Meet DEXI‑10.
Research takes flight.
DEXI‑10 is the top of the flight path, a research-class quadcopter, with a full Raspberry Pi 5 + custom HAT for serious onboard compute. Big payload, optical flow indoors and out, and the headroom for autonomy research and industry applications.
A research platform that flies.
DEXI‑10 is the top rung of the flight path, a research-class airframe built for the work professional PX4 developers prototype on. A full Raspberry Pi 5 on a custom DroneBlocks HAT gives it more onboard compute than any other DEXI: enough for serious computer vision, AI, and ROS 2 autonomy.
A real research platform
A research-class airframe of the kind professional PX4 developers prototype on, sized and built for serious work.
Full Raspberry Pi 5 compute
A complete Pi 5 on a custom DroneBlocks HAT: more headroom than any other DEXI for vision, AI, and ROS 2.
Built for payloads
The airframe carries real sensors (cameras, LiDAR and manipulators), not just a flight controller.
Outdoor + payloads
optical flow for missions indoors and out, and the open platform for multi-drone coordination.
Industry-grade stack
PX4 + ROS 2 + C++, the exact same tools shipping in commercial and research drones today.
Caps the flight path
The top rung, the platform for applied research and capstone work after Hover Lab and Flight Deck.
Straight off the CAD.
The DEXI‑10 is in bringup, so these are renders of the actual Onshape assembly rather than photographs. The geometry is the real thing: the same model the build guide is generated from, down to the arm spacing and the payload bay under the center plate.
These are renders. Photography follows once the aircraft is out of bringup.
The top rung
of the flight path.
DEXI‑10 is introduced at Flight Deck and carries through Vertical Innovation. Students who built autonomy on a developer kit add a research-class airframe with a full Pi 5, big payload, and the headroom for applied research and industry applications.
DEXI‑3
An aircraft they build, Python and ROS 2, flown indoors, GPS-free.
DEXI‑5
ROS 2 autonomy, computer vision, and optical-flow flight indoors and out.
DEXI‑10
Research-class airframe, full Pi 5, payloads, and endurance.
Everything a student learned lower on the flight path carries up, the same code, the same curriculum spine, the same open platform, now flying applied research on a research-class drone.
One kit. Research-ready.
Every DEXI‑10 is a full developer kit, Raspberry Pi 5 + HAT, camera, and the open DEXI OS included. Choose how it arrives and how many.
| What's in every DEXI‑10 | Included |
|---|---|
| Flight stack | |
| Flight controller (PX4 1.17) | ✓ |
| Optical flow + range sensor | ✓ |
| GPS module (add-on, coming soon) | , |
| 4S LiPo battery + charger | ✓ |
| Code stack | |
| Raspberry Pi 5 + custom DroneBlocks HAT | ✓ |
| Camera | ✓ |
| Open DEXI OS (ROS 2 + vision) | ✓ |
| Capability, every unit | |
| DroneBlocks / Python / ROS 2 / C++ from day one | ✓ |
| Computer vision & AI on the Pi 5 | ✓ |
| Optical-flow hover, indoors and outdoors | ✓ |
Everything is included, there is no flight-only tier and no code upsell. Every DEXI‑10 is a full research developer kit.
The drone is the airframe.
DEXI‑10 is the research platform.
A DEXI‑10 student finishes the term running computer vision on a full Pi 5, flying autonomous outdoor missions on optical flow, and coordinating multiple drones, on a real, open, research-class platform.
Applied research & capstones
The top rung of the flight path is the platform for applied research and capstone projects: an open PX4 / ROS 2 / C++ stack with nothing locked down, sized for real college, research, and industry work.
Computer vision & AI on Pi 5
A full Raspberry Pi 5, the most onboard compute in the DEXI line, runs vision and AI models natively. Object detection, tracking, and custom neural networks have real headroom to run.
Long missions, indoors and out
DEXI‑10 flies outdoors on optical flow today; GPS waypoint missions arrive with our add-on, coming soon, and a research-class airframe carries the payload to do useful work on those missions.
Room for multi-drone work
An open platform with the compute and airframe headroom for multi-drone research, and the payload capacity for the domain applications that university labs run today.
Same Python script.
Sim today. Real DEXI‑10 tomorrow.
No drone. No toolchain. No install.
Students open code-server in their browser, write a MAVSDK Python script, and fly an autonomous mission in the Unity sim. No drone. No toolchain. No install.
One line of config different.
The exact same file runs against a DEXI‑10 on the lab Wi-Fi. Same takeoff, same waypoints, same land, just one line of config different.
address = "udpin://0.0.0.0:14540" # sim
address = "udpout://192.168.x.x:14540" # real DEXI-10
The whole bridge is one line. That promise is what lets a Vertical Innovation subscription deliver value the day a teacher buys it, weeks before the drone arrives, while batteries charge, when one drone has to serve thirty students. The simulator is not a toy version of the curriculum; it is the curriculum.
Vertical Innovation +
DEXI‑10 hardware
The full path is the DEXI‑10 developer kit plus a Flight Deck or Vertical Innovation subscription on my.droneblocks.io: one PO, one curriculum spine, students starting in the simulator on day one and running real research on a research-class drone the day it arrives.
Sim → Code → FlyNothing about DEXI‑10
is a black box.
The drone firmware, OS image, 65+ open ROS 2 packages, and 3D-printable parts are all public. Fork them, modify them, contribute back.
Pre-built Linux OS image for the Raspberry Pi 5, bundled with the full ROS 2 stack and PX4 link.
github.com/DroneBlocks/dexi-osROS 2 offboard flight control: takeoff, land, position and velocity setpoints.
github.com/DroneBlocks/dexi_offboardSTL files for replacement and customization parts, the same parts shipped on the production drone.
github.com/DroneBlocks/dexi-3d-prints