Hubbry Logo
AutopilotAutopilotMain
Open search
Autopilot
Community hub
Autopilot
logo
8 pages, 0 posts
0 subscribers
Be the first to start a discussion here.
Be the first to start a discussion here.
Contribute something
Autopilot
Autopilot
Autopilot
View on Wikipedia
from Wikipedia

The autopilot control panel of a Boeing 747-200 aircraft

An autopilot is a system used to control the path of a vehicle without requiring constant manual control by a human operator. Autopilots do not replace human operators. Instead, the autopilot assists the operator's control of the vehicle, allowing the operator to focus on broader aspects of operations (for example, monitoring the trajectory, weather and on-board systems).[1]

When present, an autopilot is often used in conjunction with an autothrottle, a system for controlling the power delivered by the engines.

An autopilot system is sometimes colloquially referred to as "George"[2] (e.g. "we'll let George fly for a while"; "George is flying the plane now".). The etymology of the nickname is unclear: some claim it is a reference to American inventor George De Beeson (1897–1965), who patented an autopilot in the 1930s, while others claim that Royal Air Force pilots coined the term during World War II to symbolize that their aircraft technically belonged to King George VI.[3]

First autopilots

[edit]
See also: Gyroscopic autopilot
A World War II-era Honeywell C-1 autopilot control panel

In the early days of aviation, aircraft required the continuous attention of a pilot to fly safely. As aircraft range increased, allowing flights of many hours, the constant attention led to serious fatigue. An autopilot is designed to perform some of the pilot's tasks.

The first aircraft autopilot was developed by Sperry Corporation in 1912.[4] The autopilot connected a gyroscopic heading indicator, and attitude indicator to hydraulically operated elevators and rudder. (Ailerons were not connected as wing dihedral was counted upon to produce the necessary roll stability.) It permitted the aircraft to fly straight and level on a compass course without a pilot's attention, greatly reducing the pilot's workload.

Lawrence Sperry, the son of famous inventor Elmer Sperry, demonstrated it in 1914 at an aviation safety contest held in Paris. Sperry demonstrated the credibility of the invention by flying the aircraft with his hands away from the controls and visible to onlookers. Elmer Sperry Jr., the son of Lawrence Sperry, and Capt Shiras continued work on the same autopilot after the war, and in 1930, they tested a more compact and reliable autopilot which kept a U.S. Army Air Corps aircraft on a true heading and altitude for three hours.[5]

In 1930, the Royal Aircraft Establishment in the United Kingdom developed an autopilot called a pilots' assister that used a pneumatically spun gyroscope to move the flight controls.[6]

The autopilot was further developed, to include, for example, improved control algorithms and hydraulic servomechanisms. Adding more instruments, such as radio-navigation aids, made it possible to fly at night and in bad weather. In 1947, a U.S. Air Force C-53 made a transatlantic flight, including takeoff and landing, completely under the control of an autopilot.[7][8] Bill Lear developed his F-5 automatic pilot, and automatic approach control system, and was awarded the Collier Trophy in 1949.[9]

In the early 1920s, the Standard Oil tanker J.A. Moffet became the first ship to use an autopilot.

The Piasecki HUP-2 Retriever was the first production helicopter with an autopilot.[10]

The lunar module digital autopilot of the Apollo program is an early example of a fully digital autopilot system in spacecraft.[11]

Modern autopilots

[edit]
The modern flight control unit of an Airbus A340

Not all of the passenger aircraft flying today have an autopilot system. Older and smaller general aviation aircraft especially are still hand-flown, and even small airliners with fewer than twenty seats may also be without an autopilot as they are used on short-duration flights with two pilots. The installation of autopilots in aircraft with more than twenty seats is generally made mandatory by international aviation regulations.

There are three levels of control in autopilots for smaller aircraft.

  • A single-axis autopilot controls an aircraft in the roll axis only; such autopilots are also known colloquially as "wing levellers", reflecting their single capability.
  • A two-axis autopilot controls an aircraft in the pitch axis as well as roll, and may be little more than a wing leveller with limited pitch oscillation-correcting ability; or it may receive inputs from on-board radio navigation systems to provide true automatic flight guidance once the aircraft has taken off until shortly before landing; or its capabilities may lie somewhere between these two extremes.
  • A three-axis autopilot adds control in the yaw axis and is not required in many small aircraft.

Autopilots in modern complex aircraft are three-axis and generally divide a flight into taxi, takeoff, climb, cruise (level flight), descent, approach, and landing phases. Autopilots that automate all of these flight phases except taxi and takeoff exist. An autopilot-controlled approach to landing on a runway and controlling the aircraft on rollout (i.e. keeping it on the centre of the runway) is known as an Autoland, where the autopilot utilizes an Instrument Landing System (ILS) Cat IIIc approach, which is used when the visibility is zero. These approaches are available at many major airports' runways today, especially at airports subject to adverse weather phenomena such as fog. The aircraft can typically stop on their own, but will require the disengagement of the autopilot in order to exit the runway and taxi to the gate. An autopilot is often an integral component of a Flight Management System.

Modern autopilots use computer software to control the aircraft. The software reads the aircraft's current position, and then controls a flight control system to guide the aircraft. In such a system, besides classic flight controls, many autopilots incorporate thrust control capabilities that can control throttles to optimize the airspeed.

The autopilot in a modern large aircraft typically reads its position and the aircraft's attitude from an inertial guidance system. Inertial guidance systems accumulate errors over time. They will incorporate error reduction systems such as the carousel system that rotates once a minute so that any errors are dissipated in different directions and have an overall nulling effect. Error in gyroscopes is known as drift. This is due to physical properties within the system, be it mechanical or laser guided, that corrupt positional data. The disagreements between the two are resolved with digital signal processing, most often a six-dimensional Kalman filter. The six dimensions are usually roll, pitch, yaw, altitude, latitude, and longitude. Aircraft may fly routes that have a required performance factor, therefore the amount of error or actual performance factor must be monitored in order to fly those particular routes. The longer the flight, the more error accumulates within the system. Radio aids such as DME, DME updates, and GPS may be used to correct the aircraft position.

