Max Robotics — Global Service Robot Directory + US Certification

About

David Fridovich-Keil is a robotics and control researcher whose work sits at the intersection of multi-agent systems, game theory, and safe autonomous decision-making. His research addresses some of the most pressing challenges in deploying robots alongside humans — particularly how autonomous systems can predict, reason about, and respond to the behavior of other agents in real time. His most cited work, "Confidence-aware Motion Prediction for Real-Time Collision Avoidance" (2019, 115 citations), introduced a principled framework for incorporating prediction uncertainty into robot motion planning, enabling safer human-robot interaction. This theme of uncertainty-aware safety extends to his probabilistically safe planning work, which addresses the inherent limitations of human motion models. On the control side, Fridovich-Keil developed reinforcement learning-based approaches to feedback linearization for systems with unknown dynamics, bridging modern machine learning with classical nonlinear control theory. A significant thread of his research applies differential game theory to multi-robot and human-robot interaction, developing efficient algorithms for computing equilibria in complex multi-player scenarios and inferring the hidden objectives of other agents from observed behavior. His contingency games framework further advances this vision by enabling robots to plan flexibly under discrete uncertainty. Collectively, his contributions offer foundational tools for building robots that interact intelligently and safely in dynamic, human-populated environments.