Why We Built Drone One

SAR is software. We're a software company. So why did we just build a drone?

The short version: because the customers we want to serve first — regional fire brigades, inland public safety units, training departments, search-and-rescue charities running a first pilot — can't buy the drone we used to recommend. Not because they don't want to. Because €25,000 per drone is outside the budget they have for a first-pilot evaluation, and the €4–8k drones they can afford run a closed flight stack that locks our software out.

So we built one in the gap.

THE FOUNDING STORY · PART 3 OF 3

Three-part company narrative. Part 1: Our First Fleet — how three Holybro X500 V2 kits in a Belgium workshop turned into a software-and-hardware company. Part 2: Why we built SAR — what was missing in existing ground stations, and why we built the whole stack. Part 3 (this post): Why We Built Drone One — why the software company added hardware.

The price-shape of the actual market

Look at the SAR drone market through the lens of a regional fire brigade or a small coastguard unit, not a US federal agency, and the pricing tiers that actually exist look like this:

Sub-€10,000: DJI Mavic 3T. Closed flight stack. €4–8k retail depending on configuration.
€10–15,000: …essentially nothing with thermal and an open MAVLink stack.
€15–25,000: Various enterprise drones with mixed sovereignty stories and inconsistent SDK access.
€25,000+: Freefly Astro Max, Quantum Systems Trinity Pro — the NDAA / EU-sovereign reference platforms.

The market we want to sell to lives in the missing middle. Inland public safety. Belgian zones de secours and equivalent regional emergency services across Europe. Rural fire services. Search-and-rescue charities. Training departments at police academies and university aviation programmes. None of them have NDAA constraints; few have €25k-per-drone procurement headroom either. The ones who buy DJI today buy it because nothing else fits the budget — not because they prefer the locked stack.

If we keep recommending Astro Max as the entry option, we wait two years to ship a first pilot. If we keep saying "we're software-only, bring your own drone", customers in this segment bring DJI, and SAR can't do half its job on DJI's closed SDK — no companion-computer integration, no third-party autonomy injection, no fleet relay across drones from different vendors.

We needed a third option.

The DJI / locked-stack ceiling, and the geopolitical clock

It's worth being specific about why DJI doesn't work for the autonomy stack SAR needs to run, and why even where DJI is technically buyable today, the procurement risk on a 5–10 year horizon is real.

The technical ceiling

The Mavic 3T is a genuinely good aircraft for what it is. The thermal sensor is competitive, endurance is solid, the imaging stack is the most polished in the industry. For a single-pilot, single-drone, manually-flown mission — one operator, one drone, one search — it works. For everything beyond that, the closed Payload SDK and the progressively-restricted Onboard SDK make the kind of integration SAR needs unreachable. You can't run a third-party autonomy stack on the flight controller. You can't connect a companion computer to the autopilot for onboard detection. You can't get the kind of mid-flight parameter access that fleet-relay handoffs depend on. The customer who picks DJI gets DJI's autonomy ceiling, not ours — which is fine for the segments where DJI's built-in behaviour is enough, and not fine for any mission where you want fleet patrol, multi-drone relay, custom detection models, integration with a dispatch system, air-gapped operation, or the weather-aware re-planning SAR's safety architecture requires.

The geopolitical clock

DJI is a Chinese company, and the political environment around Chinese-origin drone hardware in Western public-safety procurement has shifted hard over the last three years and is still shifting:

US federal: the Countering CCP Drones Act and NDAA Section 848 effectively bar federal agencies from operating DJI hardware. Florida, Tennessee, Arkansas, Mississippi and a growing list of US states have followed with restrictions on state and municipal use. The trajectory is one-way.
UK: the CAA hasn't restricted DJI directly, but UK police, military, and increasingly civilian-emergency procurement prefer Blue UAS-listed platforms for interoperability with allied agencies.
EU: defence and civilian-emergency procurement has drifted toward sovereign supply chains since 2022 — partly because of the war in Ukraine, partly because of long-running data-sovereignty concerns. Several member states now write sovereignty preferences into civilian-SAR tenders.
Data sovereignty: DJI's default telemetry behaviour and firmware-update channels run through Chinese-jurisdiction infrastructure. Even when a specific drone is configurable to local-only operation, the procurement risk of a forced firmware update or a telemetry channel re-opening is non-zero.

A drone fleet you procure in 2026 will typically run for 5–8 years before replacement. The probability that the political constraints on Chinese-origin hardware tighten further over that window is high; the probability that they loosen is essentially zero. A locked DJI fleet bought today may not be operationally usable in three years if your jurisdiction joins the restriction list — and there is no software-portability escape: you can't move your investment in mission planning, training, and operator habits to a different airframe because DJI's stack doesn't speak open MAVLink. The hidden procurement cost of proprietary drone stacks is real, and DJI is the largest single instance of it.

