Virtual part repair programming for robotic thermal spray applications  (Advanced Robotics for Manufacturing (ARM) Institute)
This project focuses on implementation of a virtual method to allow robotic thermal spray operators to automatically generate the robotic programming over a repair area on a physical part within a thermal spray booth using an immersive interface. The project team at UConn will design and develop a sensor system for measuring thickness buildup over a part during various thermal spray processes, such as High Velocity Oxygen Fuel (HVOF) and Cold Spray. (This research is in collaboration with Titan Robotics.)

Physics informed reduced order modeling (ONR)

Simulation-based uncertainty quantification of manufacturing technologies (Air Force Research Laboratory)
The goal of this research is to develop computational methodology to evaluate the impact of uncertainty in manufacturing conditions on part charcteristics and performances.  Our task is to synthesize a new probabilistic framework utilizing latest advancements in computational intelligence and statistical inference to perform uncertainty quantification of composites manufacturing, cutting, casting, and additive manufacturing processes.

Intelligent wave guiding (ONR)

Automated defect inspection of complex metallic parts (Advanced Robotics for Manufacturing (ARM) Institute)
This project aims at automating manufacturing inspection of metallic parts.  At UConn, we will generate the surface inspection software that provides the necessary artificial intelligence (AI) to detect and classify faults on complex surfaces.  (This research is in collaboration with University of Washington.)

Adaptive control of vibration and damping (ONR)

Resilient extraterrestrial habitat (NASA) (This research is in collaboration with Purdue University, University of Texas at San Antonio, and Harvard University.)

Structural fault diagnosis and prognosis utilizing adaptive sensor-structure interaction: a physics-guided data analytics approach (National Science Foundation)
The overarching goal of this research is to create a new framework of fault diagnosis and prognosis enabled by physics-guided data analytics.  This framework is built upon the integration of computational intelligence with high-fidelity modeling/analysis and the adaptization of a highly promising, non-contact sensor-structure interaction mechanism through megneto-mechanical coupling.

Large metallic sanding and finishing (Advanced Robotics for Manufacturing (ARM) Institute)
The goal of this research is to develop a robotic system capable of carrying out sanding operation autonomously.  At UConn, we focus on sanding quality inspection utilizing deep learning and convolutnional neural network.  (This research is in collaboration with Wichita State University and University of Washington.)

BIGDATA: IA: Collaborative Research: From bytes to watts – A data science solution to improve wind energy reliability and operation (National Science Foundation)
The wind energy research has so far been carried out by domain-specific experts in aerospace, civil, electrical, and mechanical engineering; data science methods have historically played a rather minor role. The proposed research will make a paradigm shift in the wind industry by demonstrating how dramatically data science innovations can benefit its development.  The efforts of the research team encompass  modeling of dynamics in spatio-temporal wind data using machine learning, capturing high-order factor interactions, and reliability assment using a small run of simulations in combination with historical data, and on-site computation.

CPS/Synergy/Collaborative Research: Cybernizing mechanical structures through integrated sensor-structure fabrication (National Science Foundation)
This research aims at developing a new framework of utilizing emerging additive manufacturing technology to produce a structural system with integrated, densely distributed active sensing elements. The outcome will lead to paradigm-shifting progress in structural self-diagnosis.  Our research focuses on the acquisition of high-quality, active interrogation data throughout the entire structure, which can then be used to facilitate highly accurate and robust decision-making through data analytics.

Analysis, design, and control of ultrasonic wheel probe for high-speed inspection of rails (Sperry Rail Service)
This research concerns the nonlinear dynamic analysis of wheel probe membrane as well as the ultrasonic wave propagation.  The goal is to improve the sensing robustness in high-speed inspection.

Progressive fault identification and prognosis in aircraft structure based on dynamic data driven adaptive sensing and simulation (Air Force Office of Scientific Research)
The objective of this research is to create a new methodology of progressive fault identification and prognosis for Air Force applications based on the framework of dynamic data-driven applications systems (DDDAS). This system is different from the current technologies, as it is inspired by the latest research progress of adaptive sensor, and enabled by a suite of new modeling techniques and mathematical and statistical algorithms that can fully utilize the sensor adaptivity and enriched measurements.

A system-level framework for operation and maintenance: synergizing near and long term cares for wind turbines (National Science Foundation)
This research tackles the system-level operation and management issues of wind turbines through synergizing the near-term and long-term cares of turbines. The near-term care is through engagement of adaptive blade pitch control, which re-designs the control logic by involving a trade-off between power generation and reliability. The long-term care is through cost-effective maintenance scheduling, which plans maintenance while considering stochastic logistic and weather constraints.