Thesis topics
Adaptive Manufacturing Process Optimization with Artificial Intelligence in 5-axis 3D Printing
title
Adaptive Manufacturing Process Optimization with Artificial Intelligence in 5-axis 3D Printing
institution
doctoral_school
supervisor
co-supervisor
discipline
description
This research is focused on the development of methods based on Artificial Intelligence (AI) and Reinforcement Learning (RL) for the automation and optimization of 5-axis additive manufacturing. Multi-axis manufacturing offers significant advantages over conventional layer-by-layer printing, including the elimination of support structures, reduced post-processing, and enhanced mechanical properties. To realize this potential, complex, automated algorithms are required for adaptive process control.
The research is defined by three primary objectives: (1) Sensor-Based Process Control: An AI framework, utilizing methods such as reinforcement learning, will be developed to process data from on-printer sensors. This will enable real-time interventions in the manufacturing process, through which key parameters such as printing speed, extrusion rate, cooling intensity, and tool orientation are dynamically optimized. (2) Adaptive Error Handling and Quality Assurance: For multi-axis machining, the developed AI model will be enabled to perform early detection of typical manufacturing anomalies (e.g., delamination, warping, clogging). Automatic corrective movements will be executed by the system to increase process reliability and the quality of the final product. (3) Real-time, Adaptive Surface Recognition and Tracking: The printer will be equipped with 3D scanning systems, allowing an unknown surface geometry to be mapped in real-time. A conformal toolpath will then be immediately generated from the scanned data by the AI. This capability is intended to be used for the repair of damaged parts or for adding functional features to existing complex objects.
Relevant literature: https://doi.org/10.1016/j.jmapro.2024.04.084
The research is defined by three primary objectives: (1) Sensor-Based Process Control: An AI framework, utilizing methods such as reinforcement learning, will be developed to process data from on-printer sensors. This will enable real-time interventions in the manufacturing process, through which key parameters such as printing speed, extrusion rate, cooling intensity, and tool orientation are dynamically optimized. (2) Adaptive Error Handling and Quality Assurance: For multi-axis machining, the developed AI model will be enabled to perform early detection of typical manufacturing anomalies (e.g., delamination, warping, clogging). Automatic corrective movements will be executed by the system to increase process reliability and the quality of the final product. (3) Real-time, Adaptive Surface Recognition and Tracking: The printer will be equipped with 3D scanning systems, allowing an unknown surface geometry to be mapped in real-time. A conformal toolpath will then be immediately generated from the scanned data by the AI. This capability is intended to be used for the repair of damaged parts or for adding functional features to existing complex objects.
Relevant literature: https://doi.org/10.1016/j.jmapro.2024.04.084
student count limit
2
location
Szombathely, Budapest
deadline
2026-05-31
required language
English
additional requirements
The following skills are needed:
Hands-on experience with 3D printers and an understanding of their hardware.
Experience with CAD modelling
Optimization techniques
Programming knowledge (e.g. C++, Python)
Openness to learning about physical backgrounds