Skydio X2

What the Skydio X2 can learn.

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

1. 01Take off1 pathway · 7 s testThe quadrotor hovers low and climbs to 1.2 m only while the “take off” pathway is ON.
2. 02Fly left / right2 pathways · 8 s testTwo pathways: the drone slides 0.6 m to its left or right, whichever pathway is ON.

What the neurons can move.

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

Flight set-points

altitude above the floor (0.15 to 2.5 m) · forward from the start (−2 to 2 m) · lateral from the start, left positive (−2 to 2 m) · yaw (−π to π rad)

Real physics, on our servers.

The Skydio X2 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 · skydio_x2

In the simulator

Quadrotor flown by an onboard cascaded PD controller. Controls are setpoints: altitude (m above the floor), forward / lateral (m from the start position, +lateral = left), yaw (rad).