Thai Duong

I am a Ph.D. candidate in the Department of Electrical and Computer Engineering at University of California, San Diego. I work at the Existential Robotics Laboratory and am fortunate to be advised by Prof. Nikolay Atanasov. I am grateful to have collaborated with Prof. Michael Yip, Prof. Melvin Leok, Prof. Eduardo Montijano and Prof. Quan Nguyen.

Before moving to San Diego, I worked as a software engineer at Microsoft. I obtained my M.S. degree from Oregon State University, Corvallis, Oregon and B.S. degree from Hanoi University of Science and Technology, Hanoi, Vietnam.

Email  /  Google Scholar  /  Github  /  LinkedIn

profile photo
Recent News
Research Interests

I am interested in robotics, machine learning, control theory, and optimization. My work focuses on robots' understanding of the environments, e.g. probablistic mapping, navigation and exploration; and of their own dynamics model, e.g. robot dynamics learning, model-based reinforcement learning, and learning from demonstration. I am also interested in modeling uncertainty in map representations and robots' dynamics for safe and active planning and control.

Projects for robot dynamics learning and control

Trajectory tracking with learned model

Port-Hamiltonian Neural ODE Networks on Lie Groups For Robot Dynamics Learning and Control
Thai Duong, Abdullah Altawaitan, Jason Stanley, Nikolay Atanasov
Journal version, submitted to T-RO , 2023.
website / video / arxiv / code
Conference version accepted to Robotics: Science and Systems (RSS), 2021.
website / video / arXiv / code

The dynamics of many robots are described in terms of their generalized coordinates on a matrix Lie group, e.g. on SE(3) for ground, aerial, and underwater vehicles, and generalized velocity, and satisfy conservation of energy principles. This paper proposes a (port-)Hamiltonian formulation over a Lie group of the structure of a neural ordinary differential equation (ODE) network to approximate the robot dynamics. In contrast to a black-box ODE network, our formulation guarantees energy conservation principle and Lie group's constraints by construction and explicitly accounts for energy-dissipation effect such as friction and drag forces in the dynamics model. We develop energy shaping and damping injection control for the learned, potentially under-actuated Hamiltonian dynamics to enable a unified approach for stabilization and trajectory tracking with various robot platforms.

Data collection from manual flights

Hamiltonian Dynamics Learning from Point Cloud Observations for Nonholonomic Mobile Robot Control
Abdullah Altawaitan, Sambaran Ghosal, Jason Stanley, Thai Duong, Nikolay Atanasov
Accepted to ICRA, 2024.
website / arXiv / code

This paper develops an approach for learning robot dynamics directly from point-cloud observations, removing the need and associated errors of state estimation, while embedding Hamiltonian structure in the dynamics model to improve data efficiency. We design an observation-space loss that relates motion prediction from the dynamics model with motion prediction from point-cloud registration to train a Hamiltonian neural ordinary differential equation. We demonstrate dynamics learning and tracking control on a real nonholonomic wheeled robot

Physics-Informed Multi-Agent Reinforcement Learning for Distributed Multi-Robot Problems
Eduardo Sebastian, Thai Duong, Nikolay Atanasov, Eduardo Montijano , Carlos Sagues
Journal version, submitted to T-RO. .
website / arXiv / code

We propose a physics-informed reinforcement learning approach able to learn distributed multi-robot control policies that are both scalable and make use of all the available information to each robot. It imposes a port-Hamiltonian structure on the control policy, capturing the networked nature of robot team interactions and uses self-attention to handle time-varying information at each robot from the interaction graph. We present a soft actor-critic reinforcement learning algorithm to train our self-attention port-Hamiltonian control policy.

LEMURS: Learning Distributed Multi-Robot Interactions
Eduardo Sebastian, Thai Duong, Nikolay Atanasov, Eduardo Montijano , Carlos Sagues
Accepted to ICRA'23. .
website / arXiv / code

We propose a port-Hamiltonian description of the multi-robot system to exploit universal physical constraints in interconnected systems and achieve closed-loop stability. We represent a multi-robot control policy using an architecture that combines self-attention mechanisms and neural ordinary differential equations. The former handles time-varying communication in the robot team, while the latter respects the continuous-time robot dynamics. Our representation is distributed by construction, enabling the learned control policies to be deployed in robot teams of different sizes.

