Led by Thomas, HEAL brings together scientists, clinicians, and research institutions across the United States, as well as partners in Finland and Australia, to develop new approaches for treating lymphatic diseases.
]]>George W. Woodruff School of Mechanical Engineering
]]>Andrei Fedorov and his team have found a way to take some of the guesswork out of mass spectrometry imaging (MSI), potentially giving scientists better tools to explore the molecules that influence our health.
A small sample of human tissue can hold a remarkable amount of information.
Researchers can use it to search for molecular clues associated with cancer, investigate the chemical landscape of the brain, or determine where a drug travels after it enters the body. Mass spectrometry imaging, or MSI, helps answer those questions by mapping where different molecules are located within tissue.
But there's a catch. Despite MSI's sophisticated capabilities, finding the right conditions to get useful results can still require a considerable amount of trial and error. Georgia Tech professor Andrei Fedorov and Ph.D. student John Sentmanat are working to change that.
“Mass spectrometry imaging is probably the most powerful tool to enable new discoveries in cell biology and protein biochemistry. It helps provide a direct insight not only to how the body works but also the underlying cause of disease,” said Fedorov, Associate Chair for Graduate Studies and Rae S. and Frank H. Neely Chair in the George W. Woodruff School of Mechanical Engineering. “But the techniques for using it were evolving in a very empirical manner guided by experience and intuition. It’s like shooting arrows in the dark.”
In response to this barrier, the team has developed what Fedorov likens to a recipe, a framework designed to help researchers understand how different ingredients and conditions work together to produce the molecular information they want. Published in Analytical Chemistry, the approach could make MSI more predictable and give scientists a better starting point for developing new ways to study the molecular workings of the human body.
“This paper provides a very simple recipe that everyone can read and use. You don’t need to be a chef, or in this case, be able to write complicated equations,” Fedorov said. “It even allows you to come up with a new recipe using our model as the base.”
Though the framework was designed to be accessible to researchers across the field, Fedorov and Sentmanat didn’t set out to solve a problem for the scientific community — they were trying to solve one in their own lab.
In 2020, Fedorov, who is also a faculty member in the Parker H. Petit Institute for Bioengineering and Bioscience, was awarded $1.2 million from the National Institutes of Health (NIH) for the development of a new imaging instrument that could change the study of tumor growth, disease progression, and cell function.
The technology, called BeamMap, has continued to develop with additional support from the National Science Foundation Engineering Research Center for Cell Manufacturing Technologies (CMaT), led in part by Georgia Tech. By bringing together multiple imaging techniques, including MSI, BeamMap allows researchers to study both the physical surface of a biological sample and its molecular makeup.
While developing the instrument, Fedorov and Sentmanat encountered the same challenge that has made MSI difficult to optimize: Under some conditions, they successfully extracted the molecules they wanted to study. Under others, they damaged a sample without getting the information they needed. Rather than continue relying on trial and error, the team decided they needed a theory that could point them toward the right conditions.
“As we continued developing the instrument, we realized we needed to develop a theory to identify the optimal operating conditions,” Sentmanat said. “What began as an effort to fine-tune our own technology led to a broader theory that can be applied to a wide range of other approaches.”
To build the recipe, the team returned to the basic physics that determines when molecules leave a sample and become available for detection. They distilled those complex interactions into a simpler framework, then compared its predictions with experimental data from previously published studies. When the framework held up across different MSI approaches, the researchers realized the recipe could be useful well beyond their own lab.
“It's all measurable data that any scientist in this field would be able to use,” Fedorov said. “It’s like having a magic tool that allows you, all of a sudden, not to be blind.”
Fedorov sees the new framework as one step toward making MSI easier to understand and use. By giving researchers a better starting point for choosing how to operate an MSI system, the recipe could help them spend less time searching for the right conditions and more time using the technology to answer biological questions.
That could become increasingly important as medicine moves toward understanding disease and treatment at the molecular level.
“More and more, physicians base their diagnoses on much more specific molecular information,” Fedorov said. “The drugs of the future will not be this little pill that you will be given; they will be your own immune cells. Your diagnostic tools have to match that future.”
For now, the framework is intended primarily as a tool for researchers. But by helping scientists make better use of technologies like MSI, Fedorov hopes the work can contribute to discoveries that ultimately improve how diseases are understood, detected, and treated.
Like any recipe, the framework leaves room for adjustment. The team deliberately simplified some of the complex physics behind MSI to make the framework practical for other researchers, which means it is designed to point them toward promising conditions rather than predict exactly what will happen in every experiment.
“The theory provides a recipe that researchers can adapt by adding their own flavors,” Sentmanat said. “Rather than prescribing an exact solution, it offers a framework for identifying the right ingredients and refining the conditions to achieve the best results.”
Researchers can adapt that starting point to their own instruments, samples, and questions. What began as an effort to improve BeamMap could ultimately help others refine existing MSI technologies or develop new approaches of their own.
“It’s kind of empowering.”
Researchers can use a small sample of human tissue to search for molecular clues associated with cancer, investigate the chemical landscape of the brain, or determine where a drug travels after it enters the body. Mass spectrometry imaging, or MSI, helps answer those questions by mapping where different molecules are located within tissue. But there's a catch. Despite MSI's sophisticated capabilities, finding the right conditions to get useful results can still require a considerable amount of trial and error. Georgia Tech professor Andrei Fedorov and Ph.D. student John Sentmanat are working to change that.
]]>Parker H. Petit Institute for Bioengineering and Bioscience
]]>In a study accepted for the 2026 ACM Conference on Computer and Communications Security, Georgia Tech researchers interviewed 20 military sailors about their cybersecurity experiences, training, and responses to hypothetical scenarios. Their findings point to a gap between how mariners are trained to handle emergencies and how cyberattacks can compromise the systems they depend on.
The study found several gaps in how crews prepare for cyber threats:
These gaps matter because cyberattacks can compromise a system without causing it to stop working. For example, an attacker could manipulate navigation data, leaving a crew to make decisions based on false information.
“I’d assume it was anything but cyber at first. It would probably take me quite some time to get to that conclusion,” one participant told the researchers.
The team examined three hypothetical scenarios: GPS spoofing, loss of propulsion control in a high-traffic area, and failure of electronic navigation systems in a congested channel at night.
In each scenario, mariners would first focus on keeping the vessel safe, using established procedures to stabilize the ship and troubleshoot its systems. While this response is essential during an emergency, it could delay recognition of an attack if crews do not consider that the system or its data may have been compromised.
For example, a ship navigating a narrow channel depends on accurate location information, but what happens when its navigation systems appear to work normally while a hacker manipulates that data?
An attacker could use GPS spoofing to falsify a vessel’s reported position. If the false position is close enough to the ship’s actual location, a mariner might not notice anything wrong. The vessel could gradually head toward shallow water or another obstacle without triggering an obvious error or alarm.
“By the time the mariner realizes something is wrong, the ship’s size and limited maneuverability could make it too late to avoid a grounding or collision,” said Anna Raymaker, a Georgia Tech Ph.D. student who studies maritime security.
The scenario illustrates a central challenge in maritime cybersecurity: An equipment failure may be easy to recognize, but a cyberattack that provides false information can be much harder to detect.
Batten the Hatches: Cybersecurity with Military Mariners was written by master’s student Ryan Von Brock, Raymaker, Research Scientist Animesh Chhotaray, Associate Professors Frank Li and Saman Zonouz, and Provost & Executive Vice President for Academic Affairs Raheem Beyah.
The research was conducted in Georgia Tech’s Cyber-Physical Security Lab, a collaboration between the School of Cybersecurity and Privacy and the School of Electrical and Computer Engineering.
Photos by Terence Rushin/College of Computing
]]>Sivakumar currently serves as Georgia Tech’s vice president of Commercialization and chief commercialization officer, leading the Office of Commercialization. The office serves as the Institute’s central hub for student entrepreneurship, research translation and commercialization, and venture development and scaling, enabling faculty and students to bring new technologies to market. Sivakumar previously served as the founding director of CREATE-X, a Georgia Tech initiative that fosters entrepreneurial confidence and helps students launch real startups. He has also founded and co-founded multiple technology companies and held technical roles in industry.
In addition to his administrative leadership, Sivakumar is the Wayne J. Holman Chair and professor in the School of Electrical and Computer Engineering (ECE). His research focuses on wireless networking and mobile computing.
