Our Research
Our work lies at the intersection of robotics, machine learning, and symbolic reasoning. We build robots that operate in human environments rather than around them, pairing learned policies with symbolic structure so that behavior generalizes from few demonstrations and stays inspectable enough to reason about safety and contact. We are always looking to recruit new lab members with interests in these topics.
Adaptation
We develop approaches that adapt to complex and dynamic environments, and to the people working alongside the robot.
Hybrid Models
We pair learned policies with symbolic structure so that behavior generalizes from few demonstrations and stays inspectable enough to reason about safety and contact.
Robotics
We build robots that work in human environments, covering manipulation in unstructured settings, human-robot interaction, and beyond.