Control Wheel Steering

[edit]
Servo motor for Autopilot applications

An option midway between fully automated flight and manual flying is Control Wheel Steering (CWS). Although it is becoming less used as a stand-alone option in modern airliners, CWS is still a function on many aircraft today. Generally, an autopilot that is CWS equipped has three positions: off, CWS, and CMD. In CMD (Command) mode the autopilot has full control of the aircraft, and receives its input from either the heading/altitude setting, radio and navaids, or the FMS (Flight Management System). In CWS mode, the pilot controls the autopilot through inputs on the yoke or the stick. These inputs are translated to a specific heading and attitude, which the autopilot will then hold until instructed to do otherwise. This provides stability in pitch and roll. Some aircraft employ a form of CWS even in manual mode, such as the MD-11 which uses a constant CWS in roll. In many ways, a modern Airbus fly-by-wire aircraft in Normal Law is always in CWS mode. The major difference is that in this system the limitations of the aircraft are guarded by the flight control computer, and the pilot cannot steer the aircraft past these limits.[12]

Computer system details

[edit]

The hardware of an autopilot varies between implementations, but is generally designed with redundancy and reliability as foremost considerations. For example, the Rockwell Collins AFDS-770 Autopilot Flight Director System used on the Boeing 777 uses triplicated FCP-2002 microprocessors which have been formally verified and are fabricated in a radiation-resistant process.[13]

Software and hardware in an autopilot are tightly controlled, and extensive test procedures are put in place.

Some autopilots also use design diversity. In this safety feature, critical software processes will not only run on separate computers, and possibly even using different architectures, but each computer will run software created by different engineering teams, often being programmed in different programming languages. It is generally considered unlikely that different engineering teams will make the same mistakes. As the software becomes more expensive and complex, design diversity is becoming less common because fewer engineering companies can afford it. The flight control computers on the Space Shuttle used this design: there were five computers, four of which redundantly ran identical software, and a fifth backup running software that was developed independently. The software on the fifth system provided only the basic functions needed to fly the Shuttle, further reducing any possible commonality with the software running on the four primary systems.

Stability augmentation systems

[edit]

A stability augmentation system (SAS) is another type of automatic flight control system; however, instead of maintaining the aircraft required altitude or flight path, the SAS will move the aircraft control surfaces to damp unacceptable motions. SAS automatically stabilizes the aircraft in one or more axes. The most common type of SAS is the yaw damper which is used to reduce the Dutch roll tendency of swept-wing aircraft. Some yaw dampers are part of the autopilot system while others are stand-alone systems.[14]

Yaw dampers use a sensor to detect how fast the aircraft is rotating (either a gyroscope or a pair of accelerometers),[15] a computer/amplifier and an actuator. The sensor detects when the aircraft begins the yawing part of Dutch roll. A computer processes the signal from the sensor to determine the rudder deflection required to damp the motion. The computer tells the actuator to move the rudder in the opposite direction to the motion since the rudder has to oppose the motion to reduce it. The Dutch roll is damped and the aircraft becomes stable about the yaw axis. Because Dutch roll is an instability that is inherent in all swept-wing aircraft, most swept-wing aircraft need some sort of yaw damper.

There are two types of yaw damper: the series yaw damper and the parallel yaw damper.[16] The actuator of a parallel yaw damper will move the rudder independently of the pilot's rudder pedals while the actuator of a series yaw damper is clutched to the rudder control quadrant, and will result in pedal movement when the rudder moves.

Some aircraft have stability augmentation systems that will stabilize the aircraft in more than a single axis. The Boeing B-52, for example, requires both pitch and yaw SAS[17] in order to provide a stable bombing platform. Many helicopters have pitch, roll and yaw SAS systems. Pitch and roll SAS systems operate much the same way as the yaw damper described above; however, instead of damping Dutch roll, they will damp pitch and roll oscillations to improve the overall stability of the aircraft.

Autopilot for ILS landings

[edit]

Instrument-aided landings are defined in categories by the International Civil Aviation Organization, or ICAO. These are dependent upon the required visibility level and the degree to which the landing can be conducted automatically without input by the pilot.

  • CAT I – This category permits pilots to land with a decision height of 200 feet (61 m) and a forward visibility or Runway Visual Range (RVR) of 550 metres (1,800 ft). Autopilots are not required.[18]
  • CAT II – This category permits pilots to land with a decision height between 200 feet (61 m) and 100 feet (30 m) and a RVR of 300 metres (980 ft). Autopilots have a fail passive requirement.
  • CAT IIIa – This category permits pilots to land with a decision height as low as 50 feet (15 m) and a RVR of 200 metres (660 ft). It needs a fail-passive autopilot. There must be only a 10−6 probability of landing outside the prescribed area.
  • CAT IIIb – As IIIa but with the addition of automatic roll out after touchdown incorporated with the pilot taking control some distance along the runway. This category permits pilots to land with a decision height less than 50 feet or no decision height and a forward visibility of 250 feet (76 m) in Europe (76 metres, compare this to aircraft size, some of which are now over 70 metres (230 ft) long) or 300 feet (91 m) in the United States. For a landing-without-decision aid, a fail-operational autopilot is needed. For this category some form of runway guidance system is needed: at least fail-passive but it needs to be fail-operational for landing without decision height or for RVR below 100 metres (330 ft).
  • CAT IIIc – As IIIb but without decision height or visibility minimums, also known as "zero-zero". Not yet implemented as it would require the pilots to taxi in zero-zero visibility. An aircraft that is capable of landing in a CAT IIIb that is equipped with autobrake would be able to fully stop on the runway but would have no ability to taxi.

Fail-passive autopilot: in case of failure, the aircraft stays in a controllable position and the pilot can take control of it to go around or finish landing. It is usually a dual-channel system.

Fail-operational autopilot: in case of a failure below alert height, the approach, flare and landing can still be completed automatically. It is usually a triple-channel system or dual-dual system.

Radio-controlled models

[edit]

In radio-controlled modelling, and especially RC aircraft and helicopters, an autopilot is usually a set of extra hardware and software that deals with pre-programming the model's flight.[19]

Flight Director

[edit]
Main article: Flight director (aeronautics)
Depicted here is the PFD of a G1000. The purple triangle in the center above the yellow attitude indicator is the Flight director.

A flight director (FD) is a flight instrument that is overlaid on the attitude indicator that shows the pilot of an aircraft the attitude required to execute the desired flight path. While the flight director is separate from the autopilot, they are closely linked. With a flight plan programmed into the flight computer, the flight director will command rolls when turns are required.

Without a flight director, the autopilot is limited to more basic modes, such as maintaining an altitude or a heading, or turning on to a new heading when commanded by the pilot.

When the autopilot and flight director are used together, more complex autopilot modes are possible. The autopilot can follow flight director commands, thus following the flight plan route without pilot intervention.

See also

[edit]
Wikimedia Commons has media related to Autopilots.

