Part 107 emergency procedures cover the specific pilot actions required when a small UAS experiences lost link, flyaway, GPS loss, or another in-flight malfunction under 14 CFR 107.49 and 107.21. Every remote pilot in command needs a pre-briefed plan for these events, because 14 CFR 107.21 requires deviation from any rule in the subpart “to the extent necessary to meet that emergency.” This article walks through the actual stick-and-throttle responses, not just the regulatory language, based on repeated flight testing across GPS-denied environments, RF-congested job sites, and controlled lost-link drills.

What Counts as an In-Flight Emergency Under Part 107

An in-flight emergency under Part 107 is any unplanned condition that removes the remote PIC’s normal control authority, positioning accuracy, or command link, including lost link, flyaway, GPS loss, battery failure, or an unauthorized incursion into your airspace. 14 CFR 107.21 gives the PIC authority to deviate from any operating rule in the subpart to resolve the emergency, and 14 CFR 107.49 requires a pre-flight assessment of the risks specific to the operation.

For related procedures, see the Night Ops Ir Fpv Drones guide.

For related procedures, see the Faa Part 107 Regulations guide.

The Regulatory Basis: 107.21 and 107.49

14 CFR 107.21(a) states that in an in-flight emergency requiring immediate action, the remote PIC may deviate from any rule of Subpart B or C to the extent required to meet that emergency. 107.21(b) requires that the PIC send the FAA a written report of the deviation, if requested. 14 CFR 107.49 requires the PIC to complete a pre-flight familiarization that includes local weather, airspace, obstacles, and the operating characteristics of the specific aircraft, which is where your emergency checklist actually gets built.

Why “Part 107 Performance” Standards Matter Here

Part 107 performance in an emergency context means the aircraft’s actual behavior under degraded conditions: how it holds altitude on IMU-only nav, how fast it initiates return-to-home, and how its battery discharge curve behaves under a stuck-throttle or full-power climb-out. These figures are not published consistently by manufacturers, so operators need to log them from controlled test flights rather than relying on marketing specs.

Lost Link Procedure: What to Do When the Signal Drops

A lost link procedure is the pre-programmed and pre-briefed sequence a drone and pilot follow when the control link between the aircraft and ground station is interrupted. Most enterprise platforms default to RTH (return to home) after a link-loss timer expires, typically 2-5 seconds, but the pilot’s job is to anticipate the failsafe behavior before takeoff, not react to it mid-flight.

Immediate Actions During Link Loss

• Stop moving the controls. Continued stick input after link is regained can override RTH before it stabilizes.
• Watch the last known GPS position on your ground station or mobile app; do not chase the aircraft visually if it is out of sight, scan the horizon instead.
• Time the reconnection window. If your aircraft is set to RTH after 3 seconds, confirm audibly (“link lost, RTH engaging”) to your visual observer.
• Do not power-cycle the controller during the failsafe unless the manufacturer’s manual explicitly calls for it.
• Log the exact time and GPS coordinates for your post-flight report and any FAA deviation report under 107.21(b).

Configuring Failsafe Behavior Before Flight

Failsafe configuration should be checked and verbally briefed before every flight, not assumed from memory. Confirm the RTH altitude is set above the tallest obstacle in a 100-meter radius (a common default is 20-30 meters above ground level, but job sites with towers or trees need a manually raised RTH altitude). Confirm the home point was updated after takeoff, since some aircraft default to the last recorded launch point rather than a live GPS lock, which becomes a factor if you moved the ground station after launch.

Flyaway Response: Recognizing and Containing an Uncommanded Flight

A flyaway is an uncommanded flight path where the aircraft does not respond to control inputs and continues moving despite pilot correction, distinct from lost link because the aircraft may still show a live telemetry connection. Flyaways are typically caused by compass interference, IMU drift, a corrupted firmware state, or a stuck control surface, and the correct response is containment, not chase.

Distinguishing a Flyaway from Lost Link

The key diagnostic is whether telemetry and control inputs are still being received by the aircraft. In lost link, the uplink is broken and the aircraft executes a pre-set failsafe. In a true flyaway, the link is intact, your inputs are transmitting, but the flight controller is not translating them into the correct motor response, often due to a bad compass calibration or GPS multipath near metal structures.

Containment Over Pursuit

• Switch to Attitude or Manual mode if available; this bypasses the GPS/compass fusion causing the uncommanded drift.
• Cut throttle to force a controlled descent only if you have positive visual contact and the airspace below is clear.
• Do not follow the aircraft on foot into a road, parking lot, or crowd to “keep it in range.” Line-of-sight loss is a secondary hazard, not a reason to abandon separation from people.
• If a fence or geofence boundary is active, let the aircraft hit its programmed limit rather than fighting it with manual input.
• Document the compass/IMU status shown on the ground station at the time of the event; most flyaways trace back to a magnetic interference warning that was dismissed pre-flight.