Sivakumar joined ECE in 2000 as an assistant professor. He earned his bachelor’s degree in computer science from Anna University in Chennai, India, and his master’s and doctoral degrees in computer science from the University of Illinois Urbana-Champaign. He is a Fellow of the Institute of Electrical and Electronics Engineers.
]]>"The most meaningful breakthroughs happen when intelligent hardware and thoughtful software come together to genuinely help people,” said Tapas Roy, vice president of Device Software and Services at Amazon. “Georgia Tech's world-class expertise in robotics and engineering, paired with Amazon's work in on-device intelligence and artificial intelligence, will accelerate our understanding of systems that can both respond to prompts and anticipate customer needs. I'm excited to see how this collaboration advances the science of how humans and intelligent machines work together.”
The Hub’s initial focus areas will explore challenges in autonomous systems, next-generation logistics infrastructure, and low-Earth-orbit communications, drawing on Georgia Tech’s well-known interdisciplinary research approach.
Georgia Tech is also joining the Fauna Robotics' ecosystem, using the Sprout research platform to study how humans and robots can live, work, and play side by side. This collaboration opens new ways to study mobility, perception, manipulation, and human-robot interaction. Leadership and alumni recently celebrated the launch at Lab 126, with guest appearances from Sprout and Buzz, the Georgia Tech mascot.
The Hub will support:
• Joint research projects between Georgia Tech faculty and Amazon scientists across scientific disciplines
• Regular symposia and workshops to foster collaboration between researchers from both institutions and the broader Atlanta tech community
• Community outreach programs designed to broaden participation from interdisciplinary scholars and innovators
“Georgia Tech and Amazon have worked together for years across research, technology, and student development," said Tim Lieuwen, Georgia Tech executive vice president for Research. "This new hub puts our partnership to work for Georgia. It connects research, education, and industry in ways that will strengthen our state’s innovation economy and create new opportunities for our students.”
About Georgia Tech
Georgia Institute of Technology is one of the nation’s top public research universities, preparing leaders who advance technology and improve the human condition. The Institute serves more than 56,000 students from all 50 states and over 146 countries and drives innovation and economic growth through research that benefits Georgia, the nation, and the world.
About Amazon in Georgia
Since 2010, Amazon has invested more than $30 billion in Georgia, supporting more than 30,000 full- and part-time jobs. Amazon’s research and technology teams in the Atlanta area work on challenges spanning supply chain optimization, logistics technology, and cloud computing. The establishment of the Georgia Tech–Amazon Science Hub further deepens Amazon’s engagement with one of the nation’s leading research institutions.
About Fauna Robotics
Fauna Robotics builds Sprout, a development platform designed to put advanced robotics capabilities into the hands of researchers and developers. Sprout gives builders hands-on access to study mobility, perception, motor control, and manipulation with the ambition to deploy robots into human-centric spaces.
Their system means devices can work together like never before, sensing in one part of the body and triggering a therapeutic response elsewhere — perhaps releasing medicine or stimulating a nerve.
Described Sept. 24 in the journal Science, their communication method is expandable to include multiple interconnected devices across the body, even deep inside the stomach.
Read the full story on the College of Engineering website.
]]>Those conversations led Raymaker to focus her research on maritime security, including vulnerabilities in the Automatic Identification System (AIS), which ships use to broadcast their location, speed, and course.
This summer, Raymaker brought this research to Massachusetts Institute of Technology (MIT) Lincoln Laboratory, where she built a test bed to study AIS security in a controlled environment. The experience has also led to new research questions, from the security of shipboard Starlink connections to autonomous vessels and other maritime technologies.
We spoke with Raymaker about her research, her experience at Lincoln Laboratory, and what she hopes to investigate next.
After submitting our paper based on this summer’s work to a cybersecurity conference, we are continuing our collaboration and starting a new project to identify critical infrastructure that uses unique protocols. The project is still in the early stages of planning and design, but I’m excited to continue working with the team.
The biggest challenge was getting equipment designed to be installed and operated on a boat to function properly in a lab setting. That meant using a Faraday cage so that our AIS transmissions and security experiments wouldn’t accidentally broadcast to nearby ships or interfere with real maritime traffic. I also had to simulate GPS signals and set up additional hardware to trick the AIS transponders into thinking they were operating on a real vessel. Putting all those pieces together into a safe, realistic test bed was challenging, but it ultimately let us study the equipment in a controlled environment.
Working with Hamed Okhravi, a senior staff member at Lincoln Laboratory, was an amazing experience. He was my mentor and consistently went out of his way to connect me with people who could help me find datasets, access equipment, or get the lab space I needed for my work. For example, when I realized I needed a Faraday cage for my experiments, he connected me with the Mobile Device Lab and helped me get access very quickly. The lab manager was also immediately willing to let me use the space and equipment. That experience really reflected the culture I saw throughout Lincoln Laboratory. Everyone genuinely wanted to help each other get what they needed to move their research forward.
My experience at Georgia Tech was instrumental in preparing me for my work at Lincoln Laboratory. The way I was taught to think through difficult research problems and work independently toward solutions translated well to a large research lab. I was able to hit the ground running and take ownership of a project because Georgia Tech had prepared me to lead research independently. The internship also gave me valuable experience in a research environment outside academia, which broadened my perspective on how I could pursue research in the future.
Next, I’d love to look at Starlink connectivity for ships or for autonomous shipping. Mariners we’ve spoken with have raised concerns about both. With Starlink and increasingly autonomous vessels, they worry that greater remote connectivity could give attackers new ways to access onboard systems and disrupt ship operations. I think those concerns are worth investigating.
We recently had a paper accepted at CCS that examines exposed solar distributed energy resources, and I think applying a similar perspective to exposed maritime equipment could be very interesting. That paper is available here.
We also recently had a paper on measuring GPS spoofing of ships accepted at IEEE S&P. The full paper isn’t online yet, but the project was directly motivated by mariners telling us that they were struggling with GPS spoofing. That made me want to find a way to systematically measure where spoofing was occurring.
Ultimately, the goal is to turn that kind of research into something useful for mariners, such as warning them when they are entering an area experiencing spoofing or when their own vessel may be affected and they should no longer trust their GPS. That is the kind of research I’d like to continue doing: listening to the problems mariners are actually experiencing and finding ways to investigate and address them.
Photos by Kevin Beasley and Terence Rushin/College of Computing
]]>Communications Officer for the School of Cybersecurity and Privacy
]]>The team also found that the new standard's core rules for sending messages and handling errors retain every known security weakness of older versions.
The vulnerabilities could allow an attacker who has already gained access to a vehicle's internal computer network to change or intercept messages, take individual vehicle computers offline, send different information to different systems, or disrupt network communication.
CAN Extra Long (CAN XL), the latest version of the controller area network (CAN), is designed for future cars and other vehicles that need to move large amounts of data between onboard computers. The technology is not yet widely deployed in production vehicles, giving manufacturers an opportunity to address security concerns before it becomes more common.
“Because it isn't widely deployed yet, we have a narrow window to get its security right,” said Associate Professor Saman Zonouz of the School of Cybersecurity and Privacy and the School of Electrical and Computer Engineering, one of the researchers on the project.
“Fixing a standard now is far easier than fixing it once it's built into the hardware of millions of cars that stay on the road for a decade or more.”
Zonouz said CAN XL is expected to become the primary version of CAN, the network that enables computers inside a vehicle to communicate. It is also expected to serve as the main network for driver-assistance systems in future vehicles.
The researchers confirmed the seven vulnerabilities in commercial CAN XL hardware and demonstrated attacks using a physical test setup that mimics a vehicle network.
Modern vehicles can have dozens of electronic control units that manage systems such as sensors, brakes, steering, and entertainment. These computers need to communicate quickly and reliably.
For decades, many vehicles have used CAN for this communication. However, the original version, known as classic CAN, was not designed to handle the volume of data generated by newer vehicle technologies.
CAN XL was developed to provide faster communication, larger messages, and new security features, but the researchers wanted to know whether the new standard’s basic rules were secure. Their testing revealed that they weren’t. They discovered that an attacker who controls one computer on the network could intercept and replace messages or take a targeted computer offline.
“These flaws are in the standard itself, so every device built to follow it inherits them,” Zonouz said. “With CAN XL, there's still time to fix these problems before they reach the road.”
