From AlphaGo to AlphaZero, reinforcement learning produced strategies that surprised expert players. AlphaZero learned from experience at scale, without human demonstrations.
AlphaNIST brings this philosophy to contact-rich, high-precision robotic assembly in simulation. An end-to-end policy learns how to complete the task, without prescribed manipulation stages.
Four successful ten-part runs, using a policy trained separately from the single-component policy below.
The NIST board has twenty assembly tasks, with parts as small as 4 mm. Fitting one part can take several changes of grasp and orientation. Tight clearances leave little room for error: a tilted nut can jam or cross-thread.