References

[edit]
[edit]
Revisions and contributorsEdit on WikipediaRead on Wikipedia

Autopilot

View on Grokipedia
from Grokipedia
An autopilot is a system used to control the path of an , ship, , or other without requiring constant manual intervention by a operator, typically relying on sensors, gyroscopes, and control algorithms to maintain a predetermined course or respond to environmental inputs. Originating primarily in , autopilots have become essential tools for reducing operator fatigue, enhancing precision, and improving safety across various modes of transportation. The history of the autopilot traces back to early 20th-century aviation innovations, with the first practical system developed by the in , which utilized gyroscopic technology and hydraulic actuators to enable an aircraft to fly straight and level without pilot input. This invention, demonstrated publicly in 1914 when Lawrence Sperry's plane flew hands-off during a flight exhibition in , marked a pivotal advancement just nine years after the ' first powered flight. Early autopilots were limited to basic functions like maintaining heading and altitude, but by the mid-20th century, they incorporated more axes of control—such as pitch, roll, and yaw—and integrated with navigation aids like the (ILS) for automated approaches. In contemporary , autopilots form part of integrated automated flight control systems that manage the entire , from takeoff to , including (VNAV), lateral navigation (LNAV), and for speed and thrust adjustments. Advanced features, such as for Category III instrument approaches in zero-visibility conditions, rely on components like flight management systems (FMS), inertial navigation, and GPS to follow flight plans autonomously while allowing pilot override at any time. These systems have significantly reduced pilot workload and fatigue, contributing to overall improvements in , though they require rigorous monitoring to mitigate risks like mode confusion. Beyond aviation, autopilot technology has extended to maritime vessels, where it steers ships using radar, GPS, and dynamic positioning for offshore operations, and to spacecraft for orbital adjustments and re-entry guidance. In the automotive sector, modern implementations like Tesla's Autopilot—introduced in 2014—represent advanced driver-assistance systems (ADAS) that combine adaptive cruise control, lane-keeping, and automatic lane changes to enable semi-autonomous driving on highways. Tesla's Autopilot hardware has evolved significantly, with Hardware 4 (HW4) providing roughly 3-5 times the computational power of Hardware 3 (HW3), while Hardware 5 (HW5), expected by late 2026, is projected to offer order-of-magnitude improvements—approximately 10 times more powerful than HW4—in effective capability, marking a larger performance leap than from HW3 to HW4. In January 2026, Elon Musk stated that approximately 10 billion miles of real-world training data are required for safe unsupervised self-driving in Tesla vehicles, with the company having accumulated about 7.1 billion miles and estimates suggesting the target could be reached by mid-2026. Also in January 2026, Musk announced that Tesla would discontinue outright purchases of Full Self-Driving after February 14, 2026, transitioning to a subscription-only model thereafter. As of 2026, ongoing developments in artificial intelligence and sensor fusion continue to push autopilots toward higher levels of autonomy, with applications in drones and urban air mobility promising further transformative impacts on transportation.

History

Early Concepts and Inventions

The development of autopilot systems originated in the early , driven by advancements in gyroscopic technology pioneered by Elmer Sperry. Sperry's early work on gyrocompasses for ships began around 1911, providing directional stability, but the full gyro-pilot (autopilot) system for maritime vessels—adapting innovations—was first installed in 1922 on the tanker J.A. Moffett, using a to detect deviations and servomotors to adjust the , representing the first practical automatic steering system for ships. This maritime application built on prior concepts, as Sperry's Sperry Gyroscope Company integrated electrical and mechanical elements to stabilize vessels against wave motion. Sperry's innovations quickly extended to aviation. In 1914, his son Lawrence Burst Sperry adapted the gyroscopic principles for , demonstrating the world's first airplane autopilot during a flight over the . The system, installed on a Curtiss C-2 , used gyroscopes to sense pitch and roll attitudes, automatically actuating control surfaces via servomotors to maintain stable flight, allowing Lawrence to fly hands-free while his passenger walked along the wing to demonstrate the system's stability. Central to these early designs were gyroscopes for precise attitude sensing—detecting deviations in heading, pitch, and yaw—and servomotors for reliable actuation of rudders, elevators, and ailerons, enabling automatic corrections without constant pilot input. The outbreak of in 1914 intensified interest in autopilot technology, particularly for long-range bombing missions that demanded sustained stability over extended flights. This wartime urgency prompted Elmer and to file key patents in 1916, including US1415003 for an automatic pilot using gyroscopic pendulums and servomotors to stabilize aeroplanes, and related filings for unmanned aerial torpedoes capable of precise to targets. These inventions addressed the challenges of and inaccuracy in early , though initial implementations faced significant hurdles. Early autopilot systems suffered from reliability limitations due to their mechanical complexity, with gyroscopes prone to drift from errors and servomotors susceptible to wear in harsh environments. Integration with manual controls also proved difficult, as the devices provided stabilization rather than full , requiring pilots to override systems manually during turns or , often leading to inconsistent performance and concerns. Despite these issues, Sperry's foundational work established the core principles of automatic flight control that would evolve in subsequent decades.

First Practical Autopilots

The first practical autopilots were developed and deployed in the , transitioning from experimental demonstrations to operational use in commercial and . In 1930, the Sperry Gyroscope installed an experimental model of its Gyro-Pilot autopilot on a Ford airliner, marking one of the earliest functional installations in a passenger aircraft. This system allowed the aircraft to maintain stable flight without pilot input for extended periods during tests, significantly reducing fatigue on long-haul routes. A key adoption milestone came in 1932 when () integrated the Sperry autopilot into its fleet for transcontinental flights across the , pioneering its routine commercial application on routes like New York to . By 1934, had made the Sperry system standard equipment to further alleviate pilot workload on extended journeys. These installations demonstrated the autopilot's viability for , though early versions had limitations, including high sensitivity to that often necessitated manual overrides to prevent oscillations or deviations. Technically, these pioneering systems relied on three-axis gyroscopes—one each for pitch, roll, and yaw—to sense deviations from the desired flight path. Any misalignment generated error signals, which were amplified electronically and transmitted to hydraulic actuators connected to the control surfaces, automatically applying corrections to restore stability. This closed-loop feedback mechanism formed the core of analog autopilot operation, using gyroscopic to detect changes and hydraulic power for precise adjustments without direct mechanical linkages to the controls. World War II accelerated advancements, with the C-1 Autopilot entering service in 1940 on bombers, providing automated control of heading and altitude to enable more accurate long-range missions. The C-1 built on Sperry's designs, incorporating similar three-axis gyros, vacuum-tube amplifiers for , and hydraulic servos to drive the aircraft's elevators, ailerons, and , allowing pilots to focus on navigation and bombing amid intense combat conditions. Despite its effectiveness in straight-and-level flight, the system retained vulnerabilities to severe and required periodic calibration, highlighting the evolutionary challenges of early .