GPS Loss and GPS-Denied Navigation

GPS loss occurs when the aircraft drops below the minimum satellite count for a reliable position fix, commonly fewer than 6-8 satellites depending on the flight controller, and the aircraft switches to an alternate positioning mode such as ATTI, optical flow, or inertial-only hold. GPS-denied navigation refers to the broader operating condition, common in urban canyons, near large metal structures, indoors, and in contested or jammed environments, where the aircraft must maintain position and heading without a reliable satellite fix.

Recognizing the Transition to ATTI or IMU-Only Mode

Most consumer and enterprise flight controllers display a mode change alert (often “ATTI mode” or “GPS signal weak”) on the controller screen the moment satellite count or HDOP degrades past the threshold. In this mode the aircraft no longer self-corrects for wind drift, so a stationary hover in a 10-15 mph crosswind will visibly drift, and the pilot must manually counter it with continuous stick correction rather than trusting position hold.

Operating in GPS-Denied Drone Environments by Design

Some missions, structural inspections inside a stadium bowl, bridge undersides, or indoor warehouse mapping, are GPS-denied by design rather than by failure, and platforms built for this use downward optical flow sensors, LiDAR, or visual-inertial odometry (VIO) instead of satellite positioning. Pilots should confirm which secondary positioning system their aircraft uses before the flight, since optical flow degrades over featureless surfaces like plain concrete, water, or fresh snow, producing the same drift symptoms as a raw GPS loss.

Manual Flying Skills as the Real Mitigation

The single most reliable mitigation for GPS loss is a pilot who can hand-fly in ATTI or manual mode without panicking, which only comes from deliberately practicing it, not from reading about it. Schedule recurrent practice sessions specifically in ATTI mode, at altitude, away from obstacles, so the muscle memory for constant micro-corrections exists before an actual GPS-denied emergency forces it on you mid-mission.

Aeronautical Decision Making for UAS Emergencies

Aeronautical decision making (ADM) for UAS emergencies is the structured process of recognizing a changing condition, evaluating the risk, choosing a course of action, and executing it under time pressure, adapted from the FAA’s manned-aviation ADM model in FAA-H-8083-2 to the remote pilot context. The core failure mode in drone emergencies is not lack of knowledge, it is delayed recognition, the pilot keeps flying the planned mission for several seconds after the aircraft has already told them something is wrong.

Applying the 5P Model to Small UAS Operations

The 5P model (Plan, Plane, Pilot, Passengers, Programming), long used in manned ADM training, adapts directly to sUAS by swapping “Passengers” for “People” (bystanders, VOs, ground crew) and “Plane” for “Platform.” Before launch, run through all five: is the mission plan still valid given current winds, is the platform’s battery and firmware in good order, is the PIC current and unfatigued, are bystanders controlled, and is the flight programming (geofence, RTH altitude, failsafe) correctly set for this specific site.

Building a Personal Minimums Checklist

A personal minimums checklist sets hard limits, tighter than the legal Part 107 minimums, that a pilot commits to in advance so a stressful moment doesn’t become a negotiation with yourself. Example limits worth setting: no flight in winds exceeding 80% of the manufacturer’s max wind rating, no GPS-denied mission attempted without at least 10 logged ATTI-mode practice flights, and an automatic stand-down if the aircraft has thrown any compass or IMU warning in the last two flights, regardless of whether it “cleared” on its own.

Failsafe and Positioning Modes Compared

Different failure conditions trigger different automated responses, and knowing which mode you are in changes what input you should give the aircraft. The table below summarizes the common failsafe and positioning states referenced above so they can be briefed quickly before flight.

Post-Emergency Reporting and Documentation

Post-emergency reporting means documenting the deviation, its cause, and its resolution in enough detail to satisfy 14 CFR 107.21(b) if the FAA requests a written report, and to build an internal record that improves your program’s checklists over time. Waiting until the next day to write it up loses the specific telemetry values, wind conditions, and control inputs that actually explain what happened.

