2021 Speakers
Cynthia Brosque
Comparative analysis of robotic and traditional construction methods
Cynthia is a Ph.D. Candidate in Civil and Environmental Engineering (CEE) at Stanford University. She is conducting research under the supervision of Martin Fischer (CEE-CIFE). Her research interests are Virtual Design and Construction (VDC) and Construction Robotics. She has a Master of Science in Civil Engineering (Stanford University - 2019) and an Architecture Degree (Universidad ORT Uruguay - 2016).
Robots have increased the safety, productivity, and quality of manufacturing. Recently, sensing, computing, and mapping technologies have started to enable the use of robots in unstructured environments like construction. As robotic construction methods are being prototyped and adopted on site, innovation leaders in construction must analyze the safety, productivity, quality, and cost impacts of the deployment of robots. The researchers gained access to engineering, planning, and production data for the first use of a concrete drilling robot on site. Compared to manual drilling for installation hangers, the robot achieved a 10%-time reduction, increased task ergonomics by cutting 98% of muscle strain work hours, and reduced rework from 5% to 3%. Our comparison pays particular attention to the three levers a project team has to influence project outcomes—the product, the organization, and the process—and found that decisions like implementing a building information model (BIM) at Level of Development (LOD) 400 facilitated the robot use. This study shares a careful analysis of the application of a drilling robot to offer insights into the applicability of robots on site. Second, we suggest key elements and procedures of a framework to compare robotic and traditional construction methods. Future research should establish the generality of this analysis framework.
Timothy Sandy
Why do robots get to have all of the fun? Using computer vision to enhance manual craft
Tim Sandy received his PhD from ETH in robotic systems in control in 2018. His work focused on robotic sensing and control systems for high-accuracy on site robotic building construction, including the development of the In situ Fabricator mobile construction robot. He is now spinning out a company called incon.ai, which aims to use computer vision, digital fabrication, and augmented reality to extend human building capabilities.
In this talk, we will look at how augmented reality can be used for construction at building scale. Motivated by work in robotic bricklaying, we will see how characteristics of robotic sensing systems can be integrated into an augmented reality system to enhance the skills of craftsmen and enable new digitally guided fabrication processes.
Moritz Bächer
On Designing Robotic Characters and Architectural-Scale Structures
Moritz Bächer is a Research Scientist and group leader at Disney Research. His research is concerned with the development of versatile differentiable simulators to tackle complex design, control, and characterization problems in robotics, computational fabrication, and robotic construction. Before joining Disney, he received a Ph.D. from the Harvard School of Engineering and Applied Sciences and graduated with a master's from ETH Zurich.
Sean Hanna
Representation and Design: AI as Creative Agent
Sean Hanna is Professor of Design Computing and Director of Research at the Bartlett School of Architecture, University College London, and a member of the UCL Space Syntax Laboratory. His research is primarily in developing computational methods for dealing with complexity in design and the built environment, including the comparative modelling of space, and the use of machine learning and optimisation techniques for the design and fabrication of structures.
Machine learning is valuable for discerning complex phenomena beyond the usual range of our intuition. The task of learning is typically one of optimisation, of fitting to samples or minimising error, which can result in models useful in design. But optimisation, the notion of an ideal target, or a correct model, is inherently imperfect and misses out on essential features of the way representations are actually used in design. With reference to examples of my work and that of colleagues and students over two decades, I suggest that understanding the latter is the basis for creativity in artificial intelligence.
JenJen Chung
Robots in human spaces
Jen Jen Chung is a Senior Researcher in the Autonomous Systems Lab (ASL) at ETH Zürich. Her current research interests include perception and learning for mobile manipulation, algorithms for robot navigation through crowds, informative path planning and adaptive sampling. Prior to working at ASL, Jen Jen was a postdoctoral scholar at Oregon State University researching multiagent learning methods and she completed her Ph.D. on information-based exploration-exploitation strategies for autonomous soaring platforms at the Australian Centre for Field Robotics in the University of Sydney. She received her Ph.D. (2014) and B.E. (2010) from the University of Sydney.
The tremendous potential offered by robotic automation has been evidenced by its revolution of large scale manufacturing, mining and agriculture. Now, with increasing social and economic pressures to transfer this success to service-focused sectors, we're seeing robots moving into our homes, onto our roads and into our airspace. These human-centred spaces heighten the challenges of robotic perception for scene understanding, and emphasise the need for robust planning and interaction in unstructured and dynamic environments. This talk will focus on how we are tackling these challenges in robotic mobile manipulation scenarios, i.e. tasks where the robot needs to both navigate collision-free as well as physically grasp and manipulate objects within its environment. We will discuss the importance of problem representation and how this impacts the system's ability to reason and adapt under uncertainty. Finally, we will take an outlook on how these techniques can be further developed to achieve robots that not only operate in our spaces but that are also capable of rich and intuitive collaboration with people.
