CPS: SMALL: NSF-MeitY: 5G Enabled Real-Time Digital Twins of Dynamic Construction Sites
Lead PI:
Pratik Chaudhari
Abstract
This Cyber-Physical Systems (CPS) grant, a collaboration between the US National Science Foundation and the Ministry of Electronics and Information Technology of the Government of India (NSF-MeitY), supports research to develop a digital twin of dynamic environments such as construction sites using multiple unmanned aerial and ground robots equipped with cameras and 5G radios. This research can (a) enhance worker safety by identifying potential hazards and (b) improve construction efficiency by monitoring and optimizing resources devoted to different tasks. Technology emerging from this project can bring increased productivity in the construction industry, which has lagged behind other sectors such as manufacturing or agriculture. It can lead to affordable and green housing without compromising labor safety. To achieve these goals, the researchers will collaborate closely with multiple industry partners, and ensure public sharing of robotic data. This project will also work with the Philadelphia School District, which serves a primarily low-income and underrepresented minority population, to develop school curricula involving robotics and computer vision.<br/><br/>This project will build photometric, geometric, semantic, and radiometric representations of dynamic scenes using techniques from neural radiance fields and multi-modal visual-language features. Information-theoretic formulations of active perception and path planning will be used for monitoring construction progress and safety of human workers, using multi-robot teams. These algorithms will be run on aerial and ground robots along with edge servers that communicate periodically with these size, weight and power (SWaP)-constrained platforms using the latest standards in 5G. Using these algorithms, human workers carrying 5G radios can be localized using their radiometric signatures. Effectively, this project envisions a real-time perception and control stack executed on a heterogeneous multi-robot team. This research will be demonstrated using real-world experiments in Philadelphia and Bangalore (India).<br/><br/>This award reflects NSF's statutory mission and has been deemed worthy of support through evaluation using the Foundation's intellectual merit and broader impacts review criteria.
Performance Period: 09/01/2024 - 08/31/2027
Award Number: 2415249