The computational tools that architects use to design buildings often provide static representations of built environments, which do not account for the people who will inhabit the spaces and their dynamic activities. As a result, a significant gap exists between expected and actual building performance, which may lead to occupant dissatisfaction, operational inefficiency, increased costs of operation, and substantial impact on the natural environment.
In this Research Committee talk, we will discuss our latest developments to develop dynamic representations of buildings in use, prior to their construction and occupancy with a mission to help architects analyze human-building interactions during the design phase – when design issues can be discovered and solved in a timely fashion. In particular, we will discuss computational methods to model and simulate human behavior patterns in not-yet-built environments using autonomous virtual agents that can perform individual and collaborative tasks in semantically rich environments. Integrated with established architectural design tools, human behavior simulation supports spatiotemporal analyses of building-human interactions, human-in-the-loop design explorations, and automatic refinement of building designs with respect to human-related performance criteria.
Future work involves the systematic extrapolation of virtual agents’ behavior rules using VR experiments and sensor-based information, the development of validation and calibration protocols against real-world data, and the design of novel interfaces to benchmark simulation results and explore vast solution spaces that comprise building design alternatives as well as different behavior scenarios of day-to-day occupancy and emergency egress situations.
Davide Schaumann is a Postdoctoral Fellow at the Jacobs Technion-Cornell Institute at Cornell Tech. He recently completed a Postdoc in the Department of Computer Science at Rutgers University. He received his Ph.D. from the Faculty of Architecture and Town Planning at the Technion – Israel Institute of Technology, and he holds MSc and B.A degrees in Architecture from the Politecnico di Milano, Italy. Davide’s research lies at the intersection of Architectural Design, Artificial Intelligence, and Human Behavior Science with a mission to apply digital technologies for analyzing the dynamic interactions between people, the spaces they inhabit, and the activities they engage in.
Mubbasir Kapadia is an Assistant Professor in the Computer Science department at Rutgers University, and the Director of the Intelligent Visual Interfaces Lab. Previously, he held positions at Disney Research Zurich and the Human Modeling and Simulation Lab at University of Pennsylvania. He received his PhD in Computer Science at University of California, Los Angeles. Kapadia’s research lies at the intersection of artificial intelligence, visual computing, and human-computer interaction with a mission to develop intelligent visual interfaces for human-guided content creation.