Why Drone One + SAR is the long-term bet

Drone One ships open MAVLink end-to-end. SAR is software-portable across MAVLink airframes. If sovereignty constraints tighten in your jurisdiction, the same SAR ground station and the same operator training move with you to whichever Blue UAS or EU-sovereign airframe procurement specifies — Astro Max, Trinity Pro, or any other supported MAVLink platform. The bet you place on Drone One in 2026 doesn't lock you in; it gives you an autonomy stack that survives the airframe-replacement cycle, and a hardware vendor whose roadmap is moving in the same direction as the procurement environment.

What goes in the bundle

The hardware approach is to integrate well-known industrial-grade components, not to reinvent any of them. Every part is repairable in the field with parts available worldwide; everything speaks open MAVLink end-to-end.

Open-stack autopilot. PX4 / ArduPilot-compatible, full MAVLink mission protocol, mid-flight command access — the autopilot family SAR was designed against from day one.
Dual EO/IR gimbal. 4K visible plus 640×512 thermal in a single payload, included in the base price.
AI companion computer. On-airframe inference for human and vessel detection at 5–10 FPS. Pre-integrated, no customer-side wiring.
Long-range digital datalink. Combined C2 and HD video downlink in one package.
Ruggedised tablet ground station. Runs the full SAR Ground Station — mission planner, fleet view, alerts, telemetry, live video. The same software that runs against the example compatible airframes.

Detailed component spec sheets are shared with customers under evaluation. We focus the public conversation on what the bundle does, not on commodity component pricing.

The thermal gimbal being included in the base price is the line we hold. Most real SAR callouts happen at night, in fog, in smoke, or over dense canopy — conditions where visible-light detection fails. A SAR drone without thermal isn't really a SAR drone; it's a survey drone with optimistic marketing. We refuse to ship Drone One without thermal because the use case demands it. On the example compatible airframes, thermal is a separate payload addon at significant additional cost (~€7,000 on Astro Max, ~€13,000 on Trinity Pro). Including it in the base price on Drone One is what gets a real operational SAR aircraft — not a daytime-only one — below €15k.

On EU sovereignty: where we are, and where we're heading

Drone One isn't a fully EU-sovereign airframe today, and we won't pretend it is. Some components in the current bundle ship on global supply chains — the same chains every industrial-drone vendor at this price point depends on. We're a small drone-hardware-and-software company, and full EU sovereignty is our explicit aim and the direction we're building toward, not a feature we already shipped. It's why we exist as a hardware company at all rather than just selling SAR as software: a Belgium-based company with control over the integration and the roadmap is in a position to migrate the supply chain over time, vendor by vendor, in a way a software-only stack riding on third-party airframes is not.

What that means in practice in 2026:

The Drone One bundle is European-assembled in Belgium, with the autonomy stack, integration, flight-campaign verification, and aftermarket support all built in-house.
Some components (chassis, certain payload elements) come from global suppliers on the same supply-chain pattern as the wider industrial-drone industry — transparently sourced, named on customer spec sheets under evaluation NDA.
We are actively working toward a Belgium-or-European-origin chassis variant within the year, with a cleaner sovereign supply chain on the components customers care about most. That work is the company's medium-term roadmap, not aspirational marketing.

For customers in jurisdictions where sovereignty constraints are binding today — US federal SAR (NDAA / Blue UAS), or EU-sovereign procurement that explicitly excludes non-EU components — the example compatible airframes are the right answer right now: Astro Max for NDAA / Blue UAS, Trinity Pro for EU sovereignty. SAR runs identically on both, and the customer's procurement, not our preference, picks the airframe.

For customers whose constraints are practical — regional fire brigades, inland public safety, training, charities — Drone One's price point and delivery timeline are what unlock the evaluation. And the EU-sovereignty roadmap means Drone One stays a forward-compatible bet: SAR is fully software-portable across MAVLink airframes, so a customer who buys Drone One today and migrates to a future EU-sovereign chassis variant keeps the same autonomy stack, the same operator workflow, the same training. The investment in the autonomy layer survives the hardware migration.

What shipping our own bundle actually changes

Three things shift when we ship hardware alongside the software:

1. We control the integration loop. When a relay handoff misbehaves on Drone One, the fix lives in our repository. We don't file a ticket with the airframe vendor, wait for a firmware release, and re-test. Field iteration cadence on our own hardware is hours; on a third-party airframe it is weeks. For the parts of SAR that depend on tight ground-station / autopilot / companion-computer behaviour — fleet relay, GPS-denied navigation, exception-only alerting, on-airframe detection — owning the hardware is what makes the iteration honest.