Dr. Kian Wee Chen
The Use of Algorithms, Models and Data in the Design Process
Kian Wee Chen research interest lies in the development and use of digitals tools to support the integrative design process for the built environment. In his PhD research with the Future Cities Laboratory, Singapore-ETH Centre, he focused on developing an integrated workflow for using optimisation algorithms in designing a building’s form, envelope and cooling systems for better energy performance. After his PhD, he joined Singapore-MIT Alliance for Research and Technology, Center for Environment Sensing and Modeling (SMART-CENSAM) as a Postdoctoral Associate looking at adapting his integrated workflow for the urban design process. During his stay in SMART-CENSAM, he had the opportunity to work with researchers from the fields of urban climatology, remote sensing and landscape architecture, which provided valuable insights about the built environment that influenced his research. After this postdoctoral studies, he received the Distinguished Postdoctoral Fellowship from the Andlinger Center for Energy and the Environment, Princeton University, where he extended his research to include the use of low-cost sensors and Internet of Things (IoT) devices in the prototyping of building design with innovative cooling and heating systems. With his experience so far, Kian Wee plans to further his research of facilitating integrative design in multidisciplinary design teams through developing algorithms and modular tools. He aims to support multidisciplinary collaborations in the design teams by enabling timely access to data and models for analysing and designing the built environment.
Dr. Nikola Marincic
Unreasonably High Expectations of Mean-Centric Distributions: What does big data have to do with architecture and mastership?
Nikola Marinčić is a researcher investigating the relation between Architecture and Information Technology, especially the challenges Artificial Intelligence poses to the field. Initially, he graduated as an architect in Serbia and later obtained his Master of Advanced Studies degree at ETH Zurich, where he studied the philosophy of technology and computer programming. In 2017, he was awarded the ETH Medal of distinction for an outstanding doctoral dissertation on the topic of computational models in architecture. During his postgraduate studies, Nikola worked one year as a guest researcher at the Future Cities Laboratory, an interdisciplinary research program of the Singapore ETH Centre for Global Environmental Sustainability.
Nikola strives to illuminate the exceptional relevance of digital literacy today. He recognises its applicability as universal—beyond disciplines, different practices and topical expertise. In his monograph "Computational Models in Architecture," published by Birkhäuser/De Gruyter in 2019, Nikola celebrates abstract model-thinking by offering multiple perspectives on how digitally literate architects could reinvent their field in the digital age.
Prof. Dr. Karla Saldana
Enhancing Disaster Response with Architectonic Capabilities by Leveraging Machine and Human Intelligence Interplay
Karla is an Ecuadorian architect and researcher. In June 2021, she finished her Ph.D. at ETH Zurich in the Department of Architecture with Profesor Ludger Hovestatd. Her dissertation investigated the combination of Artificial and Human Intelligence to have a precise and agile response to natural disasters. Since August 2021, Karla is a Tenure Track Assistant Professor in the School of Architecture at the University of Florida; her teaching and research focus is applying Artificial Intelligence in architectural practices at building and urban scale.
Ardavan Bidgoli
Interfacing Machine Learning Tools for Creative Users
Ardavan Bidgoli is a computational designer, roboticist, and machine learning researcher. His research concerns the interfaces that enable designers and creative users interact with computational design tools. He is currently pursuing his Ph.D. in Computational Design at the Carnegie Mellon School of Architecture, where he studies the affordances of a situated approach to machine learning in creative computing toolmaking for domain expert users.
He is also the lead instructor of the Inquiries into Machine Learning and Design course at CMU School of Architecture. His industry partnerships span over various research and product teams at AEC-Tech companies, including the Generative Components team at Bentley Systems as well as Autodesk’s OCTO team at Pier 9 and BUILD space facilities. Ardavan holds a Bachelor of Architecture and a Master of Architecture from the University of Tehran, and a Master of Architecture in Design Computing from The Pennsylvania State University.
Prof. Dr. Axel Kilian
Embodied Computation
Axel Kilian is currently a Visiting Assistant Professor at the MIT Department of Architecture. He previously was an Assistant Professor at the Princeton University School of Architecture and at the Delft University of Technology and a Postdoctoral Associate at the Department of Architecture at MIT. He holds a PhD in Design and Computation and a Master of Science in Architectural Studies from the Department of Architecture at MIT. He came to MIT as a German American Fulbright scholarship grantee after completing a professional degree in architecture at the University of the Arts Berlin. His work in architectural robotics has been exhibited at the Istanbul Design Biennial and the Seoul Biennial of Architecture and Urbanism. His current research and teaching focus is on embodied computation, exploring the extension of architecture's material form into the behavioral through physical, actuated, and sensing prototypes of space.
Dr. Ole Ohlbrock
Creativity in computational structural design?
Ole holds a degree in Civil Engineering since September 2013. He studied Civil Engineering with the minor subject Architecture at the Technical University of Munich from 2007 to 2013. Between his undergraduate and master's degree, he worked as an intern at Schlaich Bergermann and Partner, each half a year in Stuttgart and New York. In 2020, he obtained a PhD with distinction from ETH Zurich under the supervision of Prof. Dr. Joseph Schwartz.
He is currently a Postdoctoral Researcher and Lecturer at the Chair of Structural Design at the Swiss Federal Institute of Technology (ETH) in Zurich. His research focuses on the development of computational tools for the conceptual design phase. His research interests are geometry-based methods for structural design, computational modelling of strut-and-tie models, renewable building materials and structural form-finding with machine learning.