Pre-Flight Discipline: Managing Risk Before the Rotors Turn
Most drone incidents trace back to something knowable before take-off. A written, followed checklist is unglamorous and is the single highest-return practice in UAV operations.
Why checklists rather than experience
Experience makes an operator faster and better at judgement. It does not protect against the specific failure mode of a checklist: forgetting a step you have performed a thousand times, on the day you were distracted.
Aviation learned this expensively. The checklist exists precisely because competent, experienced people omit steps under time pressure, and because the omission is invisible until it matters.
Before leaving for site
Airspace and permission. Confirm the classification over the site and any restriction that applies. Check for temporary restrictions, which appear and disappear and are the most commonly missed item. Confirm authorisation is in place, not merely applied for.
Weather. Wind at altitude is typically stronger than at the surface, and gusts matter more than mean wind. Check precipitation, temperature — cold reduces battery capacity substantially — and the sun angle for imaging work.
Battery state. Charged, but also within cycle life, without swelling, and stored at an appropriate voltage if they have been idle. A swollen pack is not flown.
Firmware and software. Not on the morning of the flight. Update deliberately, then test-fly before an operational sortie. Updating on site is a well-documented way to discover a regression at the worst possible moment.
On site, before flight
Physical inspection. Propellers for nicks and cracks, and correct fitting and direction. Airframe for damage, particularly around arms and motor mounts. Motors for free rotation and absence of grinding. Fasteners for tightness. Payload and gimbal secure. Antennas intact and correctly oriented.
Site survey. Identify obstacles including those not visible from the launch point — wires are notoriously hard to see. Establish where people will be, and where they might unexpectedly come from. Choose a launch and landing area with a clear approach.
Establish an emergency plan before you need one. Where does the aircraft go if the link is lost? What is the return-to-home altitude, and is it above every obstacle on the route home? What is the abort action, and who calls it?
Set failsafes deliberately. Return-to-home altitude, low-battery thresholds, and link-loss behaviour should be chosen for this site, not inherited from the last one. An RTH altitude below the height of a nearby structure has caused many losses.
During flight
Maintain visual line of sight unless specifically authorised otherwise. Monitor battery and signal continuously rather than at intervals. Keep the aircraft where you can see its orientation.
Land early on any anomaly. An unexpected sound, an unexplained attitude correction, or a reading that does not fit — none of these improve with more flight time. The cost of an unnecessary landing is minutes; the cost of continuing is the aircraft, and possibly more.
After flight
Log the flight: duration, battery cycles, conditions, and anything unusual. This is what makes trend analysis possible — a motor gradually drawing more current, a battery losing capacity, a repeated attitude correction all appear in the log before they appear as a failure.
Inspect the aircraft after landing, not only before take-off. Damage found on the ground is a maintenance item; the same damage discovered in the air is an incident.
The habit that matters most
Write the checklist down and follow it visibly, item by item, including on routine flights, and especially when a client is watching and you feel pressure to appear efficient. The flights where you feel too experienced to need the checklist are exactly the flights where it earns its place.
References
ICAO Doc 10019, Manual on Remotely Piloted Aircraft Systems; Reason, J., Human Error; operator's manual and manufacturer maintenance schedule for the aircraft in use.