Adaptive Control of SE(3) Hamiltonian Dynamics with Learned Disturbance Features
Thai Duong, Nikolay Atanasov
Accepted to L-CSS, 2022.
website / video / arXiv

This paper develops geometric adaptive control with a learned disturbance model for rigid-body systems, such as ground, aerial, and underwater vehicles, that satisfy Hamilton's equations of motion over the SE(3) manifold. Our design consists of an offline disturbance model identification stage, using a Hamiltonian-based neural ordinary differential equation (ODE) network trained from state-control trajectory data, and an online adaptive control stage, estimating and compensating the disturbances based on geometric tracking errors. We demonstrate our adaptive geometric controller in trajectory tracking simulations of fully-actuated pendulum and under-actuated quadrotor systems.

Projects for environment understanding and navigation
Optimal Scene Graph Planning with Large Language Model Guidance
Zhirui Dai, Arash Asgharivaskasi, Thai Duong, Shusen Lin, Maria-Elizabeth Tzes, George Pappas, Nikolay Atanasov
Accepted to ICRA, 2024.

This work aims to leverage these new capabilities with an efficient task planning algorithm for hierarchical metric-semantic models. We consider a scene graph representation of the environment and utilize a large language model (LLM) to convert a natural language task into a linear temporal logic (LTL) automaton. Our main contribution is to enable optimal hierarchical LTL planning with LLM guidance over scene graphs. To achieve efficiency, we construct a hierarchical planning domain that captures the attributes and connectivity of the scene graph and the task automaton, and provide semantic guidance via an LLM heuristic function. To guarantee optimality, we design an LTL heuristic function that is provably consistent and supplements the potentially inadmissible LLM guidance in multi-heuristic planning.

Autonomous Navigation in Unknown Environments with Sparse Bayesian Kernel-based Occupancy Mapping
Thai Duong, Michael Yip, Nikolay Atanasov
Accepted to IEEE Transactions on Robotics (T-RO), 2022
website / video / arXiv / code

This paper focuses on online occupancy mapping and real-time collision checking onboard an autonomous robot navigating in a large unknown environment. We develop a probabilistic formulation based on Relevance Vector Machines, allowing probabilistic occupancy classification and supporting autonomous navigation. We provide an online training algorithm, updating the sparse Bayesian map incrementally from streaming range data, and an efficient collision-checking method for general curves, representing potential robot trajectories.

Autonomous Navigation in Unknown Environments using Sparse Kernel-based Occupancy Mapping
Thai Duong, Nikhil Das, Michael Yip, Nikolay Atanasov
International Conference on Robotics and Automation (ICRA), 2020
website / video / arXiv / code

We propose a new map representation, in which occupied and free space are separated by the decision boundary of a kernel perceptron classifier. We develop an online training algorithm that maintains a very sparse set of support vectors to represent obstacle boundaries in configuration space. We also derive conditions that allow complete (without sampling) collision-checking for piecewise-linear and piecewise-polynomial robot trajectories.

Projects for robot safety
Robust and Safe Autonomous Navigation for Systems with Learned SE(3) Hamiltonian Dynamics
Zhichao Li*, Thai Duong*, Nikolay Atanasov
Learning for Dynamics and Control (L4DC), 2022. *Equal contribution.
Journal version accepted to OJ-CSYS Special Issue: Intersection of Machine Learning with Control, 2022, as invited paper.
website / video / arXiv (L4DC) / arXiv (OJ-CSYS)

This paper proposes techniques to learn the dynamics models of a mobile robot from trajectory data and synthesize a tracking controller with safety and stability guarantees. We use a dataset of state-control trajectories to train a translation-equivariant nonlinear Hamiltonian model represented as a neural ordinary differential equation (ODE) network. The learned Hamiltonian model is used to synthesize an energy-shaping passivity-based controller and derive conditions which guarantee safe regulation to a desired reference pose. Finally, we enable adaptive tracking of a desired path, subject to safety constraints obtained from obstacle distance measurements.

Safe Robot Navigation in Cluttered Environments using Invariant Ellipsoids and a Reference Governor
Zhichao Li, Thai Duong, Nikolay Atanasov
ArXiv, 2020

We consider a control-affine nonlinear robot system subject to bounded input noise and rely on feedback linearization to determine ellipsoid output bounds on the closed-loop robot trajectory under stabilizing control. A virtual governor system is developed to adaptively track a desired navigation path, while relying on the robot trajectory bounds to slow down if safety is endangered and speed up otherwise. The main contribution is the derivation of theoretical guarantees for safe nonlinear system path-following control and its application to autonomous robot navigation in unknown environments.

Template borrowed from Jon Barron.