Projects

MRI: Development of a mmWave-Networked Robotic Testbed for Multi-Agent AI Learning and Operations
MRI: Development of a mmWave-Networked Robotic Testbed for Multi-Agent AI Learning and Operations
MRI: Development of a mmWave-Networked Robotic Testbed for Multi-Agent AI Learning and Operations
MRI: Development of a mmWave-Networked Robotic Testbed for Multi-Agent AI Learning and Operations

MRI: Development of a mmWave-Networked Robotic Testbed for Multi-Agent AI Learning and Operations

Overview: The Center for Connected Autonomy and AI develops a first-of-its-kind millimeter-wave connected robotic platform for team-AI learning and operations. The platform consists of five ground mobile robotic modules equipped with on-board mmwave programmable transceivers, LiDAR, and GPU. Two of the mobile robotic modules are equipped with a robotic arm. Additional major platform components include two fixed-point programmable mmwave transceivers and a base-station GPU tower. Under this project, the investigators develop (i) novel multi-agent learning algorithms that execute over the networked modules and (ii) protocols for networked robotic team operation upon learning.

The Center platform is made available to all FAU Engineering and Computer Science faculty for research and instruction at all levels. Remote access to the platform from other US academic institutions will be made possible via real-time interfacing.


NSF Convergence Accelerator Track G: Autonomously TunableWaveform-Agnostic Radio Adapter for Seamless and Secure Operation of DoDDevices Through Non-Cooperative 5G Network
NSF Convergence Accelerator Track G: Autonomously TunableWaveform-Agnostic Radio Adapter for Seamless and Secure Operation of DoDDevices Through Non-Cooperative 5G Network

NSF Convergence Accelerator Track G: Autonomously Tunable Waveform-Agnostic Radio Adapter for Seamless and Secure Operation of DoD Devices Through Non-Cooperative 5G Networks

Overview: The Center for Connected Autonomy and AI at Florida Atlantic University joins forces with Florida International University (FIU), Virginia Tech (VT), and PQSecure Technologies LLC to create a universal radio adapter that will enable seamless and secure operation through non-cooperative indigenous 5G networks for U.S. government and critical infrastructure systems. The goal of the proposed universal radio adapter is to enable personnel, aircraft, satellites, mobile phones, vehicles, sensors, drones, and other IoT devices to operate through either friendly or adversarial (non-trusted) 5G network infrastructure and seamlessly connect with devices on trusted networks, while providing end-to-end data integrity, confidentiality and resiliency by data hiding and by autonomously switching between communications pathways. Through this research, FIU and FAU will develop a research center in 5G and beyond technologies to train undergraduate and graduate students in 5G wireless systems, post-quantum cryptography, physical layer and network security.
 
The proposed waveform-agnostic adapter will be compatible with U.S. DoD terrestrial and satellite communication protocols that operate from HF up to the Ka-band and will be able to interact with indigenous 5G ground or space-based networks that operate from the UHF up to the Ka-band. To trust the services provided by indigenous 5G infrastructure (treated as a black-box), the proposed convergence research effort will carry out accelerated research and development to enhance security and resiliency of end systems connecting to 5G networks, leveraging zero-trust principles where possible. The team’s workplan is organized in four technical thrusts: 1) Authentication of end users and devices will be carried out by a universal radio adapter that combines multi-band and multi-functional RF front-ends to connect to the black-box 5G network; 2) Data integrity and confidentiality will be enhanced by strong security protection at the physical layer (as a first line of defense) by exploiting unique characteristics of the wireless communication channel, the universal adapter hardware, and/or 5G core network as unique entropy sources; 3) Low size, weight, power and cost (SWaP-C) implementations of post-quantum cryptography (PQC) and standard ciphers will be adopted by the universal radio adapter to enable interoperability with the existing and a future “quantum-ready” 5G RAN/core and facilitate accelerated smooth transition to public-key based mechanisms based on the latest NIST PQC standards; 4) Data hiding techniques that securely embed data to popular application traffic of the 5G network and hide the act of embedding will be implemented at the universal adapter to disguise end device sessions.

