Over-instrument during development
The data you did not capture is the data the intermittent fault needed. During development, capture generously.
For Manufacturers
The bug you cannot reproduce is in your logs
Building an aircraft means chasing intermittent faults across scattered logs, on a schedule. The data that explains most of them was already recorded — it just was not somewhere anyone could ask questions of it.
Flight test
What actually goes wrong, from experience
The interference bug that cost three weeks
GPS accuracy that degrades only when a particular subsystem is active is the classic example: intermittent, hardware-dependent, and invisible in any single flight log.- GPS and USB interferenceA well-known and still frequently rediscovered failure mode. Moving affected subsystems to Ethernet removes the coupling entirely.
- CAN with telemetryWhere Ethernet is not practical, CAN-based subsystems that report their own state let you see the fault instead of inferring it.
- Design follows requirementsMultirotor, VTOL quadplane, fixed-wing STOL — the airframe follows the mission, and the instrumentation should follow the airframe.
The interference bug
Instrumentation
Log more than you think you need
Storage is cheap. Reproducing a fault six months later is not.
Share it across the team
Logs that live on the test engineer's laptop help one person. Logs in one place help everyone who touches the aircraft.
More data, faster iteration
The rate at which you can answer questions about the aircraft sets the rate at which you can improve it.
Device · parameters and their change history: who changed what, when
Today vs. tomorrow
How analysis is done now, and what it costs
Desktop tools, one flight at a time
The industry standard is a desktop log viewer, opened per flight, by one engineer. It works — for one flight. It cannot answer anything about the campaign.- Centralised loggingEvery test flight lands in the same place automatically, so the archive builds itself rather than depending on discipline.
- Time-series across campaignsCompare this week's builds against last month's, and see the trend that no individual flight review would reveal.
- Planning and ops for developmentCloud-based flight planning and operations management for the test fleet, so campaign coordination stops being a spreadsheet.
Campaign analysis
After the sale
Your aircraft has a life after it ships
Warranty backed by data
When the fleet reports what actually happened, warranty questions are settled by evidence instead of by argument.
Field failure modes
What breaks at customer sites is the input your next design revision needs most, and the hardest to collect.
Software your customers use
Ship an operations platform with the aircraft rather than leaving each customer to assemble one.
Simulation · the same onboard client on four robots, each on its vendor’s interface
Pilot programme for manufacturers
We work with OEM teams during development, not just after launch. Tell us what you are building.