Default humanoid intake to command-following gait and support bounded Cartesian target dragging for arms. Carry model and actuator semantics through calibration, native verification and packaging. Document reusable intake decisions and cover them with regression tests.
3.4 KiB
Operator controls
controller/vocabulary.json defines the names used by the gamepad and
keyboard readers. Print them with python -m controller --vocabulary.
controller.controls.load_controls validates the shared schema_version: 2
format.
Training template controls
Generate a body-frame locomotion binding beside the robot asset:
python .claude/skills/controls/scripts/write_controls.py --task my_robot \
--out assets/my_robot/controls.yaml
The generated command is twist feeding velocity_commands. Its axes are
lin_vel_x, lin_vel_y, and ang_vel_z, in that order. Pad deflections
scale to the command range, with separate positive and negative ends.
Keyboard keys bind directly to an axis direction and ramp while held.
For a registered task the generator selects a layout by family. Bipeds and
as-yet unregistered robots default to walking; other legged families use
omni, except wheeled bipeds which use differential. An explicit
--profile overrides that choice.
The walking layout uses W/S for forward/reverse and A/D for
left/right yaw. The arrow keys mirror WASD. J/L request left/right lateral
motion on policies that support it. The pad uses Ly for forward speed, Lx
for lateral speed and Rx for yaw. Holding W+A combines forward and left turn.
The explicit omni layout keeps A/D for lateral motion and J/L for yaw.
Use --profile differential for a wheeled biped; its lateral axis is
unbound on the keyboard and has a zero command range in the template.
Use --bind AXIS=SOURCE+ or --bind AXIS=SOURCE- for explicit signs and
--release BUTTON for the pad release. Review the direction of motion and
release behavior in simulation. Point the robot specification's
controls_path to the generated file.
Deterministic controller controls
onboard.py setup-tier2 generates controls for the controller's named axes.
The config's optional operator_axes selects up to four axes. Remaining
joint sliders or targets can be addressed through the programmatic controller.
Run native verification with the exact controller and controls paths. After
changing bindings, generate a fresh report before packaging.
For end-effector IK, double-click the tool target marker, then hold Ctrl and drag with the right mouse button. Add Shift to change the translation plane. Release the button to hold the selected target.
Shared schema
The schema describes a list of command terms and the bindings for both devices. Each command declares its name, observation or controller feed, frame, ordered axes, and the full-deflection scaling rule. Each axis names its units and positive direction. The gamepad declares a signed source, and the keyboard declares key lists for both directions or an explicit reason that the axis is unbound.
The parser also supports split travel, shifted bindings, key chords,
integrated axes with reset, and auxiliary controls for compatible
consumers. A consumer can restrict that schema to the features it
implements. The locked locomotion templates take one body-frame velocity
command with zero rests and no auxiliary controls. Native mode selection
uses --app-mode. Legacy FSM mode switches are separate from a policy's
own device bindings and are declared in the legacy deployment manifest.
python -c "from pathlib import Path; from controller.controls import load_controls; load_controls(Path('assets/my_robot/controls.yaml'))"
python -m pytest tests/test_controls.py -q