Acquisition of Millimeter-Wave Platform for Joint Communication-Radar Signal Design with Applications to Connected Artificial Intelligence
Acquisition of Millimeter-Wave Platform for Joint Communication-Radar Signal Design with Applications to Connected Artificial Intelligence

A Millimeter-Wave Platform for Joint Communication-Radar Signal Design with Applications to Connected Artificial Intelligence

Overview: The Center for Connected Autonomy and AI acquired reconfigurable computing hardware, high-frequency data converters and radio equipment to build a millimeter-wave bi-directional MIMO platform for joint communication-radar signal design in the 71-76 GHz band. Under this project the PIs: (i) experimentally evaluate novel tensor data analysis algorithms for unsupervised soft characterization of sensed spectral data to enable truly robust and agile spectrum access; (ii) investigate experimentally, agile waveform designs that demonstrate robustness to clutter and agility to changing network dynamics and can offer a unique graceful tradeoff between communication data rate and high-quality sensing for situational awareness; (iii) extend dynamic spectrum access methods to facilitate MIMO communications-radar co-existence.
ARO sponsored research creates: (i) curated wireless spectrum datasets; (ii) software and hardware for testing & evaluation of radio technology capable of multitasking (comm/positioning/timing/navigation) in contested spectral environments; (iii) novel model-based & AI-assisted waveform designs for high-rate interference avoiding directional communications and communications-radar coexistence in 71-76 GHz band. 

ACE: Autonomous Conformity Evaluation of Tensor Data by Means of Novel L1-norm Principal-Component Analysis
ACE: Autonomous Conformity Evaluation of Tensor Data by Means of Novel L1-norm Principal-Component Analysis

ACE: Autonomous Conformity Evaluation of Tensor Data by Means of Novel L1-norm Principal-Component Analysis

Overview: Novel (i) blind real-time evaluation/screening of incoming sensed data of autonomous platforms for faults detection and safety and (ii) data quality assessment in training sets (training dataset curation) for improved learning.

Model-based and AI-assisted Autonomous Interference-Avoiding Directional Networking
Model-based and AI-assisted Autonomous Interference-Avoiding Directional Networking

Model-based and AI-assisted Autonomous Interference-Avoiding Directional Networking

Overview: Directional connectivity of tactical networks not only enables effective and efficient use of the available space-time-frequency continuum but also safeguards signals from would-be eavesdroppers. The objective of this effort is to develop and implement model-based and AI-assisted joint space-time waveform design algorithms for directional connectivity of multiple input multiple output (MIMO) network nodes that will enable resilient, self-sustained directional wireless networking of distributed ground/air assets that operate at maximum throughput.

Autonomous Interference-Avoiding Wireless Networks: The Need for Secure Control Plane
Autonomous Interference-Avoiding Wireless Networks: The Need for Secure Control Plane

Autonomous Interference-Avoiding Wireless Networks: The Need for Secure Control Plane

Overview: In the new generation of dynamic interference-avoiding networks, the control plane itself (and not the data plane) becomes the critical point of vulnerability. The objective of this effort is to investigate the development of user authentication and distributed crypto-key generation mechanisms for establishing a hardened control plane, which is the critical brain of the mission-driven autonomous network.

Model-based and AI-assisted Autonomous Interference-Avoiding Directional Networking - Part I

Model-based and AI-assisted Autonomous Interference-Avoiding Directional Networking - Part I

Overview: The objective of this effort is to investigate directionality versus data rate considerations for point-to-point and multicast communications and develop a blueprint of a 2x2 autonomous directional networking software-defined radio platform to be used for live directional interference avoidance.

