Apply a control command, get the next observation
One call per control tick, the response IS the post-action observation. Commands take either form: a discrete action, or the SNN-native differential wheels (quantized onto the world’s drive: right wheel faster → veers left, near-equal → forward, near-zero → idle; the response echoes applied_action).
Authorizations
Single per-partner bearer key, provisioned by Alakazam.
Path Parameters
Body
Response
Observation after the action window.
The SNN-controller observation contract. proximity = two virtual range sensors (0 = clear, 1.0 = contact range): compact obstacles grade by screen-space closeness, walls/elongated structures by time-to-contact; an object dead ahead excites both. collision is a TERMINAL SPIKE , true exactly once, on the step the contact engine confirms the hit , after which done stays true until reset.
SPIKE, true exactly once per collision event (derived from the monotonic counter, so a pulse between two step reads is never missed). With terminal_collision=false the run continues and later events spike again.
Monotonic count of collision events this run.
Only latches when terminal_collision=true (or the world ends the run).
Milliseconds survived this run.
Lag accounting, how stale the sensor side (proximity/labels) is relative to the camera frame in the same observation. Cloud-detector worlds run ≈500–1000 ms behind; local-detector worlds tens of ms. Controllers should compensate or discount accordingly.
The discrete action actually applied (echoes wheel quantization). Present on step responses.
Present when the session requested a camera resolution.