The attacks would require an attacker to first gain control of a computer connected to the vehicle's internal network. The research does not show that CAN XL itself provides a means to break into a vehicle. Instead, it shows what an attacker could do after gaining access.
Several attacks were also faster and harder to detect than similar attacks against classic CAN, according to the authors.
The researchers proposed changes to the CAN XL rules to prevent several of the attacks. They also recommended additional security measures, including message authentication and systems that can detect unusual activity.
The team reported the vulnerabilities and informed the manufacturers of the commercial devices it tested about the bugs it found. One company has already released a fix. The researchers hope their findings will help manufacturers address security weaknesses before the technology becomes more widely used.
A Formal Security Analysis of CAN XL was published in the Proceedings of the 35th USENIX Security Symposium, held Aug. 12-14 in Baltimore, MD. The paper was also named a runner-up for a distinguished paper award.
The study was conducted by Georgia Tech Ph.D. student ZhaozhouTang, Professor Vijay Ganesh, Zonouz, and Provost and Executive Vice President for Academic Affairs Raheem Beyah, along with Khaled Serag of the Qatar Computing Research Institute and Z. Berkay Celik of Purdue University.
The paper and its findings result from projects sponsored by the Hyundai America Technical Center, the Georgia Department of Transportation, and the $7 million CHORUS Center through the National Science Foundation’s Cyber-Physical System Frontier Program.
]]>The agreement was signed during a Sept. 14 visit by PNNL leadership to Georgia Tech’s Advanced Manufacturing Pilot Facility. It builds on years of successful collaboration between researchers at the two institutions.
The enhanced collaboration has three primary goals: solving big problems through multidisciplinary, multi-institutional research; sustaining and engaging human capital by creating opportunities for students, faculty, researchers, and interns to work on complex real-world challenges; and accelerating technology adoption by taking scientific discoveries and innovations into industry and the federal marketplace.
"Strengthening our partnerships with national laboratories is a strategic priority for Georgia Tech,” said Tim Lieuwen, executive vice president for Research at Georgia Tech. "PNNL is an outstanding partner, and we look forward to deepening connections among our researchers and students, expanding opportunities and access to world-class capabilities and facilities, and advancing discovery, innovation, and national impact together."
PNNL is a U.S. Department of Energy national laboratory that advances scientific discovery and innovations that support reliable energy, national security, and economic competitiveness. Georgia Tech is one of the nation's leading research universities, conducting more than $1.5 billion in research annually and pursuing a mission to develop leaders who advance technology and improve the human condition.
"PNNL and Georgia Tech have a long history of working together," said Deb Gracio, laboratory director of Pacific Northwest National Laboratory. “This agreement provides the foundation to take on difficult research challenges and to leverage our complementary expertise and investments in the years ahead.”
The Strategic Energy Institute (SEI) has played a central role in advancing Georgia Tech's engagement with the national laboratory system and connecting researchers with partnership opportunities.
"As SEI takes on a greater role in coordinating Georgia Tech's national laboratory partnerships, this agreement with PNNL demonstrates how we are building stronger pathways for faculty, researchers, and students to engage with world-class national lab capabilities," said Yuanzhi Tang, SEI executive director.
The National Laboratory Partnership Program supports engagement between Georgia Tech and the U.S. national laboratories. SEI now leads the program, with ongoing support from Georgia Tech’s Office of the Vice President for Interdisciplinary Research. The agreement with PNNL is the first formal agreement established through the program since its transition to SEI.
"This MOU formalizes and strengthens relationships that have been growing for many years," said Scott McWhorter, SEI director of national laboratory partnerships. “It leverages the unique strengths of Georgia Tech and PNNL and provides a framework for expanding engagement across areas of mutual interest.”
Georgia Tech and PNNL bring complementary strengths in energy and environmental systems, electric grid resilience, cybersecurity of critical infrastructure, alternative fuels and advanced materials, biofuels, radiation detection, data science and visual analytics, nuclear nonproliferation, and national security. These capabilities will be brough together through two upcoming Genesis Mission projects.
Bill Pike, PNNL's deputy director for Science and Technology, who joined the laboratory’s leadership delegation for the MOU signing, said the projects reflect the growing partnership and expanding collaboration between the two institutions.
About Georgia Tech
The Georgia Institute of Technology is one of the nation's leading research universities, with strengths spanning engineering, computing, science, business, design, and the liberal arts. With more than $1.5 billion annually in research awards across its Colleges and the Georgia Tech Research Institute, Georgia Tech is among the nation's most research-intensive universities. The Institute's mission is to develop leaders who advance technology and improve the human condition.
About Pacific Northwest National Laboratory
Pacific Northwest National Laboratory draws on its distinguishing strengths in chemistry, Earth sciences, biology and data science to advance scientific knowledge and address challenges in energy resiliency and national security. Founded in 1965, PNNL is operated by Battelle and supported by the Office of Science of the U.S. Department of Energy. The Office of Science is the single largest supporter of basic research in the physical sciences in the United States and is working to address some of the most pressing challenges of our time.
The agreement was signed during a Sept. 14 visit by PNNL leadership to Georgia Tech’s Advanced Manufacturing Pilot Facility . It builds on years of successful collaboration between researchers at the two institutions.
]]>Saman Zonouz, an associate professor in Georgia Tech’s School of Cybersecurity and Privacy and the School of Electrical and Computer Engineering, is helping address those concerns as a member of a planning committee for the National Academies of Sciences, Engineering, and Medicine.
Since May 2026, Zonouz has served on the planning committee for an upcoming public workshop on integrating AI into the electricity system. The workshop is scheduled for Oct. 26 and will convene experts to discuss how to safely and reliably use AI across the power grid.
The workshop will explore how AI is used today, the challenges of integrating AI into existing power systems, and how people and AI can work together. The event will also examine relationships between AI and cybersecurity, data centers, workforce development, and future research.
For Zonouz, cybersecurity and resilience are key parts of the conversation.
“AI is both a shield and a target for the grid,” Zonouz said.
Zonouz leads Georgia Tech’s Cyber-Physical Systems Security Research Laboratory. The lab’s research, much of it funded by the U.S. Department of Energy, investigates how AI can protect power systems from cyberattacks.
One recent initiative combines AI with detailed knowledge about power system operations to detect attacks in near real time. Another project sought to identify security weaknesses in solar inverters and other devices that connect energy sources, such as solar panels, to the power grid.
Regardless of the challenge, Zonouz and his lab work to identify and fix these weaknesses before attackers can exploit them. Despite their best efforts, however, Zonouz said that some AI tools meant to secure a system can create new ways for attackers to target a power grid.
The lab has demonstrated in several studies that AI systems can be tricked by false or manipulated data. Attackers can also target the software and systems that AI depends on. As AI becomes more advanced, some systems may even be able to modify grid operations.
This raises an important question: What happens if an AI system fails or is attacked?
“That is why the right frame is resilience,” Zonouz said.
For Zonouz, the goal is not to create an AI system that can never fail or be attacked. It’s to build a trustworthy power grid that can continue operating safely even if an AI system fails or is compromised.
Zonouz will bring his perspective to the National Academies workshop. In addition to helping organize the event, he will lead a session on cybersecurity and resilience. He will also help summarize the workshop’s key takeaways during the closing session. The workshop will be held virtually in late October.
“Lessons learned from the electricity grid can be applied to other critical infrastructure as AI monitoring and control tools continue to advance and become more common,” he said.
]]>Communications Officer at the School of Cybersecurity and Privacy
]]>Most people would recognize that something unexpected has happened and adjust accordingly.
For machines, it’s not always that simple.
Researchers at Georgia Tech’s H. Milton Stewart School of Industrial and Systems Engineering (ISyE) are addressing this very challenge: how can autonomous systems recognize when something unexpected has happened, determine whether it matters, and decide how to respond?
In a recent paper published in the INFORMS Journal on Data Science, the research team introduced a new framework called Mutual Information Surprise, designed to help machines identify meaningful surprises in complex environments. The work could help lay the foundation for AI agents, robots, and autonomous systems that can better navigate unexpected situations, avoid costly mistakes, and make more informed decisions when faced with situations they weren’t designed or trained to handle.
"Traditional automation systems are designed to follow instructions. They do not have a sense of surprise. Machines follow a predetermined recipe and produce the expected result," said Yu Ding, the Anderson-Interface Chair and professor in ISyE, who is leading this project. " Intelligent agents, however, must do more than simply follow instructions. They need to recognize when their current understanding or reasoning process has become inadequate. Humans routinely use surprise for this purpose. An autonomous machine needs an analogous computational capability."