 
 
 

Functional recovery (degeneracy) in biologically plausible neural architectures in partially observable environments

Overview: The wiring of brain regions involved in cognitive control can promote adaptive behavior by jointly minimizing multiple forms of uncertainty. This sub-project focuses on the emergence of such adaptation (functional recovery) from structural changes during reinforcement learning in partially observable environments.

crii
 
 
 

CRII: OAC: Scalable and Integrated Data Collection Platforms for Connected Vehicle Data

  • National Science Foundation
  • Jinwoo Jang (PI)
  • Duration: Aug. 2020- July. 2023
Overview: This project will investigate novel approaches to understand connected vehicle data, comprised of spatiotemporal trajectories and non-spatial sensor data, and develop scientific tools that can compress, impute, partition, and summarize connected vehicle data. This project will address important data challenges and scalability issues associated with the large scale of the database. This project will advance scientific knowledge in 1) trajectory data compression based on non-spatial sensor data, 2) the matrix decomposition techniques for the location inference of connected vehicle data, 3) the spectral graph partitioning methods for trajectories and non-spatial sensor data, and 4) large-scale trajectory data mining.

SII Planning: Center for Next Generation Wireless Spectrum Sharing
 SII Planning: Center for Next Generation Wireless Spectrum Sharing
 SII Planning: Center for Next Generation Wireless Spectrum Sharing
 SII Planning: Center for Next Generation Wireless Spectrum Sharing
 SII Planning: Center for Next Generation Wireless Spectrum Sharing

SII Planning: Center for Next Generation Wireless Spectrum Sharing

Overview: A planning grant for the Spectrum Innovation Initiative: National Center for Wireless Spectrum Research (SII-Center).
CAREER: A Skill-Driven Cooperative Learning Framework for Cyber-Physical Autonomy
CAREER: A Skill-Driven Cooperative Learning Framework for Cyber-Physical Autonomy

CAREER: A Skill-Driven Cooperative Learning Framework for Cyber-Physical Autonomy

  • National Science Foundation
  • Xiangnan Zhong (PI/PD)
  • Duration: Jun. 2021-May 2026
Overview: This project investigates new reinforcement learning (RL) approaches for cyber-physical autonomy to bridge the gap between current intelligent systems and human-level intelligence. The nature of many cyber-physical systems (CPS) is distributed, heterogeneous, and high-dimensional, making the hand-coded functions and task-specific information hard to design in the learning scheme. Large amount of training data is often required for achieving the desired performance, however this limits the generalization to other tasks. Hence, this project is to explore the new RL strategies to enable CPS with the capabilities of autonomous learning and generalization to rapidly adapt in unknown situations that were not assumed in the design phase. The results are expected to transform how agents interact in high-dimensional and heterogeneous environment, and therefore could potentially provide in-depth findings for exploring creativity in frontier Artificial Intelligence techniques.
CAREER: Toward Artificial General Intelligence for Complex Adaptive Systems: A Natural Concurrent “Learning-in-Learning” Control Paradigm
CAREER: Toward Artificial General Intelligence for Complex Adaptive Systems: A Natural Concurrent “Learning-in-Learning” Control Paradigm

CAREER: Toward Artificial General Intelligence for Complex Adaptive Systems: A Natural Concurrent “Learning-in-Learning” Control Paradigm

  • National Science Foundation
  • Zhen Ni (PI/PD)
  • Duration: Mar. 2021-Feb. 2026
Overview: The project will develop a natural concurrent Reinforcement Learning framework that carries three major advantages over traditional RL methods, namely the i) advantages of simultaneously learning multimodal properties of the complex system; ii) structural advantages of using a personalized learning scheme; and iii) implementation advantages of the data-driven sample-efficient design.
Autonomously Reconfigurable Hardware-Reduced Wideband Transceivers for Efficient Passive-Active Spectrum Coexistence
Autonomously Reconfigurable Hardware-Reduced Wideband Transceivers for Efficient Passive-Active Spectrum Coexistence
Autonomously Reconfigurable Hardware-Reduced Wideband Transceivers for Efficient Passive-Active Spectrum Coexistence
Autonomously Reconfigurable Hardware-Reduced Wideband Transceivers for Efficient Passive-Active Spectrum Coexistence

