AR Welding Guidance Using Expert Data on Complex Weld Paths
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Beginner welders face difficulties in performing high-quality welds on complex-shaped base materials due to the need to change welding postures, angles, speed, and arc length, leading to potential arc loss and defects.
Innovation Solution
A real-time welding variable collection device tracks expert welding behavior using IMU sensors and cameras, recording work angle, travel angle, speed, and arc length, which are then synchronized and displayed on an AR device for beginner welders to improve their technique.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Manufacturing precision
If constant welding parameters (work angle, travel angle, speed, arc length) are used, then welding quality is high on simple geometries, but welding becomes difficult on complex-shaped base materials requiring posture changes
Solution Approach 1:
The system captures expert welder behavior through IMU sensors and cameras, recording work angle, travel angle, speed, and arc length. This behavioral data is then copied and provided to beginner welders via AR devices, allowing them to replicate expert techniques without years of training experience
Solution Approach 2:
The system provides real-time feedback to beginner welders by displaying expert behavioral variables through AR devices during welding operations. This allows beginners to see what parameters an expert would use in their current position and adjust their technique accordingly
2Manufacturing precision
If welding parameters are changed according to base material shape, then welding quality on complex shapes improves, but the complexity of controlling multiple variables increases
Solution Approach 1:
The system automatically captures and processes expert welding behavior data without requiring manual input or complex control systems. The IMU sensors and cameras self-record the expert's natural welding performance, and the system automatically processes this data into actionable guidance for beginners
Solution Approach 2:
The system pre-captures and stores expert welding behavioral data across various positions and geometries before actual welding operations. This preliminary data collection creates a library of expert techniques that can be instantly retrieved and displayed to beginners during welding, eliminating the need for real-time complex calculations
3Loss of information
If real-time tracking of expert welding behavior is implemented, then beginner welders gain access to valuable behavioral variables, but the system complexity and data processing requirements increase
Solution Approach 1:
The system replaces complex manual analysis and teaching methods with automated sensor-based data collection. IMU sensors and cameras automatically capture welding behavior, replacing the need for manual observation and interpretation of expert techniques
Solution Approach 2:
The system uses multi-functional devices that serve multiple purposes: IMU sensors track both position and orientation, cameras capture both visual information and arc characteristics, and the AR device provides both guidance and feedback. This consolidation reduces overall system complexity while maintaining comprehensive data collection
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 solution enables beginner welders to improve their welding quality by referencing expert behavioral variables in real-time, reducing defects and enhancing productivity and safety while welding complex shapes.
Implementation Method 1
an IMU sensor installed in a welding torch of the expert extracts data about a work angle and travel angle of welding performed by the expert
Implementation Method 2
a welding image is acquired using a camera of a welding helmet worn by a user, and a welding speed and a contact tip to work piece distance (CTWD) are acquired by extracting arc light and a contact tip from the acquired image
Data Source
AI summary
An expert navigation device based on actual welding data of an embodiment includes: a welding variable acquisition unit that acquires welding variable data of an expert through a bigdata platform; a welding line extraction unit that analyzes a base material, which is a welding target, through a camera mounted on a welding helmet to acquire the welding line in the base material; a synchronization unit that generates a trigger, which serves as a reference point, for matching the welding variable data acquired through the welding variable acquisition unit to the welding line acquired through the welding line extraction unit; and a navigation output unit that displays information of the welding variable data to a user when a welding location reaches a trigger point while welding is performed based on the trigger point generated by the synchronization unit.