This research concept was jointly worked out by Xiao Liu, the David M. McKenney Family Associate Professor, as well as a former postdoctoral fellow in ISyE, Yinsong Wang, and Ph.D. student Quan Zeng.
While several methods already exist to measure surprise in machines, such as the "Shannon Surprise" and "Bayesian Surprise," the ISyE team saw limitations in both approaches.
Shannon Surprise tends to focus on the rarity of an event, while Bayesian Surprise focuses on how much a new observation changes a system's beliefs.
To illustrate the limitations of existing surprise measures, Ding pointed to the examples of someone noticing a specific type of car parked nearby or a student who consistently earns top grades but suddenly performs poorly on an exam.
Existing surprise measures don't always flag the right kinds of moments to be surprised, Ding said. To the examples given, the Shannon Surprise would flag the specific car parked nearby, although people typically wouldn’t be surprised by that event. Conversely, people would be surprised by the student performing poorly, but the Bayesian Surprise would not flag it.
"We introduce this new surprise definition because it measures the gain of knowledge and epistemic progression, thus capturing the right moments when a system should be surprised and ignoring the moments when it shouldn't," Ding added.
The framework could eventually help systems determine when to continue as planned or when to stop and reassess, whether that's an AI assistant struggling to answer a question, a robot encountering a situation it has never seen before, or an autonomous factory detecting a disruption that could affect production and require human intervention.
Looking ahead, the ISyE research team is exploring potential applications in engineering autonomous systems and is developing a proposal for federally funded programs.
]]>Despite decades of research, the chaotic nature of turbulence makes it difficult to predict and control. This challenge impedes more than smoother air travel. Solving the turbulence problem could lead to advances in areas ranging from sustainable energy to training a smarter workforce for the era of artificial intelligence (AI).
To advance understanding of turbulence, the National Science Foundation is establishing a new, $30 million Science and Technology Center (STC) at Michigan State University. Georgia Tech is among eight universities supporting the center.
Assistant Professor Qi Tang will join the STC for Transformative Explorations in Multi-Physics and Engineering of Scientific Turbulence (TEMPEST). TEMPEST, launching on Sept. 1, aims to build trustworthy, predictive models of real-world turbulence.
Tang will lead TEMPEST’s modeling and scientific machine learning (ML) efforts. NSF will fund Tang and Georgia Tech with over $1 million from the center’s five-year, $30 million award.
By understanding and predicting turbulence, TEMPEST can unlock new applications, with a focus on fusion energy and national security. The center will also generate long-term research projects with applications in air and space flight, manufacturing, chemistry, and broaden science education and AI fluency.
“Turbulence has resisted prediction for a century,” said Tang, a faculty member in the School of Computational Science and Engineering (CSE).
“STC TEMPEST aims to develop a unified, predictive science that advances fusion energy, improves hypersonic technologies, and deepens our understanding of how stars created the elements that make up our world and ourselves. For everyday people, the center points toward abundant clean energy and faster, more efficient flight."
Turbulence is the motion of a fluid characterized by chaotic changes in pressure and speed. Turbulent flows can occur across all scales, from interactions between subatomic particles to astrophysical scales, including supernovas, black holes, and cosmic rays.
While turbulent flows are common and occur naturally, scientists still do not fully understand them. Small changes in a turbulent flow can produce dramatically different outcomes. Combined with the countless interactions across multiple scales of time and space, this makes turbulence extraordinarily difficult to predict.
“Rather than studying individual pieces of this enormously complex problem in isolation, we are bringing together theory, experimentation, computation and artificial intelligence to develop a deeper understanding of turbulence across scales,” said Michael Murillo, an MSU professor and the director of TEMPEST.
“Our goal is not simply to understand turbulence better, but to make it predictable and controllable in ways that will enable new technologies and scientific discoveries.”
The NSF TEMPEST award supports students and researchers at Michigan State University, Auburn University, Baylor University, Georgia Tech, San José State University, Texas A&M University-Corpus Christi, University of Rochester, and Yale University to combine theory, computation, AI techniques, and experimentation to build trustworthy predictive models of real-world turbulence for high-consequence applications.
Additional partners include Los Alamos National Laboratory, Sandia National Laboratories, Lawrence Livermore National Laboratory, Pacific Fusion, and General Atomics.
Together, TEMPEST researchers will build more accurate and reliable physics-grounded models. The center will test new ideas against real-world observations, then use those results to refine the models.
Tang contributes expertise in ML, scientific computing, and plasma physics to STC TEMPEST.
Earlier this year, Tang received an Early Career Research award from the Department of Energy’s Office of Science. He is using the award to build ML and data science tools that help scientists analyze massive datasets from fusion experiments and simulations.
This interdisciplinary approach is intended to move science from understanding why turbulence behaves as it does to predicting how it will behave. Through prediction and simulation, scientists could eventually engineer solutions to control turbulence in real-world scenarios.
The center will make its data and software broadly available and engage the public through museum exhibitions, immersive media, and educational programs that are expected to reach more than 10,000 K-12 students annually. TEMPEST will also help train an AI-fluent scientific workforce prepared to tackle complex problems across disciplines.
“One reason I am excited to work in TEMPEST is because it aligns perfectly with our School of CSE mission. As a discipline, CSE complements theory and experimentation as a mode of scientific discovery,” Tang said.
“We build computational models to simulate scientific and engineering concepts, like turbulence in this case, so that we can test theories that are too difficult, expensive, or risky for physical experiments.”
]]>This award is part of NASA’s Space Technology Research Grants Program, which supports groundbreaking projects with the potential to transform space science and technology.
Through the three-year, $750,000 award, Mazumdar and her research team will develop laser-based methods to simultaneously measure temperature, chemical species, gas velocity, and other properties in the gases surrounding heat shield materials.
“We aim to implement our measurement techniques in arc jet or plasma jet facilities, like the ones at NASA, to assist with the evaluation of thermal protection systems for next-generation spacecraft,” Mazumdar said.
The project uses ultrafast lasers, which produce extremely short, high-intensity pulses. These pulses allow researchers to study the behavior of molecules on extremely short timescales and measure properties that can improve understanding of the gases surrounding a spacecraft during atmospheric entry.
Mazumdar said she and her students are excited about this project and hope it will help NASA develop new heat shields for spacecraft traveling to destinations such as Mars, Venus, and Titan.
Read Full Story on the ME Webpage
]]>This award is part of NASA’s Space Technology Research Grants Program, which supports groundbreaking projects with the potential to transform space science and technology.
Through the three-year, $750,000 award, Mazumdar and her research team will develop laser-based methods to simultaneously measure temperature, chemical species, gas velocity, and other properties in the gases surrounding heat shield materials.
“We aim to implement our measurement techniques in arc jet or plasma jet facilities, like the ones at NASA, to assist with the evaluation of thermal protection systems for next-generation spacecraft,” Mazumdar said.
]]>Communications Manager I
]]>Instead of becoming steadily stickier as they are pulled apart, soft hydrogels first become less adhesive and then more adhesive as the pulling speed increases, revealing a previously unknown "sweet spot" where detachment is easier. The discovery challenges a long-held assumption in adhesion mechanics and reveals a new physical mechanism that governs how soft materials attach and detach.
"We expected the adhesion to increase continuously with pulling speed, as predicted by classical theories," said Yuhang Hu, associate professor in the George W. Woodruff School of Mechanical Engineering and the School of Chemical and Biomolecular Engineering. "Instead, we found the exact opposite over a broad range of speeds. That told us there had to be a completely different physical process at work."
]]>George W. Woodruff School of Mechanical Engineering
]]>Eunhye Song, a Coca-Cola Foundation Early Career Professor and associate professor in the H. Milton Stewart School of Industrial and Systems Engineering (ISyE) at Georgia Tech, has been named vice president/president-elect of the INFORMS Simulation (I-SIM) Society, recognizing her leadership and growing impact in the field of simulation.
I-SIM is a specialized community within the Institute for Operations Research and the Management Sciences (INFORMS) dedicated to advancing simulation research and practice while fostering collaboration and professional development among members of the simulation community.
Simulation has become increasingly important as advances in computing and data availability enable researchers to build more detailed models of complex real-world systems, helping organizations evaluate decisions before implementing them and avoid costly trial and error.
