Connectivity, Information & Intelligence Lab
The mission of the Connectivity, Information & Intelligence Lab (Ci2Lab) is to fully realize the potential of Artificial Intelligence (AI) in order to improve our infrastructure and enhance our quality of life.
Research
Here are some examples of our ongoing research projects in Applied Machine Learning for monitoring and managing infrastructure and natural resources.
GridEyeS: Satellite Data Analytics for Infrastructure Monitoring
The rise of satellite imagery data and machine learning provide an opportunity to close the loop with continuous data-driven vegetation monitoring around infrastructure. We have developed an automated framework for monitoring vegetation along roadways and electricity lines using high-resolution satellite imagery and a semi-supervised machine learning algorithm. The proposed satellite-based vegetation monitoring framework, GridEyeS, aims to reduce the cost and time of asset monitoring by partially replacing ground patrols and helicopter or drone inspection with satellite data analytics. GridEyeS is supported by the European Space Agency. It is implemented on a Norwegian DSO system.
Project Webpage | Video | Lecture | Paper 1 | Paper 2
DisasterView: Disaster Impact Assessment with Space-based AI
Natural disasters, such as earthquakes, hurricanes, and floods, affect large areas and millions of people, but responding to such disasters is a massive logistical challenge. Emergency responders need fast and accurate assessments in the aftermath of disasters to plan how best to allocate limited resources. We develop advanced ML and AI algorithms to analyze high-resolution satellite imagery and SAR data automatically. We build crucial frameworks to give responders an unprecedented breadth of visual information about terrain, infrastructure, and populations affected by disasters.
For example, we assessed the impacts of Hurricane Michael (2018) in Tallahassee, Florida using multispectral satellite images.
TreeWatcher: Tree Species Classification Using Satellite Imagery
Knowing the vegetation type in an area is crucial for several applications, including forestry, land use management, wildfire prevention, infrastructure risk assessment, etc. Conventional methods for documenting tree species are based on field surveys that are intensive, time-consuming, and often impractical in mountainous or inaccessible areas. However, remote sensing technologies such as optical and LiDAR imaging by helicopters or airborne vehicles have been explored over the last few decades to address such challenges. However, they have high costs of labor and computing. Therefore, we developed a machine learning-based framework to estimate tree species relying on multi-spectral high-resolution satellite images.
Paper 1 |
SAR - Optical Fusion:
Synthetic aperture radar (SAR) and optical are the most common sensors for earth observation, each providing a unique livel of details. Optical and SAR sensors produce different images with different characteristics and complementary information. Combining the two sensors will increase the information it is possible to extract from satellite imagery, improving the quality of existing and future remote sensing applications.
In this project, we investigate methods for optical-SAR image matching and forest structure estimation.
AI4Hydro: AI Powered Forecasting for Hydro-Power
The complex nature of cascaded reservoirs for hydropower generation plus climate change impose massive uncertainty and dynamics into classic hydropower scheduling tools. We are developing artifactual intelligent-based hydropower forecasting algorithms using hydrological and meteorological data to deal with spatiotemporal interdependencies among various reservoirs' inflow and water capacity for the Norwegian power system. This project is in collaboration with NTNU and Smart Innovation Norway.
Project Webpage | Paper 1 | Paper 2 |
Co-Resilience: Multi-Domain Infrastructure Networks Resilience
Most studies on resilience treat it as a single dimension attribute of a system or investigate the different dimensions of the resilience separately without considering its multi-domain nature. We developed an advanced causal inference approach combined with machine learning to characterize the spatiotemporal and multi-domain vulnerability of an urban infrastructure, coined as "co-resilience.'' We performed resilience assessments for combined electricity and transportation networks by considering the meteorological, topographic, and demographic attributes of Tallahassee, Florida after Hurricane Hermine (2016) and Hurricane Michael (2018).
Project Webpage | Lecture | Paper 1 | Paper 2
Co-Mobility: Causal Inference for Complex Multi-Layer Networks
Urban mobility and electricity demand are multidimensional characteristics of city infrastructure including places, people, and information. The emergence of electric vehicles creates even more interdependency between electricity and transportation networks. Therefore, the study of electricity and transportation systems should go beyond an individual network and merge with other networks. However, understanding relationships between driving patterns, electricity demand, weather conditions, and demographics is a complex problem that is out of reach for classic machine learning. We tackle such a complex problem by introducing the "Co-Mobility" concept based on the causal Bayesian multi-network models. We have implemented our framework on actual data from the City of Tallahassee, Florida.
