Franka Panda

What the Franka Panda can learn.

Each skill pairs neural pathways with target poses and a test the robot must pass with stimulation and fail without it.

1. 01Reach on command1 pathway · 6 s testPanda arm reaches forward and closes its gripper while the “reach” pathway is ON.
2. 02Sweep left / right2 pathways · 7 s testBase turns left or right depending on which pathway the neurons answer on.

What the neurons can move.

8 controls in all. A skill sets their targets; the simulator does the rest.

Arm joints

1 · 2 · 3 · 4 · 5 · 6 · 7

Gripper

open or close (0 = closed, 255 = open)

Real physics, on our servers.

The Franka Panda is simulated with MuJoCo 3.12, the official WebAssembly build, on the server; your browser loads the same model only to draw it. The model comes unmodified from Google DeepMind’s MuJoCo Menagerie.

Engine

MuJoCo 3.12, WebAssembly, server-side

Model

MuJoCo Menagerie · franka_emika_panda

In the simulator

7-joint arm with a parallel gripper (gripper: 0 = closed, 255 = open).