“We have much more data, computing power and the ability to build more detailed simulation models than ever before,” Song said. “That’s exciting because we’re getting closer to accurately representing what’s happening in the real system.”
Song’s connection to the simulation community spans over a decade. She first attended I-SIM's flagship Winter Simulation Conference as a first-year Ph.D. student in 2013 and has participated ever since. After joining the faculty ranks in 2017 – first at Penn State University between 2017 and 2022, then at ISyE where she joined in 2022 as an assistant professor – she became increasingly involved in service and leadership activities within the society.
As vice president/president-elect, which is a two-year term that began in July, Song will help guide the organization’s priorities and support initiatives that advance simulation research while identifying emerging areas of focus for the field.
“I’m honored to help the society think about the most relevant and emerging areas in simulation,” Song said.
Beyond its research mission, Song said the simulation community has played a key role in shaping her professional journey through mentorship, collaboration and service.
"I-SIM is a very tight-knit community, and I have received so much advice and mentorship from previous and current members,” Song said. “By the time it’s your turn to lead, everyone is happy to step up. I look forward to helping continue that great tradition in this new role.”
With more than 12,000 members worldwide, INFORMS is the leading international association for professionals in operations research, analytics, management science, and related fields. The organization advances the science and technology of decision making through publications, conferences, competitions, networking communities, and professional development services.
Communications Manager I
]]>Recycled materials promise a cleaner future, but recycled content alone does not necessarily make a product sustainable. At Georgia Tech’s Daedalus Lab, assistant professor, National Science Foundation CAREER Award recipient, and Brook Byers Institute for Sustainable Systems Faculty Fellow Christos Athanasiou and postdoctoral researcher Danqi Sun are working to provide greater certainty to designers and engineers by rethinking how materials are tested for their fracture characteristics. Their article in Science Advances details a new testing protocol that reduces cost, increases speed, and simulates real-world conditions.
Materials shape nearly every part of modern life, from packaging and consumer products to bridges and medical devices. Choosing the right material involves balancing durability, cost, manufacturability, and environmental impact. Yet those decisions are not always guided by a clear understanding of how materials age and fail under real-world conditions, especially for recycled materials like plastics.
One way that materials frequently fail is by cracking. A small crack can begin almost invisibly. Over time, it can spread from regular wear and tear and exposure to common environmental factors like moisture, temperature fluctuations, or even dirt. Eventually, the crack expands, and the part fails. Engineers have studied fracture for more than a century, but connecting the science of cracking to practical decisions about sustainability remains a major challenge.
The characteristics of recycled plastics often vary from those of the same material in unrecycled, or virgin, form. Products made from recycled plastics may be sold with sustainability claims under the assumption that they will perform as if they were made with virgin material. However, premature failure requiring repair or replacement can quickly change the sustainability equation as well as the acceptance of recycled materials by manufacturers and consumers.
Materials rarely fail due to a single factor. They may be exposed to several factors simultaneously, such as mechanical loading, chemical environments, temperature changes, moisture, and time. Traditional fracture protocols test one specimen at a time under carefully controlled laboratory conditions, which bear little resemblance to the real world. To move beyond this limitation, Sun developed an in-situ, high-throughput platform capable of studying how materials degrade and fail under more realistic conditions.
The platform changes conventional fracture testing in three important ways. First, it can test multiple specimens simultaneously rather than one at a time. By monitoring samples in parallel, testing time can be reduced by more than 60%. Second, it allows materials to be tested in realistic environments. In this study, researchers examined virgin and recycled plastics in alkaline environments that resemble conditions encountered in applications like landfill liner membranes and geotextiles. Third, the platform incorporates an imaging technique known as photoelasticity, which reveals the formation of stress fields that form around the origin of a newly developing crack. This allows researchers to see cracks develop earlier than before, giving them a clearer picture of the forces that drive crack growth.
The researchers have made the technology available for licensing through Georgia Tech’s Office of Technology Licensing. “Our goal was to make fracture testing not only faster but also more informative,” Sun said. “By combining high-throughput testing, realistic environments, and full-field stress imaging, we can better understand how materials fail under conditions closer to real-world applications.”
Recycled plastics are often viewed as a greener choice. But according to the study, it’s not always so straightforward. If a recycled product fails prematurely and needs to be replaced frequently, its environmental and economic costs can increase despite its recycled content. As Athanasiou puts it, “Failing materials don’t just break products. They can break sustainability promises.”
For example, comparing virgin polyethylene terephthalate (PET) with recycled PET (rPET) in applications such as landfill geotextiles, the researchers discovered that rPET showed lower resistance to environmental stressors, particularly in alkaline conditions over a pH of 9. In this application, specifying rPET over virgin PET would likely eliminate all of the presumed economic and environmental advantages of using a recycled material.
“Recycling is essential, but recycled content alone does not tell the full story. If a material fails too soon, the environmental benefits can disappear,” Athanasiou said.
For the researchers, the significance of the work extends beyond recycled plastics. The broader goal is to provide a fast, affordable, and realistic platform for evaluating the sustainability of any material choice. Because current testing protocols are costly, not widely available, and limited in the information they yield, engineers, manufacturers, and policymakers often have little choice but to continue to specify non-recycled materials because they will perform as expected. Having cheap and accurate data on recycled materials will help to accelerate their adoption because matching the engineering properties of recycled materials to their most appropriate applications will become more obvious.
The researchers also hope to expand the platform to simulate even more complex environments and apply it to a wider range of materials. Because the system generates large amounts of detailed data, it may enable opportunities to use computational modeling or artificial intelligence to digitally simulate mechanical testing, driving down costs and expanding availability even more.
The larger vision is a future in which sustainability is judged not by labels or assumptions, but by evidence for how a material performs, how long it lasts, how it fails, and what it costs society and the environment over its full lifetime.
Please visit the Daedalus Lab YouTube channel to see an explainer video about this new testing protocol: https://youtube.com/watch?v=zmRhiRIiAkQ
Read the paper here: https://www.science.org/doi/10.1126/sciadv.aeh0456
This research was supported by the National Science Foundation CAREER Award No. 2338508.
]]>Artificial intelligence (AI) is changing far more than hospital paperwork. Researchers say it's beginning to help doctors make clinical decisions, identify mental health risks earlier, expand care through smartphones and wearable devices, and extend healthcare to people who might otherwise go without it.
While adoption varies across healthcare systems, Georgia Tech researchers say AI is becoming increasingly integrated into medicine. At the same time, questions about privacy, trust, equity, and human oversight remain central to how the technology will be used in the years ahead.
“Medicine has clearly embraced the human-AI fusion approach,” said Jon Duke, director of the Center for Health Analytics & Informatics at the Georgia Tech Research Institute. “The use of AI is encouraged, but reliance on AI is not.”
From Clinical Workflows to Decision Support
Many of healthcare's earliest AI applications focused on reducing administrative burdens. Today, researchers say the technology is increasingly helping clinicians access and interpret information.
Duke points to the growing use of clinical AI tools such as OpenEvidence, which provide healthcare professionals with AI-generated, source-backed answers to medical questions.
“AI is moving beyond administrative tasks and beginning to play a meaningful role in clinical decision support,” Duke said.
Within electronic health records, AI is being used to generate clinical notes from patient-provider conversations and summarize complex patient charts. While healthcare systems have moved most quickly on efficiency-focused applications, Duke said AI is steadily becoming part of everyday clinical practice.
Still, he believes human oversight will remain essential.
“One of the biggest long-term questions is whether physicians trained alongside AI will develop the same level of independent clinical judgment as previous generations,” he said.
Expanding Access Beyond the Doctor's Office
Researchers also see AI creating new opportunities outside hospitals and clinics.
Munmun De Choudhury, professor in the School of Interactive Computing, said AI is helping public health systems identify emerging mental health concerns by analyzing digital and clinical data.
“AI is enabling public health systems to move from reactive to more proactive approaches by identifying emerging mental health concerns from diverse digital and clinical data sources,” De Choudhury said.
Those insights may create opportunities for earlier intervention.
“AI has the potential to identify subtle behavioral changes that precede worsening mental health, creating opportunities for earlier support before someone reaches a crisis,” she said.
Alexander Adams, assistant professor in the School of Interactive Computing, studies wearable sensing systems, remote patient monitoring technologies, and point-of-care health tools. He said advances in AI are accelerating innovation in areas such as pharmaceuticals, medical imaging, simulation, and biomarker discovery.
