We're Making It Fly Home: CDOSR at the 2026 National Finals

Published: March 26, 2026

Seven times in eight years.

That's how many times CDOSR has qualified for the Romanian CanSat National Competition. In a field where most teams struggle to qualify once, seven is not an accident, it's a culture.

And this year, we're back.

The 2026 Mission: Glider Meets Science

This year, we didn't ask ourselves how to qualify. We asked ourselves: what's the hardest problem worth solving?

The answer: make it fly home.

Most CanSats descend on a parachute, a passive, elegant, well-understood solution. Ours won't. After rocket ejection at up to 1,000 meters, our device deploys as a fixed-wing glider and navigates autonomously back toward the launch site using GPS and inertial sensors.

That alone would be enough for a competition entry. But we went further.

While gliding, the craft continuously samples the atmosphere across the full descent envelope:

• 🌡️ Temperature: profiling thermal gradients through the boundary layer
• 💧 Relative humidity: tracking moisture variation with altitude
• 📊 Barometric pressure: cross-validated against GPS for sensor fusion accuracy

A parachute descent is a fall with data attached. A guided glide is a flight with a purpose. The extended time-of-flight, the controlled trajectory, the ability to sample across a lateral transect rather than a purely vertical drop, these aren't just engineering choices. They're what turns a competition entry into a genuine scientific instrument.

Why This Matters

Guided autonomous recovery at this scale is an unsolved problem for most small satellite and CanSat teams. Passive parachute descent is reliable precisely because it requires nothing from the vehicle after ejection. The moment you introduce active control, attitude sensing, GPS-based navigation, servo actuation, real-time decision-making, you're adding failure modes at every layer of the stack.

We're doing it anyway, because the data quality case is compelling. A guided glide extends time-of-flight, controls the sampling trajectory, and allows us to collect atmospheric profiles across a lateral transect rather than a purely vertical column. That's a meaningfully different dataset from what a parachute recovery produces.

The secondary objective is validation. We're testing whether low-cost MEMS-based inertial navigation, combined with GPS correction, is sufficient for reliable return-to-origin guidance on a sub-100g fixed-wing platform. If the answer is yes, the approach scales, to longer-duration atmospheric missions, to multi-payload recoveries, to any scenario where landing zone control matters.

That's what we're after: not just a competition result, but a proof of concept worth building on.

The Team Behind the Mission

Everything you will see in this mission, we built ourselves. The airframe, the PCB, the firmware, the ground station, none of it was bought off the shelf or handed to us. We designed it, broke it, fixed it, and designed it again.

We split into subsystems early in the season because we had to. There's too much going on in a single CanSat for one person to track all of it. So some of us spent weeks obsessing over PCB layouts and power budgets. Others lived in the guidance algorithm, running simulation after simulation until the numbers started making sense. The structures side went through more glider prototypes than we'd like to admit.

At some point it stopped feeling like separate workstreams and started feeling like one thing we were all building together, because every subsystem depended on every other one working.

That's probably the most useful thing this competition has taught us so far. Not any specific technical skill, but how to build something complex as a team without it falling apart.

What Comes Next

We are in the final stretch. Sensor calibration is ongoing. Flight control parameters are being tuned. The airframe is taking shape. There is still work to do — and we're doing it.

We won't wait until the competition is over to share it. Over the coming weeks we'll be publishing at each step of the process:

• PCB & hardware: a breakdown of the avionics board: what we designed, why we made the choices we did, and what the previous revision got wrong
• Firmware & guidance: how the navigation algorithm works, how we tested it, and what the simulation results looked like before we trusted it on real hardware
• Ground testing: bench results, sensor calibration data, and the edge cases that kept us up at night
• Integration: what happened when all the subsystems had to talk to each other for the first time
• Launch day: live updates from the competition site
• Mission debrief: the full picture: the data, the anomalies, what worked, what we'd change, and what we're already thinking about for the next one

Science isn't just what happens in the air. It's what you do with what you learn when you land.

Stay tuned.

Follow the Mission

• 📸 Social media: real-time updates from the lab and launch site
• 📧 Newsletter: in-depth mission reports, straight to your inbox (coming soon)
• 🔗 Official press release: CDOSR Selected for 2026 Romanian CanSat National Finals

CDOSR — CoderDojo Oradea Space Robotics — is a student space robotics team based in Oradea, Romania, competing since 2018 under the ROSPIN national CanSat and rocketry program.