2. First-pilot deployment timelines collapse. A typical Astro Max procurement runs 8–16 weeks from purchase order to operational. A typical Drone One bundle ships in 6–10 weeks because we hold inventory of the long-lead parts. For a customer running an evaluation pilot, that difference is whether the pilot finishes inside the same fiscal year or not. Public-safety budgets are annual; missing a fiscal year usually means missing the project.

3. The bundle pricing makes evaluation possible at all. A 1-drone Drone One evaluation at €11,000, or a 3-drone continuous-patrol setup at €30,000, fits inside a single line of a public-safety budget. Multiple regional fire brigades have envelopes in that range; very few have €75–150k headroom for a 3-drone Astro Max bundle. The segment opens up because the price point fits.

What stays the same: the SAR software is identical on Drone One, Astro Max, and Trinity Pro. Same boustrophedon search, same fleet relay, same SAR AI optimisation, same exception-only alerting, same multi-sector partitioning. MAVLink parity is real — we developed SAR on the X500 V2 chassis precisely so that what works in our workshop works on every supported airframe. Drone One isn't a different product; it's the same SAR product on the chassis we manufacture.

Where Drone One doesn't fit, and the rest of the supported list

Drone One is a multirotor at €11k with thermal included and ~18 minutes of flight time. That's the right answer for a regional fire brigade running urban-edge SAR, a charity covering a defined search area, or any first-pilot evaluation where the constraint is budget and time-to-first-flight. It is not the right answer for every mission profile.

Two examples where the customer needs a different airframe and we integrate SAR with the vendor's drone instead:

Long-endurance VTOL fixed-wing for large maritime / wilderness AOIs. A Trinity Pro on a 90-minute flight covers ground that would take an eight-multirotor relay to match. For coastguard agencies with hundreds of kilometres of coastline, the airframe choice is dictated by endurance, not price.
NDAA / Blue UAS-binding US federal procurement. Where the procurement floor is Blue UAS certification, Astro Max is the open-MAVLink answer with the highest customer trust today.

We are actively integrating SAR with these and other industrial-drone manufacturers, and our supported-airframes list grows whenever a customer mission requires it. The partnership posture is co-sell, not exclusive: we don't pick winners between airframes, the customer's procurement does. SAR is the autonomy layer; the airframe is whatever fits the mission, the compliance regime, the payload requirements, and the endurance class. We expect the supported list to keep growing — including with vendors not yet on it — because the point of building SAR on open MAVLink is precisely that the airframe market can change without breaking customer investment in operator training, mission planning, and detection workflows.

If you have an airframe in mind that we haven't listed, the answer is almost certainly "if it speaks PX4 or ArduPilot MAVLink and accepts a companion computer for video and detection, we can integrate". Ask — we want the conversation.

What we sell, and what we'll always sell

Drone One is the airframe we sell as a turnkey bundle today: airframe + thermal + SAR ground station + integration + 2-day on-site training. €11,000 for one drone, €30,000 for the recommended 3-drone continuous-patrol fleet, €49,000 for 5 drones to cover larger AOIs.

SAR itself runs on any MAVLink-compatible industrial drone — PX4 or ArduPilot, with a small companion computer for video and detection. Astro Max and Trinity Pro are example compatible airframes for customers whose procurement requires NDAA / Blue UAS or EU-sovereign hardware. They're illustrative, not exclusive. Bring any other MAVLink airframe and we'll licence SAR plus run the integration on it.

The strategic split is deliberate. Drone One opens the entry-tier segment, where most evaluation pilots and most first-customer deployments live. The compatible-airframe lineup serves the sovereignty-constrained segment that needs an NDAA-compliant or EU-sovereign drone. SAR doesn't pick winners between airframes — the customer's procurement does. Our job is to make sure that whichever path the customer is on, the autonomy layer above is the same.

Drone One v1 is the MVP. The job in 2026 is to get pilots flying, iterate on the integration loop, and build operational hours that turn into customer references that turn into the next round of pilots.

In parallel, the company's medium-term roadmap is the EU-sovereign Drone One chassis variant: a Belgium-or-European-origin platform that holds the same price point and the same SAR autonomy stack, with a cleaner sovereign supply chain on the components customers care about most. That's the direction we're building toward, and it's why we exist as a hardware-and-software company rather than a software-only one. Full EU sovereignty is the company's stated aim and purpose. Drone One v1 is the route to get there: each customer pilot funds the migration, each integration loop sharpens the engineering, each procurement conversation tells us which sovereign components to prioritise.

If you want to evaluate Drone One — or ask the harder question of whether our entry tier or the compatible-airframe path fits your procurement — get in touch, or run the demo if you want to see the SAR Ground Station in action before any conversation.

For the full company story, read the trilogy:

Part 1: Our First Fleet — how three Holybro X500 V2 kits in a Belgium workshop turned into a software-and-hardware company
Part 2: Why We Built SAR — what was missing in existing ground stations, and why we built the whole stack
Part 3: Why We Built Drone One (this post)