Autonomous aircraft for GPS-denied environments. Sensing, navigation, and mission autonomy — built in Scotland, UK.
SIDRA builds autonomous systems for public agencies, researchers, and industry — engineered for GPS-denied environments and held to aviation-grade standards. The platform that watches a forest inspects a turbine: one autonomy stack, many mission worlds.
We operate at the threshold — where an aircraft flies on its own estimate through denied navigation, geolocates what matters, and returns. The work begins where certainty ends.
Autonomy core for uncrewed platforms. Edge perception, on-board reasoning, and resilient flight across degraded and denied links — the guardian process every airframe runs.
Command and control layer. Fleet tasking, telemetry aggregation, and human-on-the-loop authority from a single ground station — one operator, many aircraft.
Clean-propulsion architecture for aerospace and maritime. Hybrid-electric and zero-emission drive engineered for endurance over noise.
Sensing layer. Multi-spectral Earth observation built around SPAD arrays with in-house calibration pipelines, tuned for low-light edge inference — photon-starved scenes read on board, at dusk, in smoke.
One autonomy core — CUSTOS — flies every airframe in the fleet, and every mission gets an airframe tuned to its envelope. A buildable quadcopter today, a hybrid-VTOL in development, a long-endurance concept next. Same brain, widening reach.
Buildable today — ~£610 bill of materials, 40-minute hover, flying the CUSTOS autonomy stack on commercial hardware.
Bullet fuselage, blended wing, four lift rotors and a pusher — vertical launch anywhere, fixed-wing range once airborne. Carries the ASTRUM SPAD low-light array for after-dark sector sweeps.
The watchtower of the fleet — extended-range sensor platform for wide-area environmental survey, carbon verification, and flood-corridor mapping.
Low-noise rotors and high-AGL optics for standoff orbits animals never notice.
Precision station-keeping against turbine blades and solar rows — macro imaging in ground-effect turbulence.
Salt-spray hardened BVLOS transit and structured survey where crewed flights stay grounded.
Low-tide repeat survey over mangrove restoration plots — multispectral scoring that steers the next seeding season.
The exact airframe from the engineering CAD — wheelbase to fastener. Drag to orbit.
Every mission gets the aircraft it deserves.
One autonomy core — CUSTOS — flies the whole fleet, but each mission is served by a bespoke airframe tuned to its envelope. The orbit that refines a wildfire fix inspects a turbine blade; the platform differs, the discipline does not.
Dawn and dusk patrol lines over high-risk forestry, reading the ground in thermal. A smouldering ignition is geolocated and reported minutes after it starts — while it is still one crew, not one thousand hectares. 2025 gave the UK its largest recorded wildfire; 2026 gave Scotland a national action plan.
A thermal sweep of a corrie takes minutes, not a rope team’s afternoon — and the sorties that matter most start after dark. Vertical launch from the rescue base, fixed-wing transit to the sector, and a navigation stack that holds its own estimate in the glens where GNSS gets marginal.
The watchtower of the fleet. Long-endurance repeat lines over restored blanket bog measure re-wetting and vegetation recovery for carbon verification; when the water rises, the same aircraft flies BVLOS river corridors and maps flood extents while responders still have hours to act.
High, quiet standoff orbits count herds, log movement corridors, and flag vehicles that should not be there — without a rotor note the animals ever notice.
The same orbit logic that refines a track flies a blade-following scan: leading-edge erosion and lightning strikes geotagged to the exact blade station. Over solar arrays, thermal passes find hot cells and failed strings before they show in the yield curve — no rope team, no wasted shutdown window.
BVLOS transit to platforms and substations, structured survey in salt spray that grounds crewed flights — and a navigation stack that holds on the estimate when GNSS gets ugly.
The Gulf is planting mangroves by the hundred million — Abu Dhabi alone targets 100M by 2030, seeded by drone. Someone has to measure what survives. Repeat low-tide surveys score germination and survival, feeding the next seeding pass.
July 2026: fires across Spain, Portugal, France, Greece, Italy and Türkiye, driven by the third heatwave in six weeks — and, on our own doorstep, Scotland’s worst season on record: eleven thousand hectares gone at Carrbridge and Dava, the Cairngorms National Park evacuated as fire spread north through forestry, and Arthur’s Seat burning in the heart of Edinburgh. Detection still depends on watchtowers, passers-by, and satellites that revisit hours apart. The gap between ignition and alert is where a one-crew fire becomes a thousand-hectare one — and it is exactly the gap NIGHTJAR patrols.
NIGHTJAR closes the gap: ignition to geolocated alert in under 90 seconds, on autonomous dawn-and-dusk patrol.
Watch it find the fire ↓Another record fire season is burning through headlines. The variable that decides whether an ignition becomes a paragraph or a catastrophe is time-to-detection — so here is the same fire, twice, side by side: once found by a passer-by, once found by a NIGHTJAR thermal patrol.
Accelerated model for illustration — cellular spread under a steady wind, identical seed and terrain both sides. Detection latency on the right is the figure validated in the engineering simulator (under 40 seconds from ignition to alert).
A concept study of what a forest fire-watch pass could look like — one hot spot blooming in a dark canopy, the early-warning picture CUSTOS is being built to read.
Navigation, autonomy, and flight test — the record as it is written.
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.
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.
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.
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.
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.
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.
Building SIDRA's NIGHTJAR airframe and CUSTOS autonomy stack, validated in 6-DOF simulation in-house.
Background in UAV performance engineering at Baykar Technologies, followed by managing the Baykar–SAMI–Saudi Ministry of Defense partnership for localized Bayraktar Akıncı and TB2 production in Saudi Arabia — Turkiye's largest defense export partnership.
Research at Edinburgh's APRIL AI Hub (STMicroelectronics × Google DeepMind × Royal Academy of Engineering), Bristol Drone Soc., UCL's Department of Electronic and Electrical Engineering, and Airbus UK.
Member, Royal Aeronautical Society & IMechE. Trilingual (Arabic, English, Turkish), Scout leader, building in the UK.
Outside SIDRA: nature enthusiast and parrot owner, drawn to wildlife and conservation — part of why environmental applications matter to what SIDRA builds. The instinct started with a free-flying sun conure on the Red Sea that would maneuver over the water and return to hand: loiter, maneuver, return, before any of it was code.
Recruiting technical leads across the capability pillars.
SIDRA is an early-stage venture out of Glasgow, taking one autonomy core — CUSTOS — from validated flight test toward operational deployment across wildfire early-warning, inspection, and environmental missions. We are in conversation with investors and partners who back frontier hardware and climate resilience.
