Sparse success
A full pick-and-place reward arrives too late for early learning. Staged objectives provide useful intermediate signals.
The Motion Forge / Hybrid controller · study ongoing
A hybrid UR10e controller combining a learned PPO approach with deterministic grasping, Lula transport, and verified basket placement in Isaac Lab.
* 144 / 256 stable holds in the frozen baseline.
PROOF SIGNAL
144 / 256stable holds in the frozen baselineThe mission
A hybrid UR10e controller combining a learned PPO approach with deterministic grasping, Lula transport, and verified basket placement in Isaac Lab.
Engineering log
What the system handles, and why it works this way.
A full pick-and-place reward arrives too late for early learning. Staged objectives provide useful intermediate signals.
Contact-heavy manipulation is sensitive to setup. Collision handling and controlled curriculum changes keep training interpretable.
System map
A plain technical view. The map still works without hovering.
Target-conditioned positioning
Grasp, lift, hold, and retry
Clearance and basket alignment
Release and stable placement
LOG ENTRY / WHAT I LEARNED
“Reward design is an interface between an engineering goal and what an agent can actually discover.”
Source signal