| CARVIEW |
Biography
Polar is an Assistant Professor in the Department of Civil Engineering at Stony Brook University. His research interests include human-robot collaboration, computer vision, reinforcement learning, building information modeling (BIM), digital twins, and extended reality to promote sustainability and safety in future robotized workplaces such as construction sites, manufacturing plants, and warehouses. His current research involves (1) developing a computer vision and sensor fusion system for identifying safety hazards in human-robot collaboration, (2) developing collaborative robots, extended reality, and imitation learning algorithms for sustainable construction and built environments, (3) investigating reinforcement learning methods, BIM, and digital twins for closing sim-to-real gaps in unstructured construction environments. Prior to joining SBU, he was a Robotics Fellow in the Division of Safety Research at National Institute for Occupational Safety and Health (NIOSH).
Interests
- Robotics
- Computer Vision
- Reinforcement Learning
- BIM and Digital Twins
- Extended Reality (XR)
Education
Ph.D. in Civil and Environmental Engineering, 2021
University of Michigan
M.S. in Robotics, 2017
University of Michigan
M.S. in Civil Engineering, 2015
National Taiwan University
B.S. in Civil Engineering, 2012
National Taiwan University
Selected Publications
Projects

Teaching robots to perform construction tasks through demonstration
Apply Learning from Demonstration method to teach construction robots.

Vision-based pose estimation for construction robots
Develop a vision-based DNN pose estimation system.

ARCT: AR for Clinical Training
Develop an AR stroke simulation for clinical training.
RAS: Robotic Assembly System for Steel Structure
RAS, a robotic assembly system for steel structure, remove workers from high places and perform steel beam assembly tasks semi-automatically.

Sky Classroom: Global Team Project and Virtual BIM Reviewer
Global construction collaboration course taught between seven universities using BIM and online collaboration platform.

BotBeep: Warning Device for Wheelchair Rearward Safety
BotBeep, an innovative and low-cost device, was developed to warn the user of elevation changes behind the wheelchair.



