I study how robots learn from humans. My work spans generalist robotics, neuro-symbolic policies, robust imitation learning, game-theoretic planning, and learned robot-to-human communication interfaces. I'm advised by Dylan Losey in Collab at Virginia Tech, where my dissertation is on safe, personalized robots across physical and semantic interactions. I'm also a DoD SMART Scholar and a Mechanical Engineer at the U.S. Army Engineer Research and Development Center (ERDC), Information Technology Laboratory, where I focus on autonomous, generalist robots.
News
- Oct 2026Presented our lab’s work on neuro-symbolic policies for manipulation at VRS 2026.
- Sep 2026Presented MEMO at IROS 2026.
- Sep 2026Passed my oral Candidacy Exams; I am now a PhD Candidate. See some of the slides from this exam here.
- Jun 2026MEMO has been accepted for publication in the Proceedings of IROS 2026.
- Jan 2026A Modular Haptic Display with Reconfigurable Signals for Personalized Information Transfer has been accepted for publication in IEEE ToH.
Older
- May 2025Presented Personalizing Interfaces to Humans with User-Friendly Priors at ICRA 2025.
- Jun 2024Accepted a second DoD SMART Scholar award, I will be continuing my research as a PhD student in the Collab!
- Apr 2024Defended my Master’s Thesis: Adaptive Communication Interfaces for Human-Robot Collaboration. You can read it here.
- Jan 2024Aligning Learning with Communication in Shared Autonomy has been accepted for publication in the Proceedings of IROS 2024.
- Mar 2023Towards Robots that Influence Humans over Long-Term Interaction has been accepted for publication in the Proceedings of ICRA 2023.
Research
Neuro-Symbolic Policies
Robots that use language models to generalize local corrections, utterances, or demonstrations to unseen tasks and environments.
Robust Interaction
Game-theoretic planning, robust imitation learning, and privacy-focused fine-tuning for securing user data during learning.
Personal Communication
Learned interfaces that discover robot-to-human communication mappings across modalities, with or without task knowledge.
Publications
2026
- IROS
- Preprint
Show and Tell: Unifying Physical Demonstrations and Language Corrections for Neuro-Symbolic Robots
- Preprint
- Preprint
- Preprint
2025
- Preprint
- ERDC
Online Manipulator Collision Avoidance Using SOTA Sensing Capabilities
- ERDC
Monte-Carlo Sphere Generation for Collision Detection in Manipulators
- ICRA
- ToH
2024
- THRI
- IROS
2023
- ICRA
Technical Reports
U.S. Army Engineer Research and Development Center
- 2023
- 2023
- 2023
- 2021
- 2021
- 2020
Software
- arxiv-tree — builds a citation tree from an arXiv paper, for getting up to speed in a new field.
- bchr.xyz — spatio-temporal encrypted messaging without an account. essentially a pastebin for messaging