SWIFT: Autonomously Reconfigurable Hardware-Reduced Wideband Transceivers for Efficient Passive-Active Spectrum Coexistence

Overview: Research in novel low-cost hardware-reduced and multi-parameter reconfigurable ultra-wideband transceivers that optimize passive-active spectrum sharing across a broad frequency range.
Robust Localization by Massive MIMO
Robust Localization by Massive MIMO
Robust Localization by Massive MIMO
Robust Localization by Massive MIMO
Robust Localization by Massive MIMO

Robust Localization by Massive MIMO

Overview: Research in robust massive MIMO localization using the POWDER-RENEW NSF Platform for Advanced Wireless Research (PAWR).
 Secure communication based on robust 3D localization

Secure communication based on robust 3D localization

Overview: Under the European Union program NGAtlantic.eu, the CA-AI Center collaborates with the Technical University of Crete, Greece, to develop, implement and evaluate secure communication solutions for next-generation mobile wireless networks, building on Center-developed robust 3D localization algorithms.
Making the Master's Degree in Artificial Intelligence Accessible to High-Achieving Low-Income Students
Making the Master's Degree in Artificial Intelligence Accessible to High-Achieving Low-Income Students

Making the Master's Degree in Artificial Intelligence Accessible to High-Achieving Low-Income Students

Overview: The CA-AI Center contributes to the national need for well-educated scientists, mathematicians, and engineers, by supporting the retention and graduation with a dual degree, B.Sc. in an engineering field and M.Sc. in Artificial Intelligence, of high-achieving, low-income students with demonstrated financial need at Florida Atlantic University, a Hispanic Serving Institution.
CAREER: Modeling and Control of Undulating-Fin Underwater Vessels in Close Formation
CAREER: Modeling and Control of Undulating-Fin Underwater Vessels in Close Formation

CAREER: Modeling and Control of Undulating-Fin Underwater Vessels in Close Formation

  • National Science Foundation
  • Oscar Curet (PI/PD)
  • Duration: Jun. 2018-May 2023
Overview: The objective of this work is to investigate the correlation between the hydrodynamics interaction, the far-field wake signature, maneuver control and the performance of a bio-mimetic array composed of underwater vehicles.
I-Corps Sites: Type I - Florida Atlantic University I-Corps Site for Advancing Entrepreneurship and Innovation
I-Corps Sites: Type I - Florida Atlantic University I-Corps Site for Advancing Entrepreneurship and Innovation

I-Corps Sites: Type I - Florida Atlantic University I-Corps Site for Advancing Entrepreneurship and Innovation

  • National Science Foundation
  • Dimitris Pados (PI/PD)
  • Duration: Oct. 2018-Sep. 2021
Overview: The I-Corps Site program helps (i) increase participation of Hispanic entrepreneurs at FAU in STEM innovation and (ii) retain newly formed innovative companies capable of accelerating innovations to the market and attracting investors and industry partners to the Palm Beach County and South Florida region.
Machine Learning and Game Theory for Residential Community Energy Management Systems

Machine Learning and Game Theory for Residential Community Energy Management Systems

  • Oak Ridge Associated Universities (ORAU) Ralph E. Powe Junior Faculty Enhancement Awards
  • Zhen Ni (PI/PD)
  • Duration: Oct. 2020-Sep. 2021
Overview: The technical research goal of this project is to develop a new game-theoretic reinforcement learning as a unified framework to improve the coordination, adaptation and optimization for the future residential community energy management system.
Advancing Self-Localization and Intelligent Mapping (SLIM) for Swarm of Autonomous Unmanned Underwater Vehicles using Machine Learning
Advancing Self-Localization and Intelligent Mapping (SLIM) for Swarm of Autonomous Unmanned Underwater Vehicles using Machine Learning