Team
Prof. Reza Arghandeh
Director, Ci2Lab
Professor in Data Science, Department of Comp Sci. and Elect Eng.
Leader, Data Science Group
Western Norway University of Applied Science (HVL)
Lead Data Scientist, StormGeo
Dr. Mojtaba Yousefi
Associate Professor, Department of Comp Sci. and Elect Eng.
Western Norway University of Applied Science (HVL)
Academic Advisory Board
Dr. Eren Erman Ozguven
Associate Professor, Department of Civil & Environmental Eng.
Florida State University
Current Members
Amir Miraki
PhD Candidate
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Email: amir@hvl.no
Mehak Khan
Postdoctoral Researcher
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Email: Mehak.Khan@hvl.no
Mira Kenzhebay
PhD Candidate
Western Norway Uni of App Sci
Email: Meruyert.Kenzhebay@hvl.no
Alumni
Dr. Mahyar Ghorbanzadeh
PhD
Department of Civil & Environmental Eng.
FSU-FAMU
Email: mg17x@my.fsu.edu
Sindre Aalhus
MS
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Email: 182037@stud.hvl.no
Dr. Jose David Cordova
Ph.D.
Department of Electrical and Computer Engineering
Florida State University
Dr. Lalitha Madhavi K.S
Ph.D.
Department of Electrical and
Computer Engineering
Florida State University
Dr. Mostafa Gilanifar
Ph.D.
Department of Industrial Engineering
Florida State University
Dr. Mehmet Baran Ulak
Ph.D.
Department of Civil and Environmental Engineering
Florida State University
Andres Sanchez
MS
Department of Electrical and
Computer Eng
Florida State University
Ali Sayghe
PhD
Department of Electrical and Computer Engineering
Florida State University
XiaoRui Liu
MS
Department of Electrical and
Computer Engineering
Florida State University
Dr. Ayberk Kocatepe
PhD
Connetics Transportation Group
DongLin Cai
MS
Dept of Electrical and Computer Engineering
Florida State University
Matthias Stifter
Visiting PhD Student
Dept of Electrical Engineering
Technical University of Vienna,
Austria
Mengmeng Xiao
Visiting PhD Student
Dept of Electrical Engineering
Huazhong University of Science and Technology, China
Davide Pinzan
Visiting MS Student
Dept of Electrical Engineering
University of Padova, Italy
Luca Di Narzo
Visiting MS Student
Dept of Mechanical Engineering
Politecnico di Milano, Italy
Monica Depalo
Visiting MS Student
Dept of Mechanical Engineering
Politecnico di Milano, Italy
Oscar Sommervold
MS Student
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Adrian Solheim
MS Student
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Sindre Larsen
MS Student
Electrical Eng, & Math Sci
Western Norway Uni of App Sci
Email:S.Larsen@student.uib.no
Projects
Development of innovative complex predictive maintenance system (EA-Predictive),
Dr. Arghandeh PI, EEA and Norway Grants, 2021-2024
Dr. Arghandeh Co-PI, , Research Council of Norway, 2021-2024
Prediction of ignition and spread of wildfires in Scandinavia: from experiments to models (PREWISS),,
Dr. Arghandeh Partner, Research Council of Norway, 2021-2025
Dr. Arghandeh Co-PI, , Research Council of Norway, 2021-2023
AI for Sustainable Energy,
Dr. Arghandeh PI, HVL University Project, 2020-2022
Satellite Technologies Feasibility Study for Power Lines
Dr. Arghandeh PI, Statnett & StormGeo, 2021-2021
Towards a FAIR and Open Data Ecosystem in the Low Carbon Energy Research Community
Dr. Arghandeh Co-PI, EU HORIZON 2020, 2020-2022
Information Mining with a Span of Fuzzy and Causality Approaches
Dr. Arghandeh Co-PI, the University and College Network for Western Norway, 2019-2020
GridEyeS - Smart Grid Eye, from Space to Sky
Dr. Arghandeh PI, European Space Agency, 2019-2020
IBM Faculty Award, IoT Technology for Smart Buildings
Dr. Arghandeh PI, IBM, 2018-2019
User-Centered Heterogeneous Data Fusion for Multi-Networked City Mobility UHDNetCity
Dr. Arghandeh PI and Dr. Ozguven Co-PI, U.S. National Science Foundation, 2016-2019
Resilient Alaskan Distribution System Improvements using Automation and Energy Storage
Dr. Arghandeh Co-PI, U.S. Department of Energy, 2017-2020
Bridging the Digital Divide for the Well-Being of Aging Populations in Smart Cities
Dr. Arghandeh Partner, U.S. National Science Foundation, 2017-2018
Open-source Distributed Control Platform for HIL-based Testing Advanced Ship Power Systems
Dr. Arghandeh Co-PI, U.S. Office of Naval Research, 2016-2017
High Resolution Monitoring System for Distribution Networks with Micro-Synchrophasors
Dr. Arghandeh PI, Florida State University Office of Research, FYA Program, 2016
Data and Model Integration for Multidimensional System Observability
Dr. Arghandeh PI, Florida State University Office of Research, 2015
News
This section is not updated often. A lot of cool things are happening in Ci2Lab; updates coming here soon!
