How often does the LTE link drop?
Drop frequency and duration per airframe, per site and per firmware version — so link problems become a pattern rather than an anecdote.
Log Insights
Stop grepping tlogs. Ask questions instead
Flight logs are rich and slow. The data to explain a recurring fault is almost always already recorded — it is just spread across hundreds of flights nobody has time to open. Log insights turns that archive into answers.
Questions
Things your fleet can finally tell you
Each of these is one query across the whole flight history, not a manual afternoon.
Is bad AHRS a compass or an IMU problem?
Correlate EKF bad-AHRS reports against compass and IMU health to separate the symptom from the sensor actually causing it.
Which IMU is noisiest?
Rank compasses and IMUs across the fleet by measured noise, so component selection is evidence rather than preference.
Flight logs · trends across the fleet: flight time, distance, speed, altitude per flight with the mean of the last seven
Comparison
Cheap autopilot or expensive autopilot?
Compare across the fleet, not within one flight
The interesting questions are comparative. Does this fault happen more on the budget autopilot tier? Only your own fleet history can answer that honestly.- Hardware A vs. hardware BPut two configurations side by side over their real flight history and see which failure modes actually differ.
- Time series, not snapshotsTrack a metric across an entire test campaign to catch the slow degradations that no single flight review would surface.
- Fleet-wide, automaticallyEvery flight contributes to the picture as it lands, rather than waiting for someone to decide an investigation is warranted.
Fleet comparison
Navigation stack
Add VIO or LIO, then prove it works
Extend the stack
Visual or lidar odometry can be added to the navigation stack without rebuilding the aircraft around it.
Validate with the same tooling
The log analysis that found the problem is what verifies the fix — no separate evaluation harness to build and maintain.
Built for flight test
Designed around how development teams actually iterate: instrument heavily, fly often, compare campaigns.
Where this stands today
Plots and trends today, the questions next
Today every flight log is plotted in the browser, flight statistics trend across the fleet and per device, and the health digital twin reduces each log to component KPIs judged against the device's own history.
The questions on this page — link drop rates, compass against IMU, the noisiest IMU in a batch — are what that pipeline is being built to answer. Ask us which of them run on your logs today.
Send us a log
Bring a flight log and a question about your fleet — we will show you what the analysis returns.