Advancing Self-Localization and Intelligent Mapping (SLIM) for Swarm of Autonomous Unmanned Underwater Vehicles using Machine Learning

  • Naval Sea Systems Command (NAVSEA) - Naval Engineering Education Consortium (NEEC)
  • Zhen Ni (FAU PI/PD)
  • Duration: Jun. 2020-May 2023
Overview: The project is to expand the Unmanned Underwater Vehicles (UUVs) capabilities through artificial intelligence to static undersea sensors and/or dynamic undersea groups. The proposed research is to improve the autonomous perception and data fusion in an effort to generate world models from individual sensing, while localizing the sensors and UUV swarm within the model. This project is in collaboration with South Dakota School of Mines and Technology and NUWC Keyport Division.
Unbreakable Security for Underwater Multimodal Networks
L3Harris

Unbreakable Security for Underwater Multimodal Networks

Overview: In this project, we explore the performance of secret key generation for underwater wireless networks from both acoustic and optical channel probing.
Next-generation Ocean IoT Infrastructure for Education, Habitat Restoration and Conservation in Wahoo Bay
https://wahoobay.org/

Next-generation Ocean IoT Infrastructure for Education, Habitat Restoration and Conservation in Wahoo Bay

Overview: In this project, we propose to leverage off-the-shelf hardware and open-source software to rapidly prototype ocean IoT systems and buoy-based mesh wireless networks for marine data acquisition. Through the course of the project, the team will also work on the development tutorials and hands-on demo presentations for outreach to local high-school students from the FAU A.D. Henderson high school and undergraduate students from FAU’s College of Engineering and Computer Science.
Cognitive Networking of the Oceans: Localization and Tracking Fundamentals
Cognitive Networking of the Oceans: Localization and Tracking Fundamentals

Cognitive Networking of the Oceans: Localization and Tracking Fundamentals

Overview: The researchers aim at advancing significantly (five-fold in precision) the state of the art in 3D passive undersea acoustic localization by providing novel algorithmic solutions to the problem of estimating the angle of arrival of undersea propagating signals in the presence of potentially faulty measurements. The project pursues a paradigm shift in how signal Direction-of-Arrival (DoA) estimation is carried out over undersea acoustic links by deviating from the familiar L2-norm-subspace decomposition theory and its variants and considering for the first time L1-norm signal subspace decomposition. The algorithms will be embedded in in-house developed programmable underwater acoustic modems to form an experimental four-node network for evaluation and demonstration.
Autonomous Interference Avoiding Networking on the M-Series GE Platform
Autonomous Interference Avoiding Networking on the M-Series GE Platform

Autonomous Interference Avoiding Networking on the M-Series GE Platform

  • GE Aviation |Advanced and Special Projects Division
  • Dimitris Pados (PI/PD), George Sklivanitis (Co-PI)
  • Duration: Nov. 2018-Aug. 2020
  • AFWERX Multi-Domain Challenge Showcase Finalist
Overview: Autonomous all-spectrum interference-avoiding networking to support multi-domain (underwater, surface, air, and space) connected autonomy applications.
Analysis of Reconciliation/Privacy Amplification of Physical Layer Security for Underwater Acoustic Communications
Analysis of Reconciliation/Privacy Amplification of Physical Layer Security for Underwater Acoustic Communications

Analysis of Reconciliation/Privacy Amplification of Physical Layer Security for Underwater Acoustic Communications

Overview: Analysis of physical layer security algorithms for underwater acoustic networks.
Toward Real-time Spectrum Analytics: Conformity Evaluation of RF Data by Means of Novel L1-norm Principal Component Analysis
Toward Real-time Spectrum Analytics: Conformity Evaluation of RF Data by Means of Novel L1-norm Principal Component Analysis
Toward Real-time Spectrum Analytics: Conformity Evaluation of RF Data by Means of Novel L1-norm Principal Component Analysis

