Physical AI · Robotics · Learning in public
I'm exploring the world of robotics and Physical AI. This is my field notebook documenting my experiments with robot learning, where I share lessons, discoveries, and challenges from my journey into robotics — from real-world robot learning to simulation and beyond. Stay tuned for more content coming soon!
My single-color pick-and-place policy generalized to new shapes and colors, but broke the moment I changed the tray's color. So I built a real sorting task: two trays, two colors, and a policy that has to actually look at the block before deciding where it goes.
My color-sorting ACT policy suddenly started missing the block, reaching to the same spot every time regardless of where the block actually was. Here's how I used the raw recorded training trajectories — replayed directly on the physical arm — to figure out it was a workspace calibration problem, not a model problem.
Four steps to a working pick-and-place policy: 3D printing my own tray and block, collecting a clean 40-episode dataset, training ACT, and then trying to break it to see what generalizes.