“The biggest changes I am seeing are the increased productivity in pharmaceuticals, medical imaging, simulation, and biomarker discovery,” Adams said. “That has a downstream impact on what we can measure at the point of care.”
According to Adams, smartphones and wearable devices have become increasingly important healthcare tools because of the amount of data they collect.
“No stand-alone medical devices generate nearly as much data as smartphones and smartwatches, making them a natural place for modern AI,” he said.
He also sees significant potential for AI-enabled technologies to improve access in rural and underserved communities.
“These technologies are especially important for underserved and rural populations that often have less access to care and face higher risks of complications,” Adams said.
Balancing Innovation and Human Connection
Despite growing interest in AI, researchers caution that adoption is not without challenges.
“The real opportunity lies not in building more autonomous AI, but in designing systems that strengthen human judgment, clinical expertise, and community care,” De Choudhury said.
Although AI has advanced rapidly, its impact on chronic disease management has remained limited, said Rosa Arriaga, professor in the School of Interactive Computing.
“AI holds a lot of promise but hasn't delivered much in the way of real-world applications,” Arriaga said.
She also raised concerns about how increased reliance on data-driven tools could affect interactions between clinicians and patients.
“The fear is that clinicians become ‘data checkers’ and have even less interaction with patients,” she said.
At the same time, Arriaga said AI is having its greatest impact in the mental wellness space.
“There are now randomized controlled trials showing that some people are willing to receive ‘therapy’ from an AI agent and that this intervention is beneficial,” she said.
However, she cautioned that those developments should be considered alongside potential risks.
“Greater interactions with AI may lead to greater isolation, and parasocial relationships with AI may erode the social fabric,” Arriaga said.
Shaping the Next Era of Healthcare
Despite differing views on the opportunities and challenges, the researchers see AI becoming a lasting part of healthcare.
“AI is accelerating everything,” Adams said. “Even through rapid iteration and simulation alone, it will accelerate medical devices and point-of-care technologies.”
As healthcare organizations continue to explore where AI can improve care, researchers say the future of medicine will depend not only on technological advances but also on how effectively those tools support the people who deliver and receive care.
Dunkin is an IISE Fellow, the highest classification of IISE membership that recognizes outstanding leaders of the profession who have made significant, nationally recognized contributions to industrial and systems engineering. She has served as chief information officer at the U.S. Department of Energy and the U.S. Environmental Protection Agency, as well as chief strategy and innovation officer at Dell Technologies. She has also held numerous leadership positions at Hewlett Packard, largely in the areas of manufacturing, research and development and IT. She currently serves as a distinguished external fellow with Georgia Tech’s Strategic Energy Institute.
Dunkin holds appointments as a distinguished professor of the practice in Georgia Tech’s School of Cybersecurity and Privacy and in the Jimmy and Rosalynn Carter School of Public Policy. She is also a distinguished external fellow at the Georgia Tech Research Institute (GTRI).
“The Emerging Technologies forum is a great opportunity for practicing ISEs to get exposure to a range of emerging technologies in a short period of time and identify those that they want to experiment with and implement in their organizations,” Dunkin said.
The Aug. 26 event, which will be held virtually from 1-5 p.m. EDT, will explore technologies shaping the future of industrial and systems engineering, including artificial intelligence, digital twins, and robotics and autonomy.
As event chair, Dunkin will be joined by experts from the healthcare, aerospace, logistics, manufacturing and technology industries, including representatives from Royal Caribbean, The Boeing Company, UPS and other organizations.
The complimentary event is open to engineers and improvement professionals of all backgrounds. Registration is available through the IISE Emerging Technologies Forum website.
Six years after demonstrating that membranes could separate crude oil at the molecular level, Georgia Tech researcher Ryan Lively is part of an international team that has taken the concept a significant step further.
The team, including investigators at the Korea Advanced Institute of Science and Technology (KAIST), discovered that a membrane material widely believed to be non-selective for molecules as small as those found in crude can in fact selectively separate crude oil into lighter and heavier fractions in a way researchers did not expect.
Published in Nature, their findings suggest that using membranes to separate crude oil before distillation could significantly reduce the energy, water, and carbon footprint of petroleum refining.
Lively, the Thomas C. DeLoach Jr. Endowed Professor in Georgia Tech's School of Chemical and Biomolecular Engineering, served as an advisor and corresponding author on the study. Dong-Yeun Koh, an associate professor at KAIST and a former postdoc in the Lively Lab at Georgia Tech, led the study.
Building on Earlier Research
In the 2020 Science paper, Lively and collaborators demonstrated that specially designed membranes could separate crude oil into valuable fractions without relying solely on traditional heat-driven distillation. The work helped establish membrane-based crude oil fractionation as a promising alternative for reducing energy use in refining.
"This work grew directly out of the challenges we identified in our original findings in the 2020 article," Lively said. "One of the key challenges that the KAIST team set out to tackle was the very low oil productivities of the membrane units, which has limited the ability of this concept to leave the lab. Along the way, we not only increased the productivities, but we also uncovered a surprising new mechanism that could make membrane-based crude oil separations even more practical.”
The new study built on that foundation. The researchers investigated polyacrylonitrile (PAN) membranes, a material commonly used as a non-selective support layer in filtration systems. Because the material is porous, the team generally did not expect it to perform precise molecular separations on its own.
But what they found surprised them, Lively said. As crude oil flowed through the membrane, heavier hydrocarbon molecules accumulated within the membrane's pores. Instead of clogging the membrane, the buildup created a stable internal layer that gradually narrowed the pathways through which molecules could travel. Surprisingly, the molecules that caused the buildup in the first place were eventually excluded from entering the membrane, resulting in a steady production of higher quality oil through the narrow pathways that remained.
In effect, the membrane created its own molecular-scale filter. The result was a process that allowed lighter hydrocarbons to pass through while holding back heavier components.
The membrane enriched lighter fractions such as naphtha and kerosene while achieving crude oil flow rates more than 23 times higher those reported in the 2020 paper for whole crude oils
When Buildup Becomes an Asset
In most filtration systems, buildup inside a membrane (or fouling) is considered a problem because it reduces performance.
But according to the researchers, this study demonstrates that something different can happen under the right conditions.
Using a range of analytical techniques, the researchers found that long-chain hydrocarbon molecules accumulated inside the membrane and became an essential part of the separation process. The deposits effectively transformed larger pores into stable transport pathways measuring less than two nanometers across, they deduced based on available experimental evidence.
The membrane maintained consistent separation performance during four weeks of continuous operation, suggesting the filtration pathways remained stable over time.
“The findings challenge traditional assumptions about membrane fouling and may offer new opportunities for designing industrial separation systems that take advantage of similar behavior,” Lively said.
Potential Impact on Refining
Today's refineries heat entire streams of crude oil to separate them into useful products. By using membranes to remove a substantial portion of the lighter hydrocarbons before distillation, refineries could reduce the amount of material that must undergo energy-intensive heating. Alternatively, the refinery can use the membranes to incrementally increase refinery capacity, which is currently not possible using large-scale distillation equipment.
To evaluate the potential impacts of the membrane system, the researchers modeled a refinery process that incorporated a membrane separation step before conventional distillation.
“This study reveals a new scientific principle in which a membrane interacts with a complex mixture and spontaneously forms its own separation channels," Koh said. "Working with real crude oil supplied by HD Hyundai Oilbank allowed us to validate the technology under conditions relevant to industrial operation.”
The team's technoeconomic analysis showed that incorporating the membrane process could reduce distillation energy use by 30%, carbon dioxide emissions by 35%, and water consumption by 20%.
Applied across U.S. atmospheric crude distillation capacity — about 18 million barrels per day — those savings would be equivalent to powering roughly 2.2 million homes, removing about 3 million passenger vehicles from the road, and supplying enough water for approximately 660,000 people each year.
"Turning crude oil into useful products has relied on essentially the same basic approach for more than a century," Lively said. "Membranes offer a path toward achieving those separations with dramatically lower energy requirements and emissions."
The study's findings also suggest that the phenomenon may not be limited to a single membrane chemistry. Researchers observed similar behavior in a second membrane material, raising the possibility that the approach could be extended to other membrane systems.