Fundamental Principles

Core Components

The core components of an autopilot system form the hardware foundation that enables automatic flight control by sensing aircraft state, computing necessary adjustments, and actuating control surfaces. Primary among these are the (AHRS), which integrates gyroscopes and accelerometers to measure and provide real-time data on the aircraft's pitch, roll, and yaw attitudes, as well as heading relative to magnetic north. Air data computers complement the AHRS by processing pitot-static pressure and temperature inputs to determine key parameters such as , altitude, vertical speed, and , ensuring the system has accurate environmental data for stable flight. Actuators, typically hydraulic or electromechanical devices, interface with these sensors to physically move the primary control surfaces—ailerons for roll, elevators for pitch, and rudders for yaw—translating computed commands into mechanical actions. Sensor fusion is a critical aspect of these components, particularly through Inertial Measurement Units (IMUs) that combine outputs from rate gyroscopes and accelerometers to detect angular rates and linear accelerations across the three axes, enabling precise estimation of the aircraft's orientation and motion even in the absence of external references. This integration within the AHRS or standalone IMUs allows for robust attitude determination by cross-validating data from multiple sensors, reducing errors from individual instrument drift or environmental interference. To ensure reliability, autopilot systems incorporate redundant power sources, such as multiple independent hydraulic circuits or electrical backups, which drive the actuators and prevent single-point failures in control surface operation. Error detection relies on closed-loop feedback mechanisms, where continuous monitoring of outputs compares actual parameters against pilot-set or commanded setpoints; any deviations trigger corrective signals to the actuators, maintaining stability and path adherence. These hardware elements supply essential inputs to higher-level control laws, which process the data for automated guidance.

Control Mechanisms and Laws

Autopilot systems rely on control laws to regulate attitude and by processing data and generating commands. A foundational approach is the proportional-integral- (PID) control law, which computes the control output u(t)u(t) as a function of the e(t)e(t) between the desired setpoint and the measured state: u(t)=Kpe(t)+Ki∫0te(τ) dτ+Kdde(t)dt,u(t) = K_p e(t) + K_i \int_0^t e(\tau) \, d\tau + K_d \frac{de(t)}{dt}, where KpK_p, KiK_i, and KdK_d are the proportional, integral, and gains, respectively. This law corrects deviations in parameters such as pitch, roll, or altitude by proportionally responding to the , integrating past errors to eliminate steady-state offsets, and anticipating future errors via the term. In autopilots, PID controllers are applied to stabilize axes like pitch for altitude hold or yaw for heading maintenance, with gains tuned to ensure damping and responsiveness without inducing oscillations. To manage the complexity of multi-axis flight control, autopilots employ an inner-outer loop architecture. The inner loop focuses on rapid attitude stabilization, using high-bandwidth feedback to control short-term rates such as pitch rate or roll rate via servo commands to elevators or ailerons. The outer loop, operating at a slower rate, handles tasks by generating reference commands for the inner loop, such as a heading hold achieved through proportional gain on yaw to align the with a desired course. This cascaded structure decouples fast dynamics (e.g., stabilization) from slower ones (e.g., path following), enhancing overall system stability and modularity in both analog and digital implementations. Gain scheduling adapts control parameters to varying flight conditions, preventing performance degradation or . In autopilots, gains like those in PID laws are adjusted as functions of , altitude, or flight phase—such as increasing during high-speed cruise to counter aerodynamic changes or reducing gains in low-speed maneuvers to avoid overcontrol. This technique linearizes the nonlinear dynamics across an operating envelope, ensuring consistent handling qualities; for instance, in digital systems, precomputed gain tables interpolate values based on real-time measurements. To mitigate failure modes, autopilots incorporate safeguards against integrator windup and excessive . Integrator windup occurs when the integral term accumulates during saturation, leading to delayed recovery and potential overshoot; prevention methods include anti-windup schemes like conditional integration, where the integrator is frozen when saturate, or back-calculation to feed actuator limits back to reset the integral state. Additionally, authority limits cap control outputs to predefined bounds, avoiding structural overload or pilot override conflicts while maintaining safe operation. These measures ensure robustness, particularly in transient scenarios like gust encounters.

Types of Autopilot Systems

Stability Augmentation Systems

Stability Augmentation Systems (SAS) are feedback control systems integrated into flight controls to enhance dynamic and static stability by automatically applying low-authority corrections to control surfaces, primarily in response to short-period oscillations or disturbances rather than commanding specific flight paths. These systems address inherent instabilities in design, such as those arising from relaxed static stability in high-performance fighters, by oscillatory modes like the (a coupled lateral-directional oscillation), (long-period longitudinal motion), and spiral divergence (uncoordinated roll-yaw buildup). The purpose is to improve handling qualities, reduce pilot workload, and ensure safe recovery from perturbations without overriding manual inputs, making them essential for aircraft that would otherwise exhibit poor natural . A common example of an SAS component is the yaw damper, which uses a rate gyroscope to detect yaw rate and commands rudder deflections to counteract Dutch roll tendencies, effectively increasing the damping ratio of this mode from near-neutral to well-damped levels. In the F-16 Fighting Falcon, the roll damper within the SAS employs feedback from roll rate sensors and lateral accelerometers to apply proportional aileron inputs, stabilizing the roll subsidence mode in this inherently unstable airframe designed for enhanced agility. These systems operate with limited authority—typically 10-20% of full control surface deflection—to prioritize pilot authority while providing subtle corrections. Implementation of SAS involves sensors (e.g., gyros or accelerometers), a controller for , and actuators linked to control surfaces, with gains calibrated through to match the specific airframe's dynamics and achieve target handling qualities per military specifications like MIL-STD-1797. Washout filters, implemented as high-pass filters with time constants around 1-5 seconds, are critical to eliminate steady-state biases from sensor inputs, ensuring the system responds only to transient rates and fades out during prolonged maneuvers, thus preserving pilot override capability. SAS often reference basic proportional-integral control laws for rate damping, tuned via root locus or methods during design. Historically, SAS originated in the amid the transition to supersonic military jets, where high-speed inertial coupling and low-frequency oscillations in aircraft like the F-100 Super Sabre necessitated electronic augmentation for , evolving from analog rate feedback to address issues not fully mitigated by aerodynamic fixes. In modern , digital SAS have become accessible, with systems like Garmin's Electronic Stability and Protection (ESP) providing automated attitude limiting and recovery in certified , extending military-grade stability enhancements to non-fly-by-wire platforms.

