PACMAN AI control framework tested on DIII-D fusion system

Researchers at Princeton Plasma Physics Laboratory and Princeton University tested PACMAN, an AI control framework, in five experiments on the DIII-D tokamak. The framework combines real-time measurements, machine-learning predictions and controllers while enforcing hardware safety limits. In one experiment, a model predicted a tearing-mode instability about 200 milliseconds before it occurred; the system also controlled heating and plasma targets. Researchers say it typically runs in about 20 milliseconds, while physicists retain control over experiment parameters.