellipsoidal shape detection into multiple processing stages: initial ellipse detection, centroid calculation, profile extraction in multiple directions, and iterative refinement. This segmentation allows each stage to focus on specific aspects, improving overall measurement precision while maintaining detection capability.
Solution Approach 2:
The patent performs preliminary actions by first detecting the ellipse and calculating its centroid before proceeding to profile analysis. This preliminary positioning enables subsequent measurements to be performed from an accurate reference point, reducing errors from weak edges and noise.
2Productivity
If conventional ellipse detection methods are used, then detection speed is maintained for real-time applications, but manufacturing precision deteriorates due to perspective distortion and skew effects
Solution Approach 1:
The patent employs dynamic iterative refinement where the centroid is recalculated and profiles are re-evaluated in multiple passes. This dynamic approach allows the system to converge to more accurate axis evaluations while maintaining real-time performance through optimized computation.
Solution Approach 2:
The patent transitions from 2D edge-based detection to analyzing 1D grey level profiles extracted from the 2D image space. By projecting the ellipse onto multiple 1D lines passing through the centroid, the method achieves higher precision in axis evaluation while maintaining computational efficiency.
3Reliability
If multiple scanning directions are used to improve pattern recognition, then detection reliability is improved, but device complexity increases due to multiple comparison steps
Solution Approach 1:
The patent uses a universal centroid calculation approach that works for both circular and elliptical patterns. The same centroid follower algorithm is applied regardless of the specific pattern type, reducing overall system complexity while maintaining high reliability across different detection scenarios.
Solution Approach 2:
The centroid calculation serves multiple functions simultaneously: it provides the center reference for profile extraction, enables axis orientation determination, and serves as a validation point for ellipse detection. This self-service approach reduces the need for separate processing steps, lowering algorithmic complexity while improving reliability.
Data Source
AI summary
An automated guided vehicle (AGV) has an image reader that scans indicia located on a floor. The image reader scan the indicium to obtain instructions for the AGV to follow. A processor in the AGV can perform geometry estimation of ellipsoidal objects in order to align the AGV as it moves or is stopped along a path.