Heading and Attitude Control Systems

Heading and attitude control systems in autopilots provide essential capabilities for maintaining or altering an 's orientation during flight, focusing on lateral and vertical axes to support basic without constant pilot intervention. The heading select mode enables the autopilot to intercept and hold a specified magnetic heading, typically derived from inputs like a magnetic or, in modern setups, GPS-derived track data. This mode commands the to execute turns by rolling to a predetermined bank angle, often limited to 25 degrees to ensure coordinated and efficient maneuvering while avoiding excessive structural loads or passenger discomfort. Attitude modes form the core of these systems, with pitch hold maintaining a constant nose-up or nose-down attitude to facilitate level flight or controlled climbs and descents, reducing pilot workload in turbulent conditions. Roll hold complements this by keeping the wings level relative to the horizon, preventing unintended banks and promoting stability. Advanced variants include vertical speed modes, which regulate climb or descent rates (e.g., 500-1,000 feet per minute) based on pilot presets, and altitude preselect functions that automatically adjust pitch to capture and hold a target altitude, often integrating barometric data for precision. Integration with navigation aids enhances track-following accuracy, where the autopilot couples to VOR or ILS signals to maintain radials or localizer paths by generating corrective roll commands proportional to deviation angles. For VOR tracking, the system uses the error angle from the station to adjust heading, similar to ILS localizer guidance but adapted for radial navigation. Authority limits are imposed to safeguard against overcontrol, capping bank angles at 25 degrees and pitch attitudes within ±20 degrees, with automatic disengagement if limits are approached during non-normal operations to prevent hazardous maneuvers. Advancements in the 1970s marked a pivotal shift toward dual-channel in autopilot designs, enabling fail-operational capability where a single channel failure allows continued safe operation without loss of control authority. This was driven by digital flight control studies emphasizing multiple redundant channels for reliability in . Post-2000 developments further integrated RNAV systems, allowing autopilots in equipped aircraft to track GPS-generated courses seamlessly in NAV mode, supporting performance-based without traditional ground-based aids.

Modern Aviation Autopilots

Integrated Flight Control Systems

Integrated flight control systems represent a significant advancement in , evolving from standalone autopilots to fully cohesive digital architectures that manage multiple flight parameters through computer-driven processes. The transition began in the late 1980s with the A320, the first commercial airliner to introduce a fully digital (FBW) system, where autopilot functions are embedded within the primary flight control computers that interpret and execute pilot or automated commands without mechanical linkages. In FBW designs, the autopilot integrates directly with flight control laws, processing sensor data to command actuators for precise attitude, trajectory, and envelope protection, thereby reducing pilot and enhancing stability across all flight phases. This integration marked a shift from earlier analog or hybrid systems, enabling seamless mode transitions and optimized performance in commercial operations. Key features of these systems include Lateral Navigation (LNAV) and Vertical Navigation (VNAV) modes, which couple the autopilot to the (FMS) for automated guidance along predefined waypoints. LNAV directs lateral path tracking by computing deviations from the FMS route and adjusting roll commands accordingly, while VNAV computes and maintains vertical profiles based on altitude constraints, speed schedules, and performance models stored in the FMS database. These modes enable precise en-route navigation and descent planning, with the autopilot pitching or powering to meet geometric or time-based targets derived from the . Autothrottle coupling complements this by automatically modulating engine thrust levers to regulate , ensuring consistency with VNAV objectives during climbs, cruises, or approaches without manual intervention. Together, LNAV, VNAV, and form a unified framework for computer-managed flight, reducing errors in complex . Fault tolerance in integrated systems relies on triple modular redundancy (TMR), an architecture employing three parallel processing channels that execute identical computations in , with a voter circuit selecting the majority output to mask discrepancies from faults. This design, prominent in advanced flight control computers like those on the , achieves high reliability by tolerating single failures without performance degradation, as the voting logic isolates erroneous data in real time. TMR extends to inputs and actuators, ensuring continuous operation even under partial degradation, which is critical for maintaining control authority in safety-critical environments. Enhancements since 2010 have integrated synthetic vision systems into these autopilots, providing pilots with a computer-generated 3D view of terrain, obstacles, and on primary displays to support low-visibility operations. By fusing data from GPS, inertial sensors, and databases, synthetic vision enables the autopilot to maintain coupled guidance during instrument approaches in or night conditions, equivalent to minima in some cases. These integrations, as evaluated in FAA and studies, improve descent accuracy and alignment, facilitating safer landings without external visual references.

Control Wheel Steering and Yaw Dampers

Control wheel steering (CWS) is a pilot-assisted mode in autopilot systems that enables temporary manual inputs via the control yoke or wheel to adjust the aircraft's attitude or flight path, after which the autopilot automatically resumes maintaining the new parameters. When engaged, CWS disengages the autopilot servos, allowing the pilot to directly command changes proportional to yoke displacement, such as pitch for vertical path adjustments or roll for heading alterations. Upon releasing the yoke, the system captures and holds the established attitude or heading without requiring further pilot action, facilitating smooth transitions during en route deviations or minor corrections. This mode is particularly useful for immediate path changes while preserving the core automation, as implemented in Boeing aircraft like the 737 and 777 series. Yaw dampers complement CWS and other autopilot functions by providing continuous, low-level rudder inputs to suppress sideslip and mitigate Dutch roll oscillations, enhancing directional stability without pilot intervention. These systems typically employ feedback from sensors detecting yaw rate or sideslip angle (β), often via side-slip vanes or inertial reference units, to generate corrective rudder deflections that counteract unwanted lateral motions. For instance, in transport aircraft, yaw dampers use β-dot (sideslip rate) signals to dampen oscillations, ensuring coordinated flight even in crosswinds or turbulence. Unlike full autopilot yaw control, yaw dampers operate subtly and independently, often remaining active during manual flight to improve handling qualities. Autopilot disengagement in CWS and yaw damper operations incorporates soft modes for gradual handover, minimizing abrupt control shifts and allowing pilots to resume manual authority smoothly, especially in turbulence or during tactical maneuvers. In turbulent conditions, these systems may limit aggressive servo responses to prevent unintended disconnects, with the autopilot yielding control progressively as pilot inputs exceed thresholds, such as yoke forces beyond 50-100 pounds depending on the . This design reduces workload in dynamic environments, where full disengagement could otherwise lead to mode anomalies or loss of synchronization in fly-by-wire setups. Yaw dampers, being stability augmentation tools, typically do not disengage abruptly but can be selectively turned off for takeoff and landing to avoid interfering with deliberate rudder use. In modern , particularly , CWS and yaw dampers are standard features integrated into envelope protection systems, with newer implementations incorporating haptic feedback through the control wheel to alert pilots of limits or mode transitions. For example, the 787's flight uses force feedback in the to provide tactile cues during CWS engagement, helping prevent inadvertent exceedances of flight parameters while maintaining hybrid manual-automation interactions. These enhancements build on earlier designs by adding sensory aids that improve without overriding core autopilot logic. As of 2024, advancements in are further integrating into these systems to enable predictive and support reduced-crew operations in .

