Xiao Qin's Research
Projects
Current Projects
-
AI4Trees prepares AI-literate middle and high school students to apply AI and environmental data to forestry and natural resource challenges. The project develops AI-enabled forestry learning modules, provides professional development for teachers and Extension educators, engages youth through 4-H programs, summer camps, and design challenges, and studies the impact on students' skills and career awareness. (Sponsor: USDA NIFA; Role: Co-PI and AI Lead)
-
In partnership with Talladega College, we are developing six modular data science training units, covering programming, data cleaning, statistical analysis, feature engineering, machine learning, and data visualization, for undergraduate STEM courses at Talladega College, together with faculty workshops and cross-institutional mentoring that prepare students for graduate STEM programs at Auburn University. (Sponsor: NSF HBCU-UP; Role: PI)
The AI Bridge: From AI Literacy to AI Fluency for All Auburn Students (2026 - 2028)
The AI Bridge creates a pathway from AI literacy to proficiency and fluency for all Auburn students. Students start with COMP 1100 Power of AI (no math or code required), cross a three-course Bridge, and continue to the Certificate in AI Engineering or the MS in AI Engineering. (Sponsor: Center for Inclusive Computing at Northeastern University; Role: Co-PI and GTA Training Lead)
-
Partnering with Wright Media, an Alabama digital marketing company, we are building a data mining infrastructure that collects, processes, and analyzes large-scale user data, and developing algorithms that raise click-through rates and reduce cost per click in digital advertising campaigns. The project also trains software and data engineers at Wright Media. (Sponsor: the Alabama Department of Economic and Community Affairs (ADECA) through theAlabama Research and Development Enhancement Fund (ARDEF) program; Role: PI)
AEDIS: Auburn Educational Data Intelligence System (2025 - 2026)
AEDIS supports academic program assessment at Auburn University. The system integrates course, curriculum-map, and learning-management data, produces enhanced class lists with degree-program information, and presents assessment results through interactive dashboards. (Sponsor: Auburn University Mission Enhancement Fund; Role: Co-PI and Software Development Lead)
-
This NSF S-STEM project, led by Tuskegee University with a consortium of six institutions in Alabama and Michigan, supports scholarships for low-income, academically talented STEM students, combined with mentoring, research experiences, graduate school preparation, and hands-on experience with cutting-edge technologies such as digital twins. Supported by NSF (2024 - 2029). (Role: Co-PI and Software Engineering Lead)
Completed Projects
Flight Software Development for the Alabama CubeSat Initiative (2021 - 2025)
Student teams in our capstone design course designed and tested flight software for a NASA CubeSat intended to study the radiation environment around the Moon and make space travel safer for humans. The software runs on a real-time operating system and is built with F Prime, an open-source framework for flight software and embedded systems. (Sponsors: Alabama Space Grant Consortium and NASA EPSCoR; Role: PI)
AI for Online Recruiting and Advertising (2023 - 2025)
In two industry-sponsored projects with Wright Media, an Alabama recruiting solutions company, we developed AI-driven optimization techniques for web and mobile recruiting solutions (2023 - 2024) and real-time bidding techniques for online display advertising (2024 - 2025). (Sponsor: Wright Media Corp.; Role: PI)
Synthesis of Artificial Intelligence Training and Validation Method (2023 - 2024)
As part of this National Highway Traffic Safety Administration project led by Virginia Tech, our team contributed to a synthesis of methods for training and validating artificial intelligence systems. Dr. Qin also presented a talk on zero-shot learning at NHTSA in March 2024. (Sponsor: National Highway Traffic Safety Administration; Role: PI for Auburn University)
-
This Auburn University project built interdisciplinary partnerships between the humanities and computing to apply 21st-century technologies to the study of 19th-century literature, laying the groundwork for a digital humanities initiative at Auburn. (Sponsor: Auburn University Creative Work and Social Impact Scholarship Funding Program; PI: Beverley Rilett; Role: Co-PI)
Automotive-Relevant Project Experiences for CSSE Undergraduates (2020 - 2022)
With support from the DENSO Foundation, this project enhanced computer science and software engineering undergraduate education with automotive-relevant project experiences in mobility critical skills areas, giving students hands-on experience with problems drawn from the automotive industry. (Sponsor: DENSO Foundation; PI: Hari Narayanan; Role: Co-PI)
-
This project developed techniques for answering location-based queries and tracking moving objects in indoor spaces such as office buildings, shopping malls, and subway systems, where GPS signals are unavailable. Using RFID-based positioning and Bayesian filtering, the techniques account for indoor structures such as doors, walls, and hallways that constrain how people move. (Sponsor: NSF IIS-1618669; Role: PI)
-
This project created data integrity assurance and privacy protection solutions that help scientific researchers use cloud services safely. It developed techniques to verify that query results returned by cloud databases are accurate and complete, even when a cloud database is compromised or malfunctions, while protecting sensitive data from cloud service providers. (Sponsor: NSF CICI-1642133; Role: PI)
Multicore-Based Disks for Data-Intensive Computing (2009 - 2016)
Supported by an NSF CAREER Award, we bridged the technology gap between multicore computing and parallel disk systems by offloading data processing to multicore processors embedded in disk drives.
FastStor: Data-Mining-Based Prefetching (2009 - 2014)
We built data-mining-based multilayer prefetching techniques to improve the performance of data centers with hybrid storage systems.
A Middleware Approach to Teaching Computer Security (2009 - 2013)
This project developed QoSec, an extensible middleware framework for computer security course projects. The QoSec framework provides learning materials for undergraduate students related to the design of large-scale trustworthy computer systems.
MINT: Mathematical Reliability Models for Energy-Efficient Parallel Disk Systems (2007-2011)
The objective of this project is to address the mathematical underpinnings of modeling reliability of energy-efficient parallel disk systems, where fault tolerance and energy-saving techniques are seamlessly integrated to conserve energy without sacrificing reliability in parallel disk systems.
BUD: A Buffer-Disk Architecture for Energy Conservation in Parallel Disk Systems (2007-2010)
We are taking an organized approach to implementing energy-saving techniques for parallel disks, simulating energy-efficient parallel disk systems, and conducting a physical demonstration.
Energy-Efficient Computing Systems (2005-2008)
We designed job and packets scheduling algorithms for embedded systems, cluster computing platforms, and wireless networks. The algorithms can achieve good tradeoffs between energy efficiency and high performance.
Security-Aware Real-Time Systems (2004-2007)
This project addressed the issue of maximizing quality of security for real-time systems. We developed new dynamic/static scheduling algorithms and storage resource management to support security-aware real-time systems.
I/O-Aware Load Balancing (2002-2006)
To alleviate I/O bottlenecks in high-performance clusters, we develop load-balancing policies that are capable of achieving high disk utilization in addition to those of CPU and memory resources.
Real-time Scheduling in Heterogeneous Systems (1998-2001)
This research includes not only developing real-time and fault-tolerant scheduling algorithms but also exploring various options for designing real-time scheduling algorithms that strive to improve the reliability of heterogeneous systems.
Fault-Tolerant Real-Time Collaborative Editing Systems (1999-2000)
If a real-time collaborative editor is to be effectively used over the Internet, the system should tolerant the client and link failures, for the quality of the Internet are unpredictable. In this research we have developed a novel and efficient approach to supporting crash recovery in real-time collaborative editing systems.
Real-time and Fault-tolerant Distributed Systems (1996-1999)
If a real-time collaborative editor is to be effectively used over the Internet, the system should tolerant the client and link failures, for the quality of the Internet are unpredictable. In this research we have developed a novel and efficient approach to supporting crash recovery in real-time collaborative editing systems.