Preston Culbertson
Preston Culbertson is a roboticist who draws on machine learning, computer vision and control theory to develop robots that move like humans. He studied mechanical engineering at Georgia Tech and Stanford, finishing a PhD in 2022 while working with Mac Schwager in the Multi-Robot Systems Lab. In 2018, he and Schwager won the IEEE ICRA Best Paper Award in Robot Manipulation. He was a postdoctoral scholar at Caltech and joined The AI Institute as a research scientist in June 2024; his DROP paper on in-hand cube reorientation won Outstanding Paper at the CoRL Workshop on Dexterous Manipulation in November 2024. Since 2025, he is an assistant professor of computer science at Cornell, where he leads the Praxis Lab on robust robot learning.
- Now at
- Cornell University
Timeline
-
2016–2022
PhD at Stanford with Schwager
He earns a BS in mechanical engineering from Georgia Tech in 2016 and an MS in mechanical engineering from Stanford in 2020. At Stanford, he works with Prof. Mac Schwager in the Multi-Robot Systems Lab on multi-robot manipulation, distributed and robust adaptive control, and learning for decision-making and control. He completes his PhD in mechanical engineering in 2022.
-
2018
ICRA award and NASA fellowship
He and Mac Schwager win the 2018 IEEE ICRA Best Paper Award in Robot Manipulation for Decentralized Adaptive Control for Collaborative Manipulation, which shows that a team of robots can move an unknown payload along a desired trajectory by adapting their control gains. A 2018 NASA Space Technology Research Fellowship funds his work on algorithms for multi-robot assembly of orbital structures using teams of small satellites.
-
2021
Navigating inside a NeRF
In 2021, he coauthors Vision-Only Robot Navigation in a Neural Radiance World with Michal Adamkiewicz, Timothy Chen, Adam Caccavale, Rachel Gardner, Jeannette Bohg and Mac Schwager. The algorithm navigates a robot through a 3D environment represented as a neural radiance field, using only an on-board RGB camera for localization.
-
2023
FRoGGeR grasp synthesis
In 2023, he coauthors FRoGGeR: Fast Robust Grasp Generation via the Min-Weight Metric with Albert H. Li, Joel W. Burdick and Aaron D. Ames. The differentiable, analytic grasp metric enables collision-free grasp synthesis for dexterous hands in seconds.
-
June 2024
Research scientist at The AI Institute
In June 2024, he joins The AI Institute in Cambridge, Massachusetts, as a research scientist. He works on contact-rich planning and 3D vision for dexterous mobile manipulation.
-
2024
DROP for in-hand reorientation
In 2024, he coauthors DROP: Dexterous Reorientation via Online Planning with Albert H. Li, Vince Kurtz and Aaron D. Ames. A sampling-based predictive controller and a vision-based pose estimator search for contact-rich control actions online and reach performance comparable to prior reinforcement-learning work on in-hand cube reorientation. DROP wins Outstanding Paper at the CoRL Workshop on Dexterous Manipulation in November 2024.
-
Since 2025
Assistant professor at Cornell
He joins Cornell as an assistant professor of computer science in 2025, after working as a research scientist at The AI Institute and as a postdoctoral scholar at Caltech. He draws on machine learning, computer vision and control theory to develop robots that move like humans, and leads the Praxis Lab, whose research on robustness in robot learning focuses on dexterous manipulation and tool use. In January 2026, he gives a plenary talk at MIT, Robust Robot Behavior Through Online Optimization.
Career
- Since 2025Cornell UniversityAssistant professor of computer science
- From 2024The AI InstituteResearch scientist
- California Institute of TechnologyPostdoctoral scholar
Education
- 2022Stanford UniversityPhD in mechanical engineering
- 2020Stanford UniversityMS in mechanical engineering
- 2016Georgia Institute of TechnologyBS in mechanical engineering
Sources and credits
Photo: We have no published photo of Preston Culbertson yet. Send us a free-licensed photo.
- Preston Culbertson
- News
- Preston Culbertson
- Vision-Only Robot Navigation in a Neural Radiance World
- Preston Culbertson | Department of Computer Science | Cornell Bowers
- DROP: Dexterous Reorientation via Online Planning
- Meet the new faculty: Preston Culbertson | Cornell Bowers
- Plenary Talk | Prof. Preston Culbertson - Events Calendar
- Preston Culbertson – Robotics @ Cornell