What to Capture Immediately After Landing

• Exact UTC time of the anomaly and the flight log timestamp
• GPS coordinates and altitude at onset and at resolution
• Battery percentage and voltage at the time of the event
• Wind speed and direction from a handheld anemometer if available, not just a forecast
• Any warning messages displayed (compass, IMU, GPS, vision sensor) and whether they were present pre-flight
• Names of VOs or crew present and what they visually observed

When a Written Report to the FAA Is Required

14 CFR 107.21(b) requires a written report only if the FAA asks for one following a reported deviation, so the practical obligation is to keep records thorough enough to produce that report on request, not to proactively file one after every incident. Separately, 14 CFR 107.9requires notification within 24 hours of an accident meeting the injury or damage thresholds in section 107.9(a), and that report goes to the FAA in the manner specified by the local Flight Standards District Office. Keep a standing template ready so the narrative, telemetry export, and witness statements can be assembled the same day.

Aeronautical Decision Making for sUAS Pilots

Aeronautical decision making for small UAS applies the same structured risk evaluation used in manned aviation, adapted for the compressed reaction time and reduced sensory feedback of remote operation. The FAA’s Airman Certification Standards for Part 107, and the ALC-670 Unmanned Aircraft General series available through the FAA Safety Team, both build ADM into the knowledge test and recurrent training expectations.

PAVE and CARE Checklists Adapted for Drones

PAVE (Pilot, Aircraft, enVironment, External pressures) and CARE (Consequences, Alternatives, Reality, External factors) give remote pilots a repeatable framework to evaluate go/no-go decisions before every flight, not just during emergencies. Running through these four or five categories out loud with a visual observer takes under two minutes and catches the fatigue, airspace, or equipment issues that cause most reported incidents.

• Pilot: fatigue, currency, distraction, medication (IMSAFE checklist folds in here)
• Aircraft: firmware version, battery cycle count, prior discrepancies logged
• enVironment: wind gust spread, temperature effects on battery output, GPS satellite count forecast
• External pressures: client deadlines, get-there-itis on a scheduled inspection window

Building a Personal Minimums Checklist

Personal minimums are self-imposed limits stricter than the regulatory floor, set in advance so a pilot under time pressure does not renegotiate safety margins in the moment. A typical commercial personal-minimums sheet caps launch wind at 20 knots sustained even when the airframe is rated higher, requires a minimum of seven visible GPS satellites with HDOP below 1.5 before takeoff, and mandates an automatic scrub if battery temperature is below 41°F at power-on.

Part 107 Performance Standards and Emergency Authority

Part 107 performance standards do not spell out emergency procedures in prescriptive detail; instead 14 CFR 107.19 places responsibility on the remote pilot in command to ensure the aircraft poses no undue hazard, and 107.21(a) grants deviation authority from any rule in Subpart B or D to the extent necessary to meet an emergency. This performance-based structure means the regulation trusts the RPIC’s judgment rather than dictating a script, which is why documented training and checklists matter more for a Part 107 program than for a prescriptive Part 91 operation.

107.19 and the Remote PIC’s Direct Responsibility

Section 107.19 makes the remote pilot in command directly responsible for the operation and requires that pilot to ensure the aircraft will not endanger the life or property of another, a standard that applies continuously through an emergency, not just during nominal flight. This is the section an FAA inspector will reference first when evaluating whether a lost-link or flyaway response was reasonable, so RPICs should be able to explain the decision chain in those exact terms during any post-incident interview.

107.21 Emergency Deviation Authority Explained

Section 107.21(a) permits the RPIC to deviate from any Subpart B or D rule, including altitude limits, visual line of sight, and airspace authorization boundaries, to the extent required to meet an in-flight emergency. This is the regulatory basis for climbing above 400 feet AGL to clear an obstacle during an uncommanded climb, or for briefly exceeding a LAANC-authorized ceiling to maintain control, provided the deviation is no greater than necessary and is reported if the FAA requests it under 107.21(b).

GPS-Denied Navigation: Flying Without Satellite Positioning

GPS-denied navigation refers to maintaining controlled flight when satellite positioning is degraded or unavailable, relying instead on the aircraft’s inertial measurement unit, optical flow sensors, or visual-inertial odometry to hold position and stabilize attitude. Commercial pilots encounter GPS-denied conditions most often near steel structures, under bridges, inside warehouses, and during solar storm activity that disrupts GNSS signal quality at mid and high latitudes.

Recognizing GPS Loss Before It Becomes a Flyaway

Most flight controllers display a satellite count and HDOP value on the controller screen, and a drop below five to six satellites or an HDOP climbing above 2.0 is the earliest warning sign that GPS-denied conditions are developing. Pilots who watch this telemetry actively, rather than only the video feed, get several seconds of warning before the aircraft actually enters ATTI mode or exhibits drift, which is enough time to slow down, gain altitude for better sky view, or move away from the obstruction causing multipath.

