We received an NSF award from NSF Energy, Power, Control, and Learning (EPCL) for our project Collaborative Research: Intentional Stance Learning of Belief Dynamics, Objectives, and Control Policies in Autonomous Agents.
Learning · Control · Autonomy
Wenqian Xue
Postdoctoral Associate & Adjunct Professor
Nonlinear Controls and Robotics
Laboratory
University of Florida
My research focuses on safe, reliable, and interpretable learning and control for autonomous aerial, robotic, and multi-agent systems under uncertainty, including UAVs and robotic manipulator systems. My work pursues two complementary directions: AI for Control, which develops learning-based methods for adaptive and safe control and objective and intent inference; and Control for AI, which develops control-theoretic foundations for reliable generative and sequential learning, including safe trajectory and distribution generation. Specifically, my research lies at the intersection of automatic control, (inverse) reinforcement learning, deep learning, game theory, and generative modeling.
Currently, I am a Postdoctoral Associate at the Nonlinear Control and Robotics Laboratory (NCR), Department of Mechanical and Aerospace Engineering, University of Florida. I am also an Adjunct Professor in the Department of Electrical and Computer Engineering and I teach Linear Control Systems. Before joining the NCR Laboratory in 2024, I was a Postdoctoral Associate at the Computational NeuroEngineering Laboratory, Department of Electrical and Computer Engineering, University of Florida. I received my Ph.D. in Electrical Engineering in 2022 and my M.S.E. in Control Engineering in 2018 from Northeastern University, China. During 2019-2021, I was a Research Assistant in the Deparment of Electrical Engineering at UT Arlington.
Gainesville, Florida
w.xue@ufl.edu
News
I started to teach EEL 4657C Linear Control Systems (Fall 2026) in the Department of Electrical & Computer Engineering at the University of Florida.
An award was received from Sandia LDRD for the project: Optimal Trajectories via Generative Diffusion with Formal Guarantees for Drone Kinetic Intercept. I served in a leading technical role.
Our paper “Reachability-Aware Safe Control Under Noisy Measurements Using Game-Theoretic Reinforcement Learning” was accepted by CDC 2027.
Our paper “Energy Generative Modeling: A Lyapunov-Based Energy Matching Perspective” was posted on arXiv.
Our paper “Inverse Reinforcement Learning for Disturbed Networked Nonlinear Systems With Data Dropouts” was published in IEEE Transactions on Neural Networks and Learning Systems.
Our paper “Initially Excited Asynchronous Reinforcement Learning Control With Monotonicity and Stability” was published in IEEE Transactions on Systems, Man, and Cybernetics: Systems.
Research Directions
- Adaptive and Safe learning-Based Control: Data-driven reinforcement learning and deep learning methods for adaptive, safe, robust, interpretable control to enable autonomous systems to act reliably under uncertain dynamics and imperfect information, including measurement errors and limited observations.
- Objective and Intent Inference: Data-driven, model-free inverse reinforcement learning method to infer underlying objectives from demonstrations with certain system dynamics and imperfect observations to enable autonomous systems to understand observed behavior and perform trajectory imitation and guidance.
- Representation and Generative Learning: Control-theoretic representation learning and generative modeling methods with guaranteed convergence to enable autonomous systems to extract, recover, and safely generate informative representations from data, including distribution and trajectory generation.
- Implementations: My research has been applied to autonomous aerial and robotic systems, including UAVs/drones platforms, robotic manipulator systems, and multi-agent systems.
Publications
Teaching
My teaching integrates AI/machine learning and control theory with hands-on laboratories through MATLAB and Simulink.
EEL 4657C — Linear Control Systems
Classical and modern control systems, including system modeling, system representation in the complex frequency domain, stability and response analysis, and controller design in the frequency domain and time domain, with MATLAB and Simulink laboratory exercises.
Role: Instructor/Adjunct Professor