Antonia Bronars

I am a PhD student at MIT working with Professor Pulkit Agrawal in the Improbable AI lab. I am currently investigating how tactile perception can enable more robust and dexterous robotic manipulation. I have also spent time as a Student Researcher at Google Deepmind London, and as a Research Intern with Mitsubishi Electric Research Labs (MERL). I did my Masters with Professor Alberto Rodriguez in the MCube Lab.

Prior to MIT, I got my bachelor's in Mechanical Engineering at UC Berkeley. I won the Steidel Award for undergraduate research, working with Professor Oliver O'Reilly on nonlinear dynamics, and Professor Alice Agogino on tensegrity robots.

Email  /  CV  /  LinkedIn  /  Twitter  /  Google Scholar

I am interested in physical robot intelligence, and in particular, the role of tactile feedback in reactive control of dexterous manipulation. My recent work has centered methods for solving precise manipulation tasks that leverage tactile feedback.

We approach few-shot continual learning via generative modeling.

Our framework fuses several manipulation skills including grasping, in-hand manipulation, and insertion to assemble a gear box by leveraging visuotactile and force-torque data.

Our framework estimates and tracks object pose using tactile images while simultaneously generating motion plans to achieve manipulation objectives.

We assemble tight-tolerance industrial connectors using a only simple controller, without the need for specialized search or insertion algorithms. This is possible via high-accuracy tactile pose estimation.

simPLE learns to pick, regrasp, and place objects precisely given the object CAD model. We develop three main components: task-aware grasping, visuotactile perception, and regrasp planning.

Website source code