IRIM Fall 2026 Seminar | Advancing Robotic Assembly in Construction through Learning and Adaptive Control

Abstract: Robotic assembly in construction offers opportunities to alleviate labor shortages, reduce workers’ exposure to physically demanding tasks, and expand the possibilities for novel architecture. However, reliable robotic assembly in construction remains challenging due to uncertainties associated with variability in component geometry and material properties; tolerance accumulation during fabrication and assembly; discrepancies between as-designed and as-built states; incomplete observations resulting from limited sensing coverage of large components, occlusion, and clutter; and unmodeled contact dynamics. These challenges are particularly pronounced in contact-rich robotic assembly at construction scale, where small geometric deviations or pose estimation errors can prevent insertion, generate excessive contact forces, or damage components. 

To address these challenges, this lecture presents a framework for contact-rich robotic assembly in construction that integrates perception, planning, and control. Within this framework, three complementary approaches are examined: adaptive control to compensate for model mismatch and disturbances; learned sensorimotor policies to handle complex, tight-tolerance interactions for which effective control strategies are difficult to formulate analytically; and a hybrid approach that combines learned policies with adaptive control to improve robustness and enable zero-shot sim-to-real transfer without real-world demonstrations for policy training. Through construction-scale case studies, the lecture highlights key methods, experimental findings, and current limitations of these approaches. It concludes by discussing open challenges and future directions for generalizing across material systems and component geometries, transferring skills between tasks, and extending these methods to longer assembly sequences in construction.

Bio: Dr. Arash Adel is an Assistant Professor at the School of Architecture, a Core Faculty of Princeton Robotics, and an Associated Faculty of the Department of Computer Science at Princeton University. He is the founder and director of Adel Research Group (ARG). ARG conducts interdisciplinary research at the intersection of robotics, artificial intelligence, and computational design; the research contributes to resilient, sustainable, and low-carbon construction outlooks and achievements. At the core of ARG’s comprehensive research is investigating human–robot collaborative processes, which tackle fundamental questions related to the future of the design and construction industries and their potential to have a broader impact on inclusive and equitable building culture. Prior to joining Princeton University in 2023, Adel was an Assistant Professor of Architecture at the University of Michigan’s Taubman College of Architecture and Urban Planning. Adel received his Doctorate in Architecture from the Swiss Federal Institute of Technology (ETH) and his Master’s in Architecture from Harvard University.