$$ % Define your custom commands here \newcommand{\bmat}[1]{\begin{bmatrix}#1\end{bmatrix}} \newcommand{\E}{\mathbb{E}} \newcommand{\P}{\mathbb{P}} \newcommand{\S}{\mathbb{S}} \newcommand{\R}{\mathbb{R}} \newcommand{\S}{\mathbb{S}} \newcommand{\norm}[2]{\|{#1}\|_{{}_{#2}}} \newcommand{\pd}[2]{\frac{\partial #1}{\partial #2}} \newcommand{\pdd}[2]{\frac{\partial^2 #1}{\partial #2^2}} \newcommand{\vectornorm}[1]{\left|\left|#1\right|\right|} \newcommand{\abs}[1]{\left|{#1}\right|} \newcommand{\mbf}[1]{\mathbf{#1}} \newcommand{\mc}[1]{\mathcal{#1}} \newcommand{\bm}[1]{\boldsymbol{#1}} \newcommand{\nicefrac}[2]{{}^{#1}\!/_{\!#2}} \newcommand{\argmin}{\operatorname*{arg\,min}} \newcommand{\argmax}{\operatorname*{arg\,max}} \newcommand{\dd}{\operatorname{d}\!} $$

Research

Robot Control Lab studies robotics, nonlinear control, and learning for dynamical systems at UT Dallas.

We study how robots can use their natural dynamics to move, manipulate objects, and interact with the world. Our work brings together nonlinear control, geometry, optimization, and machine learning.

A central question connects these areas: how can we build controllers that exploit physical structure while accounting for uncertainty? At Robot Control Lab, mathematical models provide a foundation for understanding motion, and learning offers tools for tackling control-design problems that are difficult to solve analytically.

Two connected research directions

Robotics and control

How can a robot use its dynamics to accomplish a task? We study locomotion, manipulation, and coordinated motion, with methods that account for energy, geometry, and actuation constraints.

Explore robotics and control →

Robust machine learning

How can learning support control when models and measurements are uncertain? We study neural approaches to controller design and Bayesian methods for representing uncertainty.

Explore robust machine learning →

From physical models to feedback

Understand the dynamics

Describe the robot’s motion, constraints, and energy to identify the structure a controller can use.

Design the controller

Combine nonlinear control and optimization with data-driven methods where analytical design becomes difficult.

Examine the behavior

Study stability and robustness under stated assumptions, alongside the behavior of the resulting robotic system.

Starting points in our work

  • Control through energy shaping. Neural approximations help address the equations that arise in passivity-based controller design. Read the study.
  • Dynamic manipulation. A coordinate-free framework studies robotic pizza tossing and catching. Publication and Scholar search.
  • Learning with uncertainty. A Bayesian control study examines parameter and state uncertainty in a simple stochastic system. Read the preprint.