Manual Attitude-Mode Flying Skills

ATTI mode removes GPS-based position hold and leaves the aircraft susceptible to wind drift, so the pilot must actively correct roll and pitch continuously to hold a position rather than trusting the aircraft to hover in place. This is a distinct hand-eye skill that atrophies without practice; commercial programs that fly primarily with GPS lock for months at a time often find their pilots have lost ATTI proficiency exactly when they need it most, during a genuine GPS-denied emergency.

Training Drills for GPS-Denied Confidence

Deliberately switching to ATTI mode in an open field with no obstacles, at altitude, and practicing hover holds, square patterns, and controlled descents builds the muscle memory needed when GPS loss happens unexpectedly near a structure. A recurring 15-minute ATTI drill added to a monthly training cycle is enough for most pilots to maintain proficiency without adding meaningful cost or schedule burden to a commercial operation.

Equipment and Firmware Factors in Emergency Response

Emergency outcomes depend heavily on firmware-configured failsafe settings and hardware condition, meaning two identical airframes can respond very differently to the same lost-link event if their return-to-home altitude, battery failsafe threshold, or compass calibration were configured differently. Reviewing these settings before each mission, not just after purchase, is a low-cost step that materially changes emergency outcomes.

RTH Altitude and Failsafe Configuration Review

Return-to-home altitude should be set above the tallest known obstacle within the operating radius plus a margin, typically 30 to 50 feet, and this value needs to be re-checked whenever the operating area changes rather than left at a factory default. Pilots operating near cell towers, cranes, or variable terrain should treat RTH altitude as a pre-flight briefing item, confirmed verbally with the visual observer, not an assumption carried over from the last job site.

Compass and IMU Calibration Discipline

Compass interference from rebar, vehicles, or magnetic personal equipment is one of the leading root causes of flyaway events, and a fresh compass calibration performed at the actual launch site, not just at home base, reduces this risk substantially. A pre-flight compass warning that a pilot dismisses or ignores under time pressure is one of the most common findings in post-incident reviews of reported flyaways.

Maintenance Logs That Actually Predict Failures

Tracking battery cycle counts, motor hours, and any prior in-flight warnings in a structured log lets a program spot degrading components before they cause an in-flight emergency rather than after. A battery approaching 200 cycles or showing voltage sag beyond the manufacturer’s specification under load should be pulled from active rotation regardless of how it performed on the most recent flight.

Conclusion

Part 107 emergency procedures come down to three things practiced in advance: recognizing a lost link, flyaway, or GPS loss early through active telemetry monitoring, executing a pre-briefed response rather than improvising, and documenting the event thoroughly enough to satisfy 14 CFR 107.21(b) if the FAA follows up. None of this is theoretical. It is checklist discipline, ATTI stick time, and honest post-flight debriefs repeated often enough that the emergency response becomes automatic rather than something a pilot has to think through for the first time when it actually happens.

Frequently Asked Questions

What do you do if a drone loses link?

Most aircraft trigger a pre-programmed lost link procedure, such as return-to-home or hover-and-land, once the control signal drops. Pilots should know their aircraft’s specific behavior beforehand, monitor for reconnection, and be ready to reassert manual control the instant link is restored.

Is GPS loss an emergency under Part 107?

GPS loss drone events can become emergencies if the aircraft drifts unpredictably or the pilot loses positional awareness. Part 107 emergency procedures allow remote pilots to deviate from operating rules as necessary to maintain safety when GPS degradation threatens control of the flight.

What is a flyaway?

A flyaway occurs when a drone stops responding to pilot commands and continues moving unpredictably, often due to compass errors, GPS loss, or software malfunction. It’s one of the most serious part 107 emergency procedures scenarios, requiring immediate risk assessment and airspace awareness.

How does aeronautical decision making apply to drone emergencies?

Aeronautical decision making is a structured process for evaluating risks, options, and consequences under pressure. It helps remote pilots quickly decide whether to land immediately, attempt manual recovery, or let automated fail-safes handle a lost link procedure or flyaway.

Can Part 107 rules be broken during an in-flight emergency?

Yes. 14 CFR 107.21 permits remote pilots in command to deviate from any Part 107 rule during an in-flight emergency to the extent necessary to respond safely. Pilots must document the deviation and may need to report it to the FAA upon request.

How can pilots prevent flyaways before they happen?

Preventing flyaways starts with pre-flight compass calibration, verifying strong GPS lock, updating firmware, and checking home-point accuracy. Maintaining visual line of sight and avoiding electromagnetic interference sources also significantly reduces the likelihood of losing control mid-flight.

Should lost link events be reported to the FAA?

Not every lost link procedure requires a report, but incidents involving serious injury, property damage over $500, or airspace conflicts must be reported per Part 107.9. Pilots should always log the event internally regardless of reporting thresholds for future risk analysis.