Researcher

Mitchell Black

Mitchell Black researches control under uncertainty and safe autonomy. His shared work includes multi-agent reinforcement learning with explicit safety constraints. Black completed a Michigan master’s in aerospace engineering from 2016 to 2017. He conducted doctoral research from 2020 to 2023 on adaptation and prediction for control barrier functions. Black co-authored the RSS 2025 paper on distributed epigraph-form multi-agent reinforcement learning with Songyuan Zhang, Oswin So, Zachary Serlin and Chuchu Fan. It studies constrained control with centralized training and distributed execution, with simulation and quadcopter experiments. Black works as technical staff at MIT Lincoln Laboratory and researches formal control methods and machine-learning-based risk quantification.

• safe autonomy
• multiagent reinforcement learning
• robot control