The record as it is written.
Navigation, autonomy, and flight test. No press releases — engineering entries.
2026.08 · VALIDATION
A safety filter added in July — the velocity D-term low-pass that contains estimator-step thrust spikes — quietly cost settled waypoint precision, 0.05 m to 0.08 m, and the 1.5 m sim gate never noticed. A pinned-environment bisect found it; halving the filter time constant kept ~13x spike attenuation and returned 0.03–0.07 m. Every validation gate across the three sims is now calibrated to the measured seed-sweep envelope at roughly 3x the worst observed run — margins that mean something, recorded next to the assertions they bound.
2026.07 · MISSION
Three ignitions lit mid-mission during a repeating patrol of the survey area. All three were autonomously detected, confirmed, and orbit-refined — latencies of 0.4, 24, and 40 s from ignition, geolocated to within 0.14 m. Full stack, controller on the estimate throughout.
2026.06 · NAV
NIGHTJAR held 0.78 m peak error on optical flow, baro, and magnetometer alone through a jammed mid-mission window. Innovation gating rejected the spoof-style jumps outright.
2026.05 · AUTONOMY
CUSTOS confirmed three ground contacts and autonomously flew a sensor-on orbit around each to tighten geolocation below 0.31 m CEP, then resumed the search.
2026.04 · FLIGHT TEST
Optical flow drops out below 0.3 m AGL, so the land detector now keys on altitude alone — ending the contact-bounce seen in earlier sorties.
ORIGIN · HERITAGE
Before any of this was an autonomy stack, there was a sun conure on the Red Sea shore at Suez. It would launch, work its own arcs over the water, and come back to hand: loiter, maneuver, return, flown on instinct. SIDRA is that instinct, engineered. Callsign: ARATINGA.