RESOURCES

Publications

European Energy Innovation Magazine

AM2PM project has been featured in the Winter edition of the European Energy Innovation Magazine! This renowned publication highlights cutting-edge advancements in energy, sustainability, and innovative technologies across Europe. Our article dives into how our project is shaping the future of construction. Don’t miss out - read it now and explore how AM2PM is driving innovation!

An aerial photograph of a building with a unique, rounded, ribbed design resembling large, curved, wooden shells, with a crane nearby.

Data-Driven Parameter Calibration in Additive Manufacturing

The researchers from TUM introduce the Learning by Printing framework, a novel approach designed to enhance performance and robustness in extrusion-based additive manufacturing for construction. Built on the foundation of Fabrication Information Modelling (FIM), the framework combines evaluation, prediction, and calibration processes into a unified, closed-loop system for fabrication learning.

Methods for Earth-Based Falsework in Multi-Material Additive Manufacturing

Researchers from Technion and Hebrew University introduce a novel large-scale 3D printing method that integrates cementitious structures with reusable earth-based falsework. By optimizing nozzle angles, deposition strategies, and sequential printing, the study demonstrates how multi-material additive manufacturing can reduce waste, enhance precision, and enable complex architectural forms.

LLM-Driven Deployment of ROS2 Modules via Semantic Affordance Discovery in a Federated Digital Twin

Researchers from Eindhoven University of Technology explore how large language models can dynamically discover and use system capabilities within federated Digital Twin environments. Using semantic affordances and ROS2-based modules, the study demonstrates an approach to AI-driven orchestration of distributed digital systems, enabling multi-step deployment workflows through natural-language instructions.

Towards a Multi-Domain Semantic Architecture for Graph-Based Orchestration of Decentralised Additive Manufacturing

Researchers from Eindhoven University of Technology present a multi-domain semantic architecture for integrating and orchestrating decentralised additive manufacturing workflows. Using an AM2PM concrete 3D printing use case, the study demonstrates how distributed software modules can be connected through a shared Digital Twin interface, enabling better coordination and end-to-end traceability of information and process outputs.

IFC2LBD-Neo: A Rust-Based Converter for Efficient Linked Building Data Generation from IFC

Researchers from Eindhoven University of Technology and RWTH Aachen University introduce IFC2LBD-Neo, a Rust-based converter for generating Linked Building Data (LBD) and linked ifcOWL from IFC models. Designed for data-intensive BIM and digital twin applications, the converter enables faster and more scalable processing, achieving up to 20× faster runtime and more than 8× lower peak memory consumption, while using a staged topology pipeline to improve topology completeness.