Factory Location Tracking Using Visual Odometry and Reference Markers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional location tracking methods in factories, such as those using Impulse Radio, Ultra-wideband, and Time Difference of Arrival, face accuracy issues in environments with abundant structures, leading to high infrastructure and maintenance costs due to the need for extensive deployment of Access Points.

Innovation Solution

A location tracking system that employs an image collecting device attached to tracking targets, utilizing Visual Odometry to recognize absolute marker locations and estimate relative positions based on these markers, thereby reducing the reliance on Access Points and improving accuracy through marker-based coordinate systems.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Measurement precision

If conventional location tracking methods (Impulse Radio, Ultra-wideband, TDoA) are used in factories with abundant structures, then location tracking can be implemented, but measurement precision deteriorates due to multipath radio signal interference

Engineering Contradiction:
Improvelocation tracking accuracyVSAvoidmultipath radio signal interference
Core Design Contradiction:
Measurement precisionVSObject-affected harmful factors

Solution Approach 1:

The patent replaces radio-based location tracking systems with a vision-based system using cameras and Visual Odometry algorithms. This substitution eliminates the harmful effect of multipath radio signal interference by using optical fields instead of radio fields, which are not affected by the same interference mechanisms in structurally complex factory environments.

Inventive Principle:
Principle #28Mechanics substitution (Replace mechanical system)

Solution Approach 2:

The patent changes the fundamental parameter of the tracking system from radio frequency signals to visual image data. By capturing images of markers through cameras and processing them via Visual Odometry, the system transforms the measurement domain from electromagnetic radio waves to optical images, thereby avoiding multipath interference inherent in radio signal-based methods.

Inventive Principle:
Principle #35Parameter changes

2Reliability

If a large number of Access Points are deployed throughout the factory for location tracking, then location coverage is improved, but device complexity and infrastructure costs increase

Engineering Contradiction:
Improvelocation tracking coverageVSAvoidAccess Point deployment quantity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

Instead of deploying multiple fixed Access Points throughout the factory to track moving targets, the patent inverts the approach by attaching the imaging device to the moving target itself. The target becomes the active element carrying the camera and markers remain passive reference points fixed in the environment, thereby reducing infrastructure complexity while maintaining tracking capability.

Inventive Principle:
Principle #13The other way round (Inversion)

Solution Approach 2:

The tracking system enables self-service by equipping each tracked target with its own imaging device and computational resources. Each target independently captures images, processes them through Visual Odometry algorithms, and determines its own location, eliminating the need for a dense network of centralized Access Points and reducing overall system complexity.

Inventive Principle:
Principle #25Self-service

3Measurement precision

If Visual Odometry with markers is used for location tracking, then measurement precision is improved, but device complexity increases due to image collecting device requirements

Engineering Contradiction:
Improvelocation tracking accuracyVSAvoidimage collecting device complexity
Core Design Contradiction:
Measurement precisionVSDevice complexity

Solution Approach 1:

The patent segments the location tracking system into two independent components: simple passive markers fixed in the environment and an active image collecting device attached to the target. This segmentation allows the complex processing functions (Visual Odometry algorithms, image analysis) to be concentrated in the moving target's device while the environmental infrastructure remains simple and low-cost.

Inventive Principle:
Principle #1Segmentation

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

This approach enhances location tracking accuracy and reduces infrastructure costs by eliminating the need for extensive Access Point deployment, while allowing for flexible marker arrangement, thereby improving tracking precision and simplifying factory layout design.

Implementation Method 1

recognize an absolute location of a reference marker detected by a visual odometry (VO) method in an image captured by a camera

Methodology Applied
Scientific EffectVisual Odometry:

Implementation Method 2

an Inertial Measurement Unit (IMU) configured to derive an inertia value including an acceleration and a moving direction of the target

Methodology Applied
Scientific EffectInertial measurement: Inertia

Data Source

PatentUS20240012394A1Location tracking system in factory and method thereof
Publication Date: 2024.01.11 HYUNDAI MOTOR CO LTD
  • US20240012394A1 patent drawing
  • US20240012394A1 patent drawing
  • US20240012394A1 patent drawing

AI summary

A location tracking system includes a plurality of markers disposed on a fixture in an area having an absolute coordinate in the factory, an image collecting device fixedly attached to a tracking target for management, configured to recognize an absolute location of a reference marker detected by a visual odometry (VO) method in an image captured by a camera, and estimate a relative location based on the absolute location of the reference marker; and a server configured to track a location of the target based on the relative location received through a wireless communication from the image collecting device.