Fluorescent Markings for Robotic Item Recognition

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current robotic technologies in materials handling facilities face challenges in automatically recognizing items and their orientations using computer vision, especially when items are stored in totes, bins, or shelves, due to insufficient contrast under normal lighting conditions.

Innovation Solution

The implementation of fluorescent markings on items that emit visible light under ultraviolet radiation, creating high contrast with their surroundings, allowing computer vision algorithms to accurately identify and manipulate items using robotic arms, even in dim or UV-lit environments.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If computer vision technology is used to recognize items under normal lighting conditions, then automated item recognition can be achieved, but recognition accuracy is insufficient due to low contrast between items and surroundings

Engineering Contradiction:
Improveitem recognition accuracyVSAvoidcontrast between items and surroundings
Core Design Contradiction:
Measurement precisionVSDifficulty of detecting and measuring

Solution Approach 1:

The patent applies fluorescent markings that change color properties under UV illumination. Items coated with fluorescent material appear bright and highly visible under UV light, creating strong contrast against dark backgrounds. This color transformation enables computer vision systems to easily distinguish items from surroundings, directly resolving the low contrast detection problem.

Inventive Principle:
Principle #32Color changes

Solution Approach 2:

The patent changes the illumination parameter from normal visible light to ultraviolet light. This parameter change activates the fluorescent markings on items, transforming them from invisible or low-contrast states to high-visibility states. The UV illumination parameter modification enables the computer vision system to detect items with high accuracy by creating optimal contrast conditions.

Inventive Principle:
Principle #35Parameter changes

2Measurement precision

If fluorescent markings are applied to items, then item visibility and contrast under UV light are improved, but additional processing steps and materials are required

Engineering Contradiction:
Improveitem recognition accuracyVSAvoidmarking application system
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent applies fluorescent markings to items during the receiving process at the beginning of their lifecycle in the fulfillment center. This preliminary action ensures that all items are pre-marked before being stored or handled, eliminating the need for later marking operations. The markings remain on items throughout their lifecycle, providing continuous visibility benefits without adding complexity to subsequent handling operations.

Inventive Principle:
Principle #10Preliminary action

Applied Scientific Principles

This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.

Function Achieved in This Case

Enhances the ability of computer vision systems to recognize items and control robotic arms with improved accuracy, facilitating automated item retrieval and storage processes in materials handling facilities.

Implementation Method 1

fluorescent markings on items that emit visible light under ultraviolet radiation

Methodology Applied
Scientific EffectFluorescence: Fluorescence

Data Source

PatentUS10442087B1Applying fluorescent markings to items to aid robotic handling
Publication Date: 2019.10.15 AMAZON TECH INC
  • US10442087B1 patent drawing
  • US10442087B1 patent drawing
  • US10442087B1 patent drawing

AI summary

Disclosed are various embodiments for using fluorescent coatings to aid in robotic handling of items. In one embodiment, an image is captured. An item in the image is recognized from the image based at least in part on a fluorescence of the item relative to its surrounding environment caused by exposure of a fluorescent marking of the item to a radiation source. A robotic arm is operated relative to the item in response to recognizing the item.