"This is a terrific piece of research that rewards curiosity," said Andrew LIvington, vice president of research and innovation and professor at Queen Mary University of London, who was not involved with the study. "This work adds significantly to the field of membrane separations of crude oil streams as it tackles the first, hard to achieve separation of heavy hydrocarbons – most work to date has focused on lighter oils – and it uses a simple and readily available membrane."
CITATION:
Jihoon Choi, Hyeokjun Seo, Minyong Lee, Woong-Chul Shin, Jaemin Choi, Keonwoo Choi, Min-Jun Jang, Sung Gap Im, Jae W. Lee, Ryan P. Lively, and Dong-Yeun Koh, "Crude oil fractionation by means of mesoporous polyacrylonitrile membranes," Nature, 2026.
That principle is especially relevant today across the artificial intelligence (AI) landscape. As systems scale, they increasingly become harder to measure, compare, and fix, particularly within proprietary environments.
A team led by Georgia Tech, working with collaborators across industry, has developed a new approached called Chakra to bring greater clarity to complex AI systems.
“Imagine a room where everyone is trying to collaborate, but each person speaks a different language,” said Tushar Krishna, an associate professor in the School of Electrical and Computer Engineering, who is leading the effort. “That’s a bit like today’s AI ecosystem. The internet worked because it was built on shared practices and protocols. In AI, we’re still building that kind of common foundation.”
The work, which Krishna leads through the nonprofit MLCommons, was released alongside a paper at the 2026 Conference on Machine Learning and Systems(MLSys) in Bellevue, Wash.
Understanding Systems Without Exposing Them
Cloud companies, chip designers, software developers, and infrastructure providers all describe their systems differently, relying largely on internal, proprietary approaches that are not publicly shared.
This slows innovation, reduces efficiency, and increases the cost of running AI at scale.
Chakra, named after the Sanskrit word for “wheel” to reflect a continuous cycle of improvement, is designed around that reality. Its release is not a single finished system, but a set of shared tools and building blocks.
Researchers are making available a standardized format for representing AI workloads, along with tools for collecting and analyzing data from what’s known as an execution trace.
“An execution trace is essentially a recording of how an AI system behaves,” Krishna said. “Rather than focusing only on outcomes like speed or accuracy, it captures what computations happened, when machines needed to communicate, and where delays or bottlenecks occurred.”
Those traces don’t expose the underlying code or data. Instead, they reflect patterns of behavior.
Those traces don’t expose the underlying code or data. Instead, they reflect patterns of behavior.
“It’s a bit like sharing a map of traffic patterns in a city, instead of handing over the blueprints for every building,” Krishna said.
The approach can also be used to explore how future systems might behave, giving researchers a way to test ideas and identify potential bottlenecks before those systems are built.
“All of this dramatically lowers the barrier to participating in AI systems innovation,” Krishna said.
Building a Shared Standard
The Chakra project began in 2023 as a collaboration between Georgia Tech and Meta, building on parallel efforts to better understand how AI workloads behave across production systems and simulation environments.
Part of that work built on ASTRA-sim, an open-source distributed AI system simulator developed and maintained by Krishna’s group, which models how large-scale AI systems perform across hardware and software.
“We knew that for AI to scale responsibly, we needed better ways to understand what’s happening under the hood,” Krishna said. “Companies struggle to compare systems fairly or reproduce why something worked well—or failed—because everyone uses different tools and proprietary setups.”
The early collaboration expanded into a broader effort called the Chakra Working Group (CWG) within MLCommons, a consortium that brings together companies and researchers to develop shared benchmarks and standards for AI systems, including widely used efforts like MLPerf.
David Kanter, co-founder of MLCommons and head of MLPerf, has praised the group for “defining an industry roadmap for AI workload tracing support and benchmarking.”
Today, CWG includes industry partners such as NVIDIA, AMD, Meta, HPE, and Keysight, along with contributions from multiple Georgia Tech faculty, students, and alumni (seven of whom are now working across partner organizations).
“Chakra is a fantastic showcase of the role ECE and Georgia Tech play in connecting academic research with real-world systems,” said Arijit Raychowdhury, Steve W. Chaddick School Chair of ECE. “We can bring together expertise spanning the full AI stack in really the only way that makes complex work like this possible.”
That level of collaboration is essential to developing something that can be used across the broader AI ecosystem, according to Winston Liu, a chief architect at Keysight Technologies and a member of CWG.
“What the Chakra community has built is meaningful, but the collaboration model that produced it is worth recognizing just as much,” he said. “That combination—early enough to shape the spec together and open enough that the output belongs to everyone—is genuinely rare.”
A Real-world Testbed at Georgia Tech
Much of the team’s work has depended on access to infrastructure capable of running AI systems at a realistic scale. Georgia Tech has built that capability through its AI Makerspace, one of the largest computing clusters in the world dedicated to supporting student-driven AI workloads while also serving as a real-world testbed for large-scale systems research.
In collaboration with the Partnership for an Advanced Computing Environment (PACE), CWG researchers utilized the AI Makerspace to run workloads across 128 advanced GPUs and collect execution traces from systems operating under real conditions.
“The AI Makerspace was built on a simple belief: AI should be accessible to as many as possible,” said Matthieu Bloch, associate dean in the College of Engineering. “It’s exciting to see our colleagues using it to amplify impact and give back to the broader community.”
That level of access allowed the work behind Chakra to move beyond theory and into environments where performance challenges actually emerge.
In one case study, Chakra helped identify a hidden communication bottleneck that only appeared under realistic conditions when different types of workloads were running at the same time. More simplified tests failed to surface the issue.
What Comes Next
As the Chakra tools and standards are released, the focus now turns to how they will be adopted and extended.
Krishna sees the current moment less as a finish line and more as a starting point for broader participation across the field.
“Five years from now, Chakra will help make AI systems development dramatically more reproducible and accessible,” he said. “Researchers could test ideas against realistic workloads without needing access to massive datacenters, and companies could identify problems much earlier in the design process.”
As AI infrastructure grows more costly, the ability to model new system designs allows researchers and companies to make informed decisions before committing to large-scale investments.
“Longer term, it could move us toward a ‘digital twin’ of AI infrastructure,” Krishna said. “A way to model and optimize systems before they’re ever built.”
Now, in a project led by the Georgia Tech Research Institute (GTRI) in collaboration with the Georgia Institute of Technology (Georgia Tech) researchers are closing the gap between Earth-based simulations and the true space environment by sending experimental materials to the International Space Station (ISS) for several months of in-orbit exposure. In a rare chance for space research, where most hardware is either left in orbit or burns up on reentry, they are getting those samples back for detailed analysis on Earth.
The materials are set to launch to the ISS in the near future as part of the Materials International Space Station Experiment 22 (MISSE-22), a testbed attached to the outside of the station. Mounted on the forward-facing side of the ISS to ensure predominant exposure to highly corrosive atomic oxygen, the test samples will spend several months enduring the extreme temperatures, radiation, and reactive environment of low Earth orbit. The team is testing a selection of lightweight, research-grade polymers designed to survive these harsh conditions. Once the samples return to Earth, engineers will examine how they held up and use that data to enhance the strategic of future satellite constellations.
This project represents a collaboration across government, academia, and industry, bringing together GTRI, Georgia Tech, the Air Force Research Laboratory (AFRL), the University of Texas at El Paso (UTEP), a California-based R&D firm Hedgefog Research Inc., and DuPont de Nemours, Inc. The research is also supported by Aegis Aerospace, which owns and operates the MISSE Flight Facility platform aboard the ISS.
Harsh conditions in low Earth orbit — the region of space extending from approximately 100 miles to over 1,000 miles above Earth, where many satellites and the ISS travel — can darken, roughen, and weaken spacecraft surfaces over time. That damage shortens satellite lifetimes and requires engineers to add extra layers of protection, increasing overall logistical burden and mission costs.
Optimizing material durability is a strategic necessity, explained Elena Plis, a GTRI senior research engineer and principal investigator for the project, because every additional unit of shielding increases the cost of getting to orbit. To design lighter, more resilient materials, researchers need to examine how they degrade in a true space environment. However, most hardware is built for a one-way trip — designed to operate in orbit and then burn up on reentry, taking that valuable material data with it.
“The beauty of this type of experiment is that the materials return to Earth,” said Plis, who is also an affiliate of the Georgia Tech Space Research Institute. “For many missions, stuff is sent up and never seen again. Being able to test returned samples from real space conditions is unique, and I can’t stress enough how exciting that is for us.”
