UCF Researchers Win DARPA Grant to Improve Autonomous Systems

Researchers at the University of Central Florida have received a $1.2 million grant from the Defense Advanced Research Projects Agency (DARPA) to improve the training of autonomous systems, such as drones and self-driving cars. The grant will fund a 1.5-year project led by George Atia and Yue Wang, associate professors in the Department of Electrical and Computer Engineering. The goal is to develop artificial intelligence-based technologies that can help these systems better adapt to unknown variables and unexpected situations.

Currently, training autonomous systems can take months or even years, and they often struggle to navigate real-world scenarios due to a “simulation-to-real gap.” The UCF team’s project, titled “Distributionally Robust Approaches to Transfer Learning,” aims to bridge this gap by designing technology that can train quickly and efficiently. This research has the potential to revolutionize various industries, including healthcare and autonomous driving, by enabling transformative approaches to complex problems.

Improving Autonomous Systems Training with AI Technologies

The development of artificial intelligence (AI) technologies is crucial for advancing autonomous systems, such as drones and self-driving cars. These systems rely on modeling and simulation to learn, but the training process can be lengthy, taking months to years, and often fails to account for real-world uncertainties. The Defense Advanced Research Projects Agency (DARPA) has recognized this limitation and launched the Transfer Learning from Imprecise and Abstract Models to Autonomous Technologies (TIAMAT) program to address it.

Researchers George Atia and Yue Wang from the University of Central Florida’s Department of Electrical Engineering have received a $1.2 million DARPA grant to develop AI-based technologies that can help autonomous systems better adapt to unknown variables. Their project, titled “Distributionally Robust Approaches to Transfer Learning,” aims to improve the simulation-to-real gap, which refers to the difference between the performance of autonomous systems in simulated environments and real-world scenarios.

The current simulation environments used for training autonomous systems may be complex and realistic, but they often fail to account for unexpected events or changes in environmental conditions. For instance, a drone flying from a city to a coast may not know how to cope with changes in flight dynamics or lighting. The UCF team plans to develop technologies that can train quickly and efficiently, enabling autonomous systems to adapt to new situations more effectively.

Bridging the Simulation-to-Real Gap

The simulation-to-real gap is a significant challenge in developing autonomous systems. Current machine learning methods often rely on large amounts of data and may not generalize well to new scenarios or environments. The UCF team’s approach aims to bridge this gap by designing AI technologies that can handle uncertainties and surprises more effectively.

By developing distributionally robust approaches to transfer learning, the researchers hope to enable autonomous systems to learn from limited real-world data and adapt to new situations more quickly. This could have significant implications for various industries, including healthcare, where treatment plans could be transferred between patients more effectively, and autonomous driving, where decision-making policies tailored for specific road conditions could be repurposed for other environments.

Overcoming Limitations of Traditional Machine Learning Methods

Traditional machine learning methods often rely on large amounts of data and may not generalize well to new scenarios or environments. The UCF team’s approach aims to overcome these limitations by developing AI technologies that can handle uncertainties and surprises more effectively.

By addressing the limitations of traditional machine learning methods, the researchers hope to enable autonomous systems to learn from limited real-world data and adapt to new situations more quickly. This could have significant implications for various industries, including healthcare and autonomous driving.

Potential Applications Across Industries

The AI technologies developed by the UCF team have the potential to revolutionize various industries. In healthcare, for instance, robust knowledge transfer methods could facilitate the transfer of treatment plans between patients, improving personalized care. Similarly, decision-making policies tailored for specific road conditions could be repurposed for other environments, enhancing safety and efficiency in autonomous driving.

The technology developed by the UCF team could also have implications for defense agencies, which are interested in developing autonomous systems that can handle unexpected events or changes in environmental conditions. By addressing the limitations of traditional machine learning methods, the researchers hope to enable autonomous systems to learn from limited real-world data and adapt to new situations more quickly.

Future Directions

The development of AI technologies that can help autonomous systems better adapt to unknown variables is a critical step towards advancing autonomous systems. The UCF team’s approach has the potential to revolutionize various industries, including healthcare and autonomous driving.

Future research directions could include exploring the application of these AI technologies in other domains, such as robotics or aerospace engineering. Additionally, further development of distributionally robust approaches to transfer learning could enable autonomous systems to learn from even more limited real-world data, making them even more effective in real-world scenarios.

More information
External Link: Click Here For More
Quantum News

Quantum News

As the Official Quantum Dog (or hound) by role is to dig out the latest nuggets of quantum goodness. There is so much happening right now in the field of technology, whether AI or the march of robots. But Quantum occupies a special space. Quite literally a special space. A Hilbert space infact, haha! Here I try to provide some of the news that might be considered breaking news in the Quantum Computing space.

Latest Posts by Quantum News:

IBM Remembers Lou Gerstner, CEO Who Reshaped Company in the 1990s

IBM Remembers Lou Gerstner, CEO Who Reshaped Company in the 1990s

December 29, 2025
Optical Tweezers Scale to 6,100 Qubits with 99.99% Imaging Survival

Optical Tweezers Scale to 6,100 Qubits with 99.99% Imaging Survival

December 28, 2025
Rosatom & Moscow State University Develop 72-Qubit Quantum Computer Prototype

Rosatom & Moscow State University Develop 72-Qubit Quantum Computer Prototype

December 27, 2025