Residential and commercial buildings account for roughly 40% of energy consumption and 30% CO2 emissions in the United States. As a result, achieving net-zero buildings to combat climate change is a longstanding national priority articulated in numerous executive orders spanning multiple administrations. To modernize our buildings, internet of things (IoT) devices can extract insightful information on building operation and how indoor conditions impact occupants. However, these IoT devices contribute to the “phantom energy problem” where standby power draw from IoT devices becomes significant at scale.

For example, a popular commercially available air quality sensor draws just 1.76 watts, but if placed in every one of the nation’s 5.6 million commercial buildings, just these sensors would add 237 MWh of energy consumption, per day. Annually, this is more energy consumption than some small countries. Therefore, we must be strategic about how we deploy and use sensors, and leverage local and remote models, correlated sensors, mobile sensors, and opportunistic sensing to enable data-driven improvements without costly energy overheads. 

The candidate working on this project will collaborate with a team of faculty, graduate, and undergraduate students on a “smart, IoT-enabled campus” project to create a community-level smart cities test-bed that can support research studies at the intersection of IoT technologies, energy consumption, CO2 reduction and public health.

As part of this project, the candidate will engage with large scale IoT deployment, data management, analyses, and modeling that could answer questions related to: (a) demand response techniques to reduce community level energy consumption and carbon emission, (b) optimization approaches for large scale IoT monitoring of different infrastructure systems (buildings, water, transportation), and (c) the impact of IoT and novel technology on health, well-being, privacy and/or security of individuals, groups, and communities. 


Outcomes from this Project

Publications

Sequential service restoration with grid-interactive flexibility from building AC systems for resilient microgrids under endogenous and exogenous uncertainties
Communication-Efficient MARL for Platoon Stability and Energy-Efficiency Co-Optimization in Cooperative Adaptive Cruise Control of CAVs
Performance evaluation of semi-supervised learning frameworks for multi-class weed detection
SoybeanNet: Transformer-based convolutional neural network for soybean pod counting from Unmanned Aerial Vehicle (UAV) images

Presentations

Large Language Models and Foundation Models in Smart Agriculture. Workshop on Artificial Intelligence in Agriculture.
Back-stepping Experience Replay with Application to Model-free Reinforcement Learning for a Soft Snake Robot. Northeast Systems and Control Workshop

Outreach

Guest Editor - The Special Issue of Applications of Artificial Intelligence (AI) in Agriculture, Electronics
Guest Editor - The Special Issue: Robotics: From Technologies to Applications, Electronics

Project Team

Image
Arsalan
Heydarian
Associate Professor
University of Virginia
Image
Brad Campbell
Brad
Campbell
Assistant Professor
University of Virginia
Image
Dong Chen
Dong
Chen
Assistant Professor
University of Mississippi
Image
mountain lake

Related News and Projects

Initiatives

All initiatives
Image
Antonios teaches a room of Climate Fellows

Related News and Projects