Instead of relying on familiar spacecraft materials like DuPont’s Kapton — a tough, heat-resistant polyimide plastic film that has coated spacecraft exteriors since the Apollo era — the team is sending up a set of new, lightweight, research-grade polymers. These materials are designed to improve the survivability of assets against space’s unforgiving elements.
Plis and her collaborators started with dozens of candidate materials they developed. To earn a spot on the MISSE-22, a sample has to be transparent or translucent, so light can pass through it, and researchers can examine how its optical properties change in orbit. The materials also have to be tough enough to withstand intense atomic oxygen exposure without fragmenting, which would create debris near the ISS. In the end, only a select number of the team’s materials made the cut.
The MISSE-22 testbed holds multiple experimental polymers. Instead of standard illumination, the team constructed a custom on-orbit polariscope: LEDs beneath each sample shine polarized light up through the material. A small camera system then slides over the top to capture these highly specific optical changes on a set schedule over the course of several months in space.
Using polarized light and machine learning to rapidly analyze color patterns in the images they receive from orbit, the researchers can track how stress inside each sample changes over time. Periodically, the system will cycle through the materials, and the images will be downlinked to Earth.
When the extended mission ends and the samples return, the team will compare those in-orbit measurements with detailed lab tests on the actual pieces that flew. Without returned materials, they would only have images and sensor data to work from. By testing the same samples in the lab, they can check how accurate the remote measurements really are and refine their methods.
If the materials perform as expected, the results could help engineers design satellites that last longer in orbit without carrying so much protective weight —providing a significant technological advantage in space domain awareness and asset longevity.
The Space Research Institute (SRI) at the Georgia Institute of Technology is an interdisciplinary hub that unites faculty, staff, and students to advance research, education, and collaboration in space science and technology. Bringing together expertise across engineering, science, policy, and the humanities, SRI drives innovative projects in areas such as astrophysics, aerospace systems, astrobiology, and space policy while fostering partnerships with academia, industry, and government. As Georgia Tech’s central nexus for space-related initiatives, SRI is committed to advancing discovery, developing the future workforce, and expanding humanity’s understanding of space and its impact on life on Earth. Learn more at space.gatech.edu.
]]>
Communications Officer II
Georgia Tech Research Institute
Christopher Zuo never thought of himself as someone mosquitoes singled out. They bit him from time to time, he said, but no more than anyone else who spent a lot of time outdoors.
“I don’t know if I would say I’m prone,” Zuo said. “I do get bitten, but I also think that’s partly because I’m just outside a lot more.”
However, that assumption did not hold up once he stepped inside a sealed mosquito chamber as part of a Georgia Tech research study.
Zuo, a Georgia Tech alum and co-author on the study, worked alongside Georgia Tech faculty member David Hu and researchers in Hu’s fluid dynamics lab — and co-authors Chenyi Fei, Alex Cohen, Jorn Dunkel from the Massachusetts Institute of Technology — on a multi-year effort to understand how mosquitoes locate people. Using high-speed cameras, careful controls, and mathematical modeling, the research examined how mosquitoes respond to carbon dioxide and visual cues. To confirm whether the data reflected real-world behavior, the team needed a human subject.
Zuo volunteered.
Before entering the chamber, he knew the mosquitoes were safe. They had been raised in a laboratory environment and were carefully controlled, making the experiment safer than being outdoors during peak mosquito activity.
“We knew exactly how all of these mosquitoes were reared, so we knew they’re disease-free,” he said. “Honestly, even if I got bitten 100 plus times, the actual danger that I was in was very little.”
Wearing a mesh suit, Zuo stood nearly motionless inside the chamber while mosquitoes were released and flew freely around him. Any movement could disrupt the data, so remaining still was critical even as mosquitoes gathered close to his face and upper body.
The response was immediate.
“You release the mosquitoes, and they’re already on top of you,” Zuo said. “Almost felt like it was instant.”
What surprised him most was not the bites but the sound.
“I didn’t realize how loud they were,” he said. “When they’re flying around your head, it’s that annoying buzzing sound. I didn’t realize how annoying it can get with just enough mosquitoes flying around.”
The experience was not limited to a single trial. Zuo entered the chamber multiple times as the research progressed, testing different variables including posture, clothing, and body positioning. In some experiments, he was required to hold his arms extended so cameras could capture a consistent silhouette.
“It felt more like an exercise at the gym,” Zuo said. “I was very much more focused on keeping my arms up and being as still as possible.”
Across those repeated interactions, patterns emerged that closely matched what the data predicted. Mosquitoes found him quickly, clustered in specific areas, and lingered only when certain conditions aligned.
“And once the conditions were right,” Zuo said, “they stayed.”
Zuo’s role helped bridge the gap between abstract modeling and human experience. It also challenged common assumptions about mosquito behavior that many people take for granted.
What follows are some of the most common mosquito myths, and what the Georgia Tech research and Zuo’s firsthand experience actually showed.
Myth: Mosquitoes swarm because they are following each other.
Reality: Mosquitoes respond independently to the same cues, which creates the appearance of swarming.
Trajectory data collected during the experiments showed no evidence that mosquitoes were coordinating or communicating with one another. Zuo explained that what people often describe as swarming is the result of multiple mosquitoes responding simultaneously to the same environmental signals. When carbon dioxide and a clear visual target are present, many mosquitoes converge on the same area independently. Zuo compared it to people arriving separately at the same crowded place because something there is attractive, not because they are following the crowd.
Myth: Mosquitoes randomly target different parts of the body.
Reality: In this study, mosquitoes concentrated near the head and shoulders, but only for the species observed, which is present in parts of the Southeast.
The Georgia Tech experiments focused on Aedes aegypti (dengue or yellow fever mosquito), a species found in parts of Georgia and other areas of the southeastern United States. Within that species, both trajectory data and Zuo’s experience inside the chamber showed mosquitoes repeatedly clustering near the head and shoulders rather than distributing evenly across the body. Zuo observed this pattern while standing still in the mesh suit, as mosquitoes returned again and again to his upper body. The study also confirmed previous biting studies showing that Aedes aegypti mosquitoes target the upper body, while other mosquitoes might focus on other areas. Researchers linked the behavior to carbon dioxide released through breathing near the mouth and nose, paired with a strong visual target. Zuo emphasized that other mosquito species behave differently and that these findings should not be applied to all mosquitoes.
Myth: Carbon dioxide alone explains why mosquitoes find people.
Reality: Carbon dioxide and visual cues work together, and neither is enough on its own.
Zuo described experiments that isolated carbon dioxide using inanimate objects before introducing a human subject. Carbon dioxide alone helped mosquitoes locate the general area of a target but did not consistently keep them there. Visual cues alone helped mosquitoes recognize an object but did not hold their attention. When both signals were combined, mosquito behavior changed significantly. The research showed the response was nonlinear, meaning the combined effect was stronger than simply adding the two cues together.
Myth: Once mosquitoes find a target, they always stay nearby.
Reality: Mosquitoes do not linger unless conditions align.
The data showed that mosquitoes often passed by targets unless both carbon dioxide and visual signals were present at the same time. Zuo observed that once those conditions aligned during the mesh suit experiments, mosquitoes stayed close and returned repeatedly to the same areas. Without the full set of cues, they were less likely to remain focused on a target.
Myth: All mosquitoes behave the same way.
Reality: Mosquito behavior varies by species and environmental conditions.
Aedes aegypti, Zuo described, are capable of feeding in well-lit conditions rather than relying solely on dusk. He contrasted this with Anopheles (marsh) mosquitoes, which require darker conditions and are closely tied to light and dark cycles during experiments. Zuo emphasized that the findings reflect the behavior of a single species and that different mosquito species respond to different cues.
While the Georgia Tech research explains how mosquitoes locate people, the Centers for Disease Control and Prevention (CDC) outlines steps people can take during mosquito season to reduce the risk of bites.
The CDC recommends using Environmental Protection Agency-registered insect repellents on exposed skin and wearing loose-fitting, long-sleeved shirts and long pants. Clothing and gear can also be treated with permethrin, which is designed for use on fabrics and not directly on skin. The agency also advises controlling mosquitoes indoors and outdoors by eliminating standing water and keeping window and door screens in good repair. The CDC notes that mosquitoes can bite during the day or night, depending on the species, and encourages precautions whenever mosquitoes are active.
Institute Communications
]]>