“Bridging Robotics and Artificial Intelligence (BRAIN) for Future Construction and Built Environments”

Date/Time
Date(s) - 04/22/2024
10:30 AM - 11:30 AM

Location
Join Zoom Meeting https://ufl.zoom.us/j/512581235 Meeting ID: 512 581 235 Join by Skype for Business https://ufl.zoom.us/skype/512581235

Categories


zoom Link: https://ufl.zoom.us/j/4369404526?omn=94291021972

Meeting ID: 436 94044526

Speaker: Shuai Li, Ph.D,Associate Professor in the Department of Civil and Environmental Engineering,University of Tennessee Knoxville.

“Bridging Robotics and Artificial Intelligence (BRAIN) for Future
Construction and Built Environments”

Abstract: Advancements in robotics and artificial intelligence (AI) present both challenges and
opportunities for future construction and built environments. In this seminar, I will present our efforts
to bridge robotics and artificial intelligence for future construction and building systems, propelled by
technological innovations in digitalization and robotization. First, for construction digitalization, I will
present an industry-inspired interdisciplinary engineering project that developed an embodied AI copilot
to digitalize the construction process and streamline communication, addressing the longstanding
issues of low productivity and high rework rates in the industry. A laser ink display-based spatial
augmented reality technology is developed to convey construction information on job sites and a large
vision-language model is adapted to provide in-situ and real-time guidance for construction workers.
For construction robotization, I will introduce our generalizable robot learning approach to transition
from robot teleoperation to human-robot collaboration and ultimately toward full construction
automation. A keyframe-guided and goal-conditional learning framework is developed to enable robots
to learn from teleoperation data for complex construction operations. To ensure seamless human-robot
collaboration in dynamic construction sequences, a federated learning approach is developed to
predict multi-level and implicit human intentions for robots to plan their actions. In addition, large
language models are exploited with learned robot intelligence to convert explicit human intentions
into robotic actions to achieve higher autonomy in complex construction tasks and new environments.
Second, for building digitalization, I will present our convergence research to develop an integrated
digital twin for cybernating the interactions among buildings, occupants, and pathogens, such as
COVID, to minimize infection risks within buildings and reduce energy consumptions while
accommodating diverse human needs and preferences through AI. In the area of building robotization,
I will introduce our work in human-centric and robot-enabled building services, including social robots
engaging occupants with persuasive suggestions for health-promoting and energy-saving behaviors,
bio-socially-adaptive robots to clean and disinfect hospitals, and robotic see-through technology that
can penetrate collapsed structures to identify and visualize survivable voids for first responders in
search and rescue operations.

Leave a Reply

Your email address will not be published. Required fields are marked *