ProgramMaster Logo
Conference Tools for 2022 AWS Professional Program
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools
About this Abstract
Meeting 2022 AWS Professional Program
Symposium 2022 AWS Professional Program
Presentation Title Machine Learning-based Process Characterization and Efficient Adaptive Control in Robotic Arc Welding
Author(s) Joseph Kershaw, Rui Yu, Yuming Zhang, Peng Wang
On-Site Speaker (Planned) Yuming Zhang
Abstract Scope Welding robots have significantly advanced the state of automotive welding by improving operational efficiency and precision. However, robotic welding is not fully autonomous yet. Still relying heavily on preprogrammed welding parameters, current welding robots cannot adaptively adjust parameters to compensate for task variability, material heterogeneity, and process uncertainty as skilled human welders do. The transition from preprogrammed to adaptive robotic welding requires the seamless integration of in-situ perception and process characterization, process-quality relationship quantification, closed-loop control, and robotic execution. This presentation presents an online adaptive control approach to robotic arc welding, where weld pool width is controlled in a closed-loop. From in-situ optical imaging, an efficient pixel-level image segmentation network is first developed to outline the pool area and estimate the pool width. Then based on a perceptron that describes the dependence of pool width on parameters (i.e., welding speed, current, torch angle), an advanced gradient descent algorithm is investigated to backpropagate the needed pool geometry changes to parameter adjustments based on individual parameters’ contributions to pool status change, while minimizing the settling time and steady-state error band. The main contribution of this work lies in the following two areas: Efficiency: in-situ images can be processed within milliseconds, thus enabling sub-second periodical process adjustment. Also, the control algorithm greatly reduces the settling time of process adjustment. Physical interpretation: no black-box mapping from pool state to action determination is involved, as compared to the RL approach. The outputs from the networks and controller can be readily interpreted and evaluated by humans. Experimental results indicate that the process can be adjusted within 7 adjustment periods to a less than 11% error band. Since weld pool status is a critical and measurable metric in all types of welding processes, choosing it as the control target enables direct physical interpretability and generalizability of the developed system. Assuming the welding path is known and preprogramed into the robotic motion path, the system can be readily transferred to other welding processes (e.g., gas metal arc welding) and application scenarios (e.g., curved welding). The developed system will be expanded by including more capabilities (e.g., sophisticated torch weave, real-time welding path calculation), and be also compared to manual welding as part of the future work.
Proceedings Inclusion? Undecided

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

2020 Thomas Medal Presentation: Keeping Pace with Change
Addressing Materials Challenges and Other Barriers to the Future of Additive Manufacturing
An AI-based Vision Methodology for Self-guided Seam Tracking in Gas Metal Arc Welding
An Investigation into the Effects of Stir Zone Chemistry on Fracture Toughness in Friction Stir Welded Pipeline Grade Steel
Analysis of Temperature and Velocity Fields in the GTAW Arc for Argon
Application of Polarity Switching Capacitor Discharge Welding to Aluminum Sheet Structures
Buried Arc GMAW for Single Pass Single Sided Erection Joints onboard Ships
Carbonitride Development and Heat Treatment Response of Additively Manufactured 17-4 PH Stainless Steel with Variations in Composition
Characterization of Dissimilar Materials 410 Martensite Stainless Steel and Mild Steel Component Produced by Wire Arc Additive Manufacturing
Deleterious Phase Avoidance in Additively Manufactured Functional Gradients through Path Planning
Developing an Automated Defect Detection TIG Welding Robot with a future Adaptive Implementation
Development of a Temperature-Dependent Material Property Database for DH36 Steel
Droplet Temperature in GMAW
Effect of Composition on Solidification Behavior and Resultant Microstructure in Dissimilar Electron Beam Welds between Commercially Pure Iron and Nickel
Energy Balance in Gas Metal Arc Welding
Evaluating the Carbide Precipitation Behavior During Short Term Tempering and its Influence on Impact Toughness.
Evaluation of Spatter Production with Deep Learning Algorithms
Evolution of Analytical Modelling Approaches for Resistance Spot Welding: A Historical Perspective
Fatigue Properties of Dissimilar Aluminum to Steel Welds Joined by Ultrasonic Interlayered Resistance Spot Welding Process
Hardness Prediction by Incorporating Heat Transfer and Molten Pool Fluid Flow in a Multi-pass, Multi-layer Weld for Onsite Repair of CSEF Grade 91 Steel
High Speed Video (HSV) and Synchronized Data Acquisition (DAQ) to Observe Welding Process Stability
How to Accurately Monitor the Weld Penetration from Dynamic Weld Pool Serial Images using CNN-LSTM Deep Learning Model?
Hybrid Manufacturing: Combining Additive Friction Stir Deposition, Metrology, and Machining
IIW Commission I Activities on Additive Manufacturing
Impact of Beam Deflection on Porosity during Keyhole Mode Laser Welding of Aluminum Alloys
In-situ Liquid Nitrogen Cryogenic Cooling for Interpass Control in High-Duty Cycle Wire Arc Additive Manufacturing of Large Components for Navy Applications
Influence of Microstructure on the Mechanism of Hydrogen-assisted Cracking in Dissimilar Metal Welds for Refinery Application
Integrated Modeling of Defect Formation during Deep Penetration Laser Welding of Creep Resistant Nickel Base Alloys
Investigation of Fe-10 wt.% Ni Steel Weld Metal Hydrogen Induced Cracking Susceptibility using the Gapped Bead on Plate (GBOP) Test
Large-Scale Hybrid Manufacturing using Wire Arc Additive Manufacturing
Machine Learning-based Process Characterization and Efficient Adaptive Control in Robotic Arc Welding
Material Characterization of Grade 91 Steel Welds using Micro-resolution Ultrasonic Imaging System.
Mechanical Design and Development of a Five Degree-of-Freedom TIG Welding Robot
Meta-Analysis of Fatigue Properties in Additively Manufactured 316L Austenitic Stainless Steel
Micro Cross Weld Tensile Testing of Dissimilar Metal Welds using Digital Image Correlation
Micro GTAW Applied to a Battery Pack for Racing Applications
Microstructure and Mechanical Properties of Electron Beam Additively Manufactured Ti-6Al-4V
Optimizing Productivity of Hyper Duplex Stainless Steels Overlay while Avoiding Sigma Phase Formation
Phase Transformation Behavior of Fe-10wt.% Ni Steel Weld Metal
Predicting Operation Windows for High-Frequency Induction Aluminum Tube Welding through Machine Learning
Process-Feature-Microstructure-Property Relationships for A9628 Directed Energy Deposition Additive Manufactured Steel Components
Real-Time Recognition of Arc Weld Pool Using Image Segmentation Network
Role of Standards in Welding Safety
Scaling Analysis of Thermal and Mechanical Process in Friction Stir Welding
The Effects of Post-Weld Processing on Friction Stir Welded Additive Manufactured AlSi10Mg
The Influence of Boride Phase Transformations on Heat-Affected Zone Liquation Cracking Susceptibility in Laser Welded 304L Stainless Steel
The Influence of Dynamic Behaviors Characteristics of Molten Pool on the Weld Formation during the High Speed Laser Welding
The International Institute of Welding: Strategic Directions for Welding and Joining Research and Industrial Applications
Use of Low Transition Temperature Steel Alloys In Welded Overlays for High Wear Applications
Welding Investigation of Wrought FeMnAl Steel
WeldVac – A Quiet, Clean Metal Removal System

Questions about ProgramMaster? Contact programming@programmaster.org