Precision Guidance and Landing

Instrument Landing System Integration

The Instrument Landing System (ILS) provides precision guidance for aircraft during the final approach phase by integrating with the autopilot to enable automated lateral and vertical tracking. The localizer component transmits a radio signal for horizontal (azimuth) guidance, aligning the aircraft with the runway centerline, while the glideslope transmitter delivers vertical guidance to maintain a stable descent path, typically at a 3-degree angle. Autopilot capture logic activates when the aircraft is within approximately 2.5 degrees of deviation from the localizer beam, ensuring reliable interception and reducing pilot workload during low-visibility conditions. In coupling modes, the autopilot's Approach (APP or APR) mode first arms the system to capture the ILS signals, then transitions to active tracking of the localizer and glideslope beams, with built-in flare logic initiating a controlled pitch-up maneuver around 30-50 feet above ground level to achieve a smooth touchdown. Error signals from the onboard ILS receiver, which detect deviations in beam alignment, directly drive the autopilot's control laws, adjusting ailerons, elevators, and rudder inputs proportionally to maintain the beam. This integration supports Category I (Cat I) operations, requiring a decision height of no lower than 200 feet above touchdown zone elevation, at which point the pilot must acquire visual references or execute a missed approach if unable. Historically, the first fully automatic landings using ILS precursors occurred in January 1945 at the Royal Aircraft Establishment in Farnborough, , where a 247D achieved blind landings using early autopilot integration with the SCS.51 system during wartime blackout conditions. In modern , redundancy is enhanced through dual ILS receivers, allowing the autopilot to switch seamlessly between primary and backup signals to mitigate single-point failures during approach.

Autopilot in Category III Approaches

Category III approaches represent the highest level of precision guidance for landings in extremely low , enabling fully automatic operations down to very low (RVR) values. These approaches are subdivided into Category IIIA, IIIB, and IIIC based on decision height (DH) and minimum RVR requirements. Category IIIA allows a DH below 100 feet (30 meters) and an RVR of at least 700 feet (200 meters), while Category IIIB permits a DH below 50 feet (15 meters) or no DH with RVR as low as 250 feet (75 meters), and Category IIIC involves no DH and RVR potentially down to zero . Autopilot systems for these approaches must incorporate to ensure safety, distinguishing between fail-operational and fail-passive configurations; fail-operational systems continue the landing after a single failure with sufficient integrity, whereas fail-passive systems disengage the autopilot upon failure, requiring manual intervention above alert height. The autoland sequence in Category III operations is a highly automated process managed by the autopilot, beginning with capture and tracking of the Instrument Landing System (ILS) signals during the approach phase. As the aircraft descends, the system transitions to flare mode typically at around 50 feet radio altitude, where pitch attitude is adjusted to reduce descent rate and achieve a smooth touchdown. Following main gear contact, the autopilot engages rollout mode, utilizing nose-wheel steering and differential braking to maintain runway centerline alignment and decelerate the aircraft, often in conjunction with autothrust reversal. Dual-channel or triple-channel autopilots provide the necessary redundancy, with at least two channels engaged throughout to monitor and cross-check each other's performance, ensuring the system can handle faults without compromising the landing. Critical sensors underpin the reliability of Category III autolands, with radio altimeters serving as the primary means for precise height measurement above the ground, essential for triggering the and rollout phases. These sensors operate by emitting radio waves and measuring the return time to detect altitude as low as a few feet, providing data independent of barometric pressure variations. System integrity is maintained through fail-active monitors, which continuously compare outputs from redundant channels and ground-based ILS monitors to detect discrepancies, alerting the if integrity falls below certification thresholds and potentially initiating a . Certification for Category III autopilot operations is governed by stringent FAA and EASA standards, requiring demonstration of system reliability, , and performance in simulated low-visibility conditions, with failures classified as "extremely remote" (typically less than 10^{-9} per flight hour). Recent advancements in the , including EASA's 2023 updates to CS-AWO for hybrid ILS/GNSS systems and for failure prediction in flight control systems, continue to enhance integrity, with applications extending to commercial and UAVs. The , equipped with triple-redundant autopilot systems, has successfully performed autolands in dense fog conditions at major airports.

Flight Director Systems

Flight director systems provide pilots with visual guidance cues on the (PFD) to assist in maintaining desired flight paths during manual or semi-automated operations. These systems compute pitch and roll commands based on inputs from the (FMS) or sources, displaying them as command bars overlaid on the . The command bars consist of a pitch bar for vertical guidance and a roll bar for lateral guidance, positioned relative to a fixed symbol to indicate deviations from the selected . Common modes include takeoff/go-around (TO/GA), which commands an initial pitch attitude of approximately 15 degrees for climb-out, and heading/track (HDG/TRK) modes for lateral . In HDG mode, the system guides the along a selected magnetic heading, while TRK mode follows the GPS-derived ground track, compensating for wind effects. These modes operate independently of the autopilot, allowing pilots to hand-fly the while cross-checking guidance against autopilot attitude modes, though the flight director serves solely as an advisory tool without actuating control surfaces. Flight directors integrate with terrain awareness and warning systems (TAWS) to display avoidance cues, such as modified command bars or alerts for proximity during deviations in . This integration enhances by overlaying -related guidance on the PFD without altering core flight control logic. Unlike autopilots, flight directors focus on display logic, including green needles on deviation scales to represent lateral and vertical offsets from navigation sources like the .

