Running a Policy on a Robot

RobotRuntime owns the control loop, hardware I/O, callbacks, and timing. It takes a required action_source: PolicySource wraps model + execution + action queue; TeleopSource forwards a leader arm. InferenceModel owns policy math. The physicalai run CLI instantiates the same objects from YAML. See the Runtime how-to.

Workflow

1. Choose API vs config: Python for notebooks/tests; YAML + physicalai run for reproducible deployment.

   • Done when: entry point matches the user's task.

2. Python minimal loop (see docs/how-to/runtime/run-policy-on-robot.md):

   from physicalai.runtime import RobotRuntime, PolicySource, SyncExecution
   from physicalai.inference import InferenceModel
   from physicalai_so101_plugin import SO101
   from physicalai.capture import UVCCamera