Crows and Zebras – Constraint-based Modeling and AI-Aided Visualization in Architecture
Benjamin will present the newly developed neuro-physics engine Crow, that combines paradigms from artificial intelligence and physics simulation to serve as an intuitive tool for constraint-based computational modeling. Furthermore he will discuss the potential of Generative Adversarial Networks (GANs) as an extended, fast method for rendering realistic architectural visualizations from simple 3D models.
About Benjamin Felbrich He is a PhD Candidate at the Institute for Computational Design and Construction (ICD) at the University of Stuttgart, Germany. During his studies of architecture at Dresden University of Technology, Germany, Benjamin spent an extended period working for Pritzker Price laureate B.V. Doshi in Ahmedabad, India where he developed a deep interest in parametric design and computational form finding. After another work stay at Behnisch Architekten in Stuttgart, Germany, he graduated with an award-winning scientific diploma thesis in which he developed a folding robotic multi-agent system. Since joining ICD in 2014, he worked on bio-inspired robotic 3D-printing, multi-machine fabrication of a large scale CFRP structure through autonomously collaborating robots and drones, physics simulation in CAD Software and artificial intelligence. Since 2016 he has been developing the physics simulation engine FlexHopper and the GPU-based AI engine Crow. Benjamin has a profound interest in high performance computation for design applications, robotic fabrication systems and machine communication infrastructures. His PhD thesis is dedicated to machine learning tools as drivers for integrative design methods in architectural fabrication.
About Aldo Sollazzo He is an architect and researcher, expert in computational design and digital fabrication. Master in Architectonic Design in 2007, Master in Advanced Architecture at IAAC in 2012, Fab Academy diploma in 2014 in the Fab Lab Barcelona, Aldo is currently involved in several projects running in parallel. Since 2011, he is the manager of Noumena, a firm embracing data-driven design, investigating between the boundaries of new digital paradigms and design strategies applied to architecture, robotics and fabrication, through a hands-on and experimental approach. He is also the founder of Fab Lab Frosinone and Director of Reshape, a digital platform promoting interdisciplinary and collaborative approach, towards the definition of new cutting-edge practices and disruptive ideas in the field of wearable tech and fashion design. Since 2015 he is faculty at IAAC, covering the positions of Director of IAAC Global Summer School, co-Director of the new Master of Robotics and Advanced Construction, and instructor of digital tools. He is one of the Supernode of Fab Academy, the educational platform founded by Neil Gershenfeld from the MIT’s Center of Bits and Atoms. For the same programme, he is also a local coordinator for Fab Academy Paris.
Application of Machine Learning within the integrative design and fabrication of robotic rod bending processes.
Presentation will focus on potentials and methodology of using Machine Learning algorithms for precision control of material deformation and increased geometrical and structural performances in robotic rod bending technology, along with integrative methods where design, material properties analysis, structural analysis, optimization and fabrication of robotically rod bended space-frames are merged into one coherent data model and allows for bi-directional information flows, shifting from absolute dimensional architectural descriptions towards the definition of relational systems.
About Maria Śmigielska She is an architect, researcher, designer and educator with a strong interest in technology, based in Zurich, Switzerland. Graduated from Architecture department at TU Poznan (Poland, 2009) and the postgraduate program at the chair for Computer Aided Architectural Design at ETH Zurich (Switzerland, 2013). Explores varied computational design and develops robotic procedures for post-industrial fabrication and performative processes (. f.ex. robotic rod bending http://bendilicious.com/. Currently a researcher at Creative Robotics at the University for Arts and Design, Linz focused on merging robotic processes with machine learning and heads her design practice in Zurich. mariasni.com
- Mary Katherine Heinrich CITA Copenhagen
- Ryan Hughes Axis Robotics Aarhus
- Paul Nicholas CITA Copenhagen
- Ilija Vukorep BTU C-S Cottbus
- Mateusz Zwierzycki BTU C-S Cottbus
More info about the speakers and topics soon...