Autopilot in Unmanned and Model Aircraft

Autopilots in unmanned and model aircraft differ significantly from those in manned due to the absence of onboard pilots, emphasizing full autonomy, , and compact designs tailored to smaller platforms like drones, radio-controlled (RC) models, and unmanned aerial vehicles (UAVs). These systems prioritize lightweight components, low power consumption, and robust for operations in diverse environments, often integrating sensors for stabilization and mission execution without human intervention. In RC models, basic autopilot functionality emerged in the 1990s through gyrostabilized servos, which provided stabilization for helicopters and basic attitude control for fixed-wing hobby planes, enabling smoother flights for enthusiasts. These early systems used rate gyros to counteract and maintain heading, marking a shift from fully manual control to assisted stability in consumer-grade models. By the 2000s, open-source platforms like advanced this further, offering stabilization modes such as self-leveling for roll and pitch in multirotors and , allowing hobbyists to implement and return-to-home features on affordable hardware. 's versatility supports RC models by running on microcontrollers with integrated inertial measurement units (), fostering community-driven enhancements for hobby applications. For professional UAVs, autopilots enable full autonomy, as exemplified by the MQ-9 Reaper, which employs a triple-redundant flight control system integrated with GPS and for precise waypoint following during long-endurance missions up to 27 hours. This setup blends GPS updates with INS data to maintain navigation accuracy even in GPS-denied areas, supporting autonomous takeoff, loiter, and sequences. In swarm operations, autopilots like the VECTOR system facilitate coordinated behaviors among multiple UAVs, using distributed algorithms for , obstacle avoidance, and task allocation in military and search-and-rescue scenarios. These systems rely on inter-UAV communication to achieve emergent intelligence, where individual autopilots adjust paths based on collective data. Key challenges in these autopilots stem from size and weight constraints, which necessitate micro-electro-mechanical systems () sensors for IMUs, offering compact, low-power alternatives to traditional gyros but introducing issues like drift from temperature variations and vibrations. MEMS integration reduces payload to under 5 kg for micro-UAVs, yet demands advanced filtering to mitigate noise in dynamic flights. Beyond-visual-line-of-sight (BVLOS) operations amplify these demands, requiring autopilots with detect-and-avoid capabilities and redundant navigation to ensure safety without pilot visibility, as outlined in FAA guidelines for performance-based certification. As of 2025, the FAA has advanced BVLOS operations through performance-based rules, enabling broader commercial applications for delivery and inspection drones. In the 2020s, AI enhancements have addressed these by enabling in delivery drones, where optimizes paths in uncertain urban environments, improving obstacle detection and energy efficiency. These AI-driven systems use neural networks for real-time decision-making, pushing toward commercial .