Oct 2022
Prof. Arghandeh book "Big Data Application in Power Systems" is translated to Korean!
Jun 2022
We officially joined the ITU and WMO joint Focus Group on AI for Natural Disaster Management (FG-AI4NDM). We lead a use case on Use Case on Situational Awareness System for Disaster Response using Space-based AI (SARA).
Mar 2022
We are excited to start our new project funded by the EEA-Norway Grants called Development of innovative complex predictive maintenance system (EA-Predictive).
We will work closely with Energy Advice company in Lithuania.
Here is the press release.
June 2021
Prof. Arghandeh will deliver the keynote speech for the 2021 International Conference & Exposition on Modern Energy and Power Systems on June 16.
June 2021
Our paper "Post-Hurricane Vegetative Debris Assessment using Spectral Indices Derived from Satellite Imagery" is accepted for publication in the Transportation Research Record.
May 2021
Our paper "City Transportation Network Vulnerability to Disasters: The Case of Hurricane Hermine in Florida" is accepted for publication in the Environmental Hazards journal.
May 2021
Our paper "Automated Satellite-based Assessment of HurricaneImpacts on Roadways" is accepted for publication in the IEEE Transactions on Industrial Informatics.
April 2021
Our paper "Resilience Characterization for Multi-Layer Infrastructure Networks" is accepted for publication in the IEEE Intelligent Transportation Systems Magazine.
April 2021
Our group will presents three papers in the ASCE International Conference on Transportation & Development,
Paper 1: Evaluating the Relationship between Vehicle Travels and Carbon Monoxide (CO) Concentrations among Florida Counties under the Impact of COVID-19 Precautions
Paper 2: Post-Hurricanes Tree Debris Detection using Satellite Imagery and Deep-Learning
Paper 3: Traffic Forecasting Using Sentinel-5P Air Pollution Data and Historical Weather Records: A Remote Sensing Approach
Aug 2019
The Smart Grid Eye, from Space to Sky (GridEyeS) is funded by the European Space Agency. The overall goal of this study is to create an end to end electric grid monitoring platform using satellite images combined with high-resolution drone data to create multi-layer situational awareness, especially in the case of extreme weather events in the electric grid.
Sep 2018, Dr. Arghandeh joined the Western Norway University of Applied Sciences, as a Professor in Data Science and AI.
March 2018
Dr. Arghandeh honored for book publication in Florida State University
January 2018
Dr. Arghandeh and Dr.Ozguven visit of NASA Challenger Learning Center
and talk on BigData for SmartCities
January 2018
Dr. Arghandeh's book "Big Data Application in Power System" published by Elsevier
October 2017
Dr. Ozguven, published an article in Tampa Bay Times
October 2017
Dr. Ozguven, published an article in Tallahassee Democrat
Oct 2017
Oct 2017
Resilient Distribution Systems Lab Call Awards, Office Of The Under Secretary For Science And Energy, U.S. Department of Energy
Oct 2016
Radio Interview," the progression and development of DigiTally", Tallahassee Talk Radio Show.
Oct 2016
Featured Research,"FSU Team Tackles Urban Mobility In Smart City Era" FSU News.
Oct 2016
Featured Research," FSU Researchers Transforming Tallahassee Into a Smart City,” Florida Flambeau.
Jul 2015
News,"EECS postdoc scholar Reza Arghandeh has been selected to receive the 2015 ASME Award"
UC Berkeley.
Jul 2015
News, "ASME Power Award for FSU New Faculty" FSU College of Engineering Website.
Jun 2015,
News, " IEEE PES Award for Paper on Data Accuracy for Inverters Control" FSU College of Engineering Website.