Toward Real-time Spectrum Analytics: Conformity Evaluation of RF Data by Means of Novel L1-norm Principal Component Analysis

  • Air Force Research Laboratory
  • Dimitris Pados (PI/PD)
  • Duration: Nov. 2019-Apr. 2020
Overview: Artificially-Intelligent (AI) methods for signal analysis and characterization.
CyberTraining: Multi-disciplinary Training of Learning, Optimization, and Communications for Next-generation Power Engineers
CyberTraining: Multi-disciplinary Training of Learning, Optimization, and Communications for Next-generation Power Engineers

CyberTraining: Multi-disciplinary Training of Learning, Optimization, and Communications for Next-generation Power Engineers

  • National Science Foundation
  • Zhen Ni (PI/PD)
  • Duration: Sept. 2019-Aug. 2023
Overview: Advanced adversarial learning algorithms for multi-layer games with applications to smart grid and connected smart city environments.
A Reflective Learning and Association Control Framework based on Adaptive Dynamic Programming: Arcitecture and Applications in Robotics
A Reflective Learning and Association Control Framework based on Adaptive Dynamic Programming: Arcitecture and Applications in Robotics

A Reflective Learning and Association Control Framework based on Adaptive Dynamic Programming: Architecture and Applications in Robotics

  • National Science Foundation
  • Zhen Ni (PI/PD)
  • Duration: Oct. 2018-Sep. 2020
Overview: AI algorithms for learning and decision making in complex engineering problems (e.g. path planning for robot-assisted pedestrian flow).
A Self-Learning Intelligent Control Framework for Networked Cyber-Physical Systems
A Self-Learning Intelligent Control Framework for Networked Cyber-Physical Systems

A Self-Learning Intelligent Control Framework for Networked Cyber-Physical Systems

  • National Science Foundation
  • Xiangnan Zhong (PI/PD)
  • Duration: Jul. 2019-Feb. 2021
Overview: Learning in a connected multi-agent environment - a new self-learning intelligent control framework.
Autonomous Hierarchical Adaptive Dynamic Programming for Decision Making in Complex Environment
Autonomous Hierarchical Adaptive Dynamic Programming for Decision Making in Complex Environment

Autonomous Hierarchical Adaptive Dynamic Programming for Decision Making in Complex Environment

  • National Science Foundation
  • Xiangnan Zhong (PI/PD)
  • Duration: Aug. 2019-Jul. 2022
Overview: An innovative reinforcement learning approach for decision making in complex environments.
Development of a low-cost, portable, in situ imaging flow cytometer for automated analysis of phytoplankton community composition

Development of a Self-Adaptive mmWave-Networked Buoy for Real-time High-Rate Marine Data Acquisition


Development of a low-cost, portable, in situ imaging flow cytometer for automated analysis of phytoplankton community composition

Development of a low-cost, portable, in situ imaging flow cytometer for automated analysis of phytoplankton community composition


Novel Ultra-Wideband Antenna Array Design for Low-Cost High Data Rate Rural Communications

Novel Ultra-Wideband Antenna Array Design for Low-Cost High Data Rate Rural Communications


Real-time Harmful Algal Bloom Detection and Monitoring: Testing a Novel Automated Classification Algorithm in Holographic Imagery
Real-time Harmful Algal Bloom Detection and Monitoring: Testing a Novel Automated Classification Algorithm in Holographic Imagery

Real-time Harmful Algal Bloom Detection and Monitoring: Testing a Novel Automated Classification Algorithm in Holographic Imagery


Reinforcement Learning for Navigation and Coordination of Bioinspired Underwater Vehicles in Close Formation
Reinforcement Learning for Navigation and Coordination of Bioinspired Underwater Vehicles in Close Formation

Reinforcement Learning for Navigation and Coordination of Bioinspired Underwater Vehicles in Close Formation