References

  1. https://www.merriam-webster.com/dictionary/autopilot
  2. https://skybrary.aero/articles/autopilot
  3. https://genesys-aerosystems.com/the-evolution-of-aircraft-autopilots-from-basic-systems-to-advanced-technology/
  4. https://www.historynet.com/lawrence-sperry-autopilot-inventor-and-aviation-innovator/
  5. https://www.aopa.org/news-and-media/all-news/2022/october/flight-training-magazine/how-it-works-autopilot
  6. https://www.faa.gov/sites/faa.gov/files/2022-01/CFIT%2520Automation%2520Overreliance.pdf
  7. https://www.history.com/articles/plane-automation-autopilot-flight-302-610
  8. https://www.tesla.com/support/autopilot
  9. https://www.jowua-life.com/blogs/jowua-blog/tesla-hw3-vs-hw4-what-has-changed-and-how-did-that-affect-tesla-accessories
  10. https://x.com/elonmusk/status/2009161554785128729
  11. https://finance.yahoo.com/news/tesla-fsd-approaches-7b-miles-143200232.html
  12. https://electrek.co/2026/01/14/tesla-tsla-stop-selling-full-self-driving-package-subscription-only/
  13. https://www.sperrymarine.com/about-us/timeline-of-sperry-gyrocompass/
  14. https://www.workboat.com/viewpoints/iron-mike-strikes
  15. https://www.invent.org/inductees/elmer-ambrose-sperry
  16. https://patents.google.com/patent/US1415003A/en
  17. https://www.history.navy.mil/content/dam/nhhc/research/histories/naval-aviation/naval-aviation-in-world-war-i/pdfs/ww1-10.pdf
  18. https://airandspace.si.edu/collection-objects/autopilot-sperry-airliner-experimental-model/nasm_A19650043000
  19. https://www.aerotime.aero/articles/autopilot-flight-automation-history
  20. https://twamuseum.org/history
  21. https://twaspirit.com/twa_firsts.html
  22. https://simpleflying.com/lawrence-sperry-the-man-who-made-the-worlds-first-autopilot/
  23. https://www.lockheedmartin.com/en-us/news/features/history/sperry.html
  24. https://www.nationalmuseum.af.mil/Visit/Museum-Exhibits/Fact-Sheets/Display/Article/195873/honeywell-c-1-autopilot/
  25. https://aeroantique.com/products/c-1-autopilot-control-panel-minneapolis-honeywell
  26. https://www.cafmn.org/news/minneapolis-honeywell-c-1-autopilot-control-panel
  27. https://skybrary.aero/articles/attitude-and-heading-reference-system-ahrs
  28. https://www.vectornav.com/resources/inertial-navigation-primer/theory-of-operation/theory-ahrs
  29. https://skybrary.aero/articles/air-data-computer-adc
  30. https://www.sentechsensors.com/news/glossary-of-terms-for-applications-of-flight-control-actuators
  31. https://www.advancednavigation.com/tech-articles/inertial-measurement-unit-imu-an-introduction/
  32. https://skybrary.aero/articles/inertial-reference-system-irs
  33. https://www.airbus.com/en/newsroom/stories/2025-07-flight-control-system-more-redundancy-to-enhance-resilience
  34. https://www.stle.org/files/TLTArchives/2013/06_June/Cover_Story.aspx
  35. https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_25_1329-1C.pdf
  36. https://ntrs.nasa.gov/api/citations/20050185598/downloads/20050185598.pdf
  37. https://ntrs.nasa.gov/api/citations/20220017318/downloads/SciTech%25202023%2520SUSAN%2520submitted%2520final.pdf
  38. https://ntrs.nasa.gov/api/citations/19760010067/downloads/19760010067.pdf
  39. https://ntrs.nasa.gov/api/citations/20110013502/downloads/20110013502.pdf
  40. https://ntrs.nasa.gov/citations/19920000213
  41. https://www.nap.edu/read/5469/chapter/6
  42. https://www.sciencedirect.com/topics/engineering/stability-augmentation
  43. https://eaglepubs.erau.edu/introductiontoaerospaceflightvehicles/chapter/aircraft-stability-control/
  44. https://www.boldmethod.com/learn-to-fly/systems/how-does-a-yaw-damper-work-in-flight-rudder/
  45. https://www.falcon-bms.com/wp-content/uploads/2021/08/FM_Developers_Notes_Part_4.pdf
  46. https://www.kybernetika.cz/content/2020/4/767/paper.pdf
  47. https://apps.dtic.mil/sti/tr/pdf/AD0717168.pdf
  48. https://aviation.stackexchange.com/questions/51106/what-is-an-autopilot-washout-filter
  49. https://www.historynet.com/supersonic-revolution/
  50. https://www.garmin.com/en-US/newsroom/press-release/aviation/2010-new-garmin-technology-brings-a-sixth-sense-to-the-cockpit/
  51. https://aviationmiscmanuals.tpub.com/TM-1-1510-262-10/TM-1-1510-262-1000150.html
  52. https://forums.flightsimulator.com/t/autopilot-behavior-not-linear-using-hdg-or-nav/326978
  53. https://www.aerostudents.com/courses/automatics-flight-control/basicAutopilotSystems.pdf
  54. https://aviationsafetymagazine.com/aircraft/autopilot-building-blocks/
  55. http://www.aerostudents.com/courses/automatics-flight-control/navigationalAutopilotSystems.pdf
  56. https://apps.dtic.mil/sti/tr/pdf/ADA006411.pdf
  57. https://www.faa.gov/sites/faa.gov/files/regulations_policies/handbooks_manuals/aviation/instrument_procedures_handbook/FAA-H-8083-16B_Chapter_4.pdf
  58. https://pilotinstitute.com/fly-by-wire/
  59. https://www.airbus.com/en/newsroom/stories/2022-06-safety-innovation-1-fly-by-wire-fbw
  60. https://helitavia.com/avionics/TheAvionicsHandbook_Cap_12.pdf
  61. https://www.cfinotebook.net/notebook/avionics-and-instruments/flight-management-and-autopilot-systems.php
  62. https://skybrary.aero/articles/autothrottleautothrust
  63. https://ntrs.nasa.gov/api/citations/19880010924/downloads/19880010924.pdf
  64. https://ieeexplore.ieee.org/document/495891/
  65. https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_90-106A.pdf
  66. https://ntrs.nasa.gov/api/citations/20130013697/downloads/20130013697.pdf
  67. https://www.faa.gov/regulations_policies/rulemaking/committees/documents/media/ACOaerT1-08131991.pdf
  68. https://ntrs.nasa.gov/api/citations/19920021916/downloads/19920021916.pdf
  69. https://ntrs.nasa.gov/api/citations/19970004404/downloads/19970004404.pdf
  70. https://aviation.stackexchange.com/questions/83748/what-protections-and-feedback-does-the-fly-by-wire-of-the-boeing-777-787-provide
  71. https://www.ainonline.com/aviation-news/futureflight/2024-11-04/ai-elicits-new-generation-avionics-and-pilot-training-tools
  72. https://www.faa.gov/air_traffic/publications/atpubs/aim_html/chap1_section_1.html
  73. https://www.boldmethod.com/learn-to-fly/systems/ils-how-the-instrument-landing-system-works/
  74. https://support.redbirdflight.com/redbird-kap140-supplement
  75. https://www.faa.gov/air_traffic/publications/atpubs/pcg_html/glossary-i.html
  76. https://www.aerosociety.com/media/4856/the-rae-contribution-to-all-weather-landing.pdf
  77. https://ifr-magazine.com/technique/ils-on-the-block/
  78. https://www.faa.gov/documentLibrary/media/Advisory_Circular/120.28C.pdf
  79. https://www.easa.europa.eu/sites/default/files/dfu/cs-awo_issue_2.pdf
  80. https://skybrary.aero/articles/autoland
  81. https://aviation.stackexchange.com/questions/73402/what-does-it-mean-when-autoland-is-fail-passive-fail-operational
  82. https://www.key.aero/article/evolution-autoland-airliners
  83. https://www.pprune.org/archive/index.php/t-569438.html
  84. https://www.faa.gov/documentLibrary/media/Order/6750.24.pdf
  85. https://www.faa.gov/documentLibrary/media/Advisory_Circular/AC_120-118.pdf
  86. https://www.easa.europa.eu/en/document-library/certification-specifications/cs-awo-amendment-5
  87. https://www.sciencedirect.com/science/article/abs/pii/S0019057822000209
  88. https://www.faasafety.gov/files/events/SO/SO09/2021/SO09104630/G1000_System_Components_QuickRef_Guide.pdf
  89. http://www.jettairx.com.br/backup/wp-content/uploads/2015/07/Bombardier_Challenger_Global_00-Automatic_Flight_Control_System.pdf
  90. https://static.garmin.com/pumac/190-00498-08_0A_Web.pdf
  91. https://skybrary.aero/articles/take-go-around-toga-mode
  92. https://preflight.dynonavionics.com/2014/02/did-you-know-dynon-autopilot-hdg-vs-trk.html
  93. https://safetyfirst.airbus.com/app/themes/mh_newsdesk/documents/archives/automatic-nav-engagement-at-go-around.pdf
  94. https://ntrs.nasa.gov/api/citations/20080018605/downloads/20080018605.pdf
  95. http://static1.1.sqspcdn.com/static/f/1003521/22282885/1364288651217/b737ng-flight-instruments.pdf?token=nqD3FXjasFkTZKb3JBf8YdLCOcU%253D
  96. https://www.mdpi.com/1424-8220/15/10/26212
  97. https://www.modelaircraft.org/sites/default/files/files/MyersGeorgeM.pdf
  98. https://www.ardupilot.org/
  99. https://www.ga-asi.com/remotely-piloted-aircraft/mq-9a
  100. https://www.airandspaceforces.com/PDF/AircraftAccidentReports/Documents/2013/111213_MQ9_NewYork.pdf
  101. https://www.uavnavigation.com/uav-swarming
  102. https://www.fierceelectronics.com/components/mems-sensors-are-heart-a-drone
  103. https://www.faa.gov/regulations_policies/rulemaking/committees/documents/media/UAS_BVLOS_ARC_FINAL_REPORT_03102022.pdf
  104. https://www.faa.gov/uas/advanced_operations/bvlos
  105. https://www.mdpi.com/2504-446X/8/7/296
Add your contribution
Related Hubs
Contribute something
User Avatar
No comments yet.