We are a research group at Northeastern University, within the Department of Electrical & Computer Engineering, the Khoury College of Computer Sciences, and the Institute for Experiential Robotics.
Our research lies at the intersection of robotics, deep learning, and control theory, with the goal of developing certifiable learning machines. This means creating robots that discover how to do cool behaviors – with guarantees on safety, reliability, and efficiency. Some specific techniques of interest are: reinforcement learning (RL), reachability analysis, learning cost-to-go functions, bridging semantic perception and motion planning, and model predictive control (MPC). A key application area of interest is navigation in challenging environments, such as off-road (e.g., forests, deserts) and alongside humans (e.g., on busy sidewalks, in crowded buildings).
