Fall 2026 GRASP on Robotics: Andrea Bajcsy, Carnegie Mellon University, “What Does Safety Mean for Generalist Robots?”

This will be an in-person event ONLY in Wu and Chen Auditorium.

ABSTRACT

Robot safety is a nuanced concept. We commonly equate safety with collision-avoidance. However, when considering the possibility of generalist robots — one robot that can do “anything” — we need to expand our notion of safety. For example, a manipulator should understand when it is not confident about a requested task, that objects should be gently manipulated to prevent breaking or spilling, and that overheating items can cause burning or melting. However, designing robots that have such a nuanced safety understanding (and that can reliably generate appropriate actions) is an outstanding challenge. In this talk, I will describe my group’s work on systematically uniting modern machine learning models (such as large vision-language models and latent world models) with classical formulations of safety in the control literature to generalize safe robot decision-making to increasingly open world interactions. Throughout the talk, I will present experimental instantiations of these ideas in domains like vision-based robotic manipulation.