Hello! I’m a robotics researcher at Toyota Research Institute (TRI) with interests spanning machine learning, robot control, and extended reality interface design. At the intersection of these interests is my overarching goal of establishing effective human-robot interactions where people live and work.

My recent work has primarily focused on improving robot foundation models, such as vision-language-action (VLA) models. Through my involvement in the Large Behavior Models project and TRI’s collaboration with Boston Dynamics to deploy these models on the eAtlas humanoid, I have had significant ownership across the entire stack, including data collection and curation, policy training and evaluation, and deployment in both simulation and on hardware. I am especially interested in methods for enabling these models to collaborate with people in a personalized and customizable manner.

Prior to joining TRI, I was a research scientist in robotics at Northeastern University, where I served as a research lead on NSF and DARPA projects. I have a PhD in Computer Engineering from Imperial College London, with work on augmented reality interfaces and shared control paradigms for robot teleoperation, representation learning algorithms for human intention estimation, and user studies involving robotic wheelchairs.

Email: mark[dot]zolotas[at]tri[dot]global