Автоматический Пилот #autopilot #crewlife #aviation
Airliner Automation by Daniel Falh
I’m often asked about the all-mighty autopilot and how truly capable the system is at flying the aircraft. We’ve all heard the saying that “airplanes practically fly themselves these days”, but little is known about the pilot workload behind managing and monitoring the automation systems on board today’s commercial aircraft. The automation used is extremely accurate, even more so in many cases than the pilots hands themselves. However, an autopilot can do an incredibly precise job of maintaining an undesired state. Thorough interfacing by the flight crew and knowing when manual intervention is necessary is crucial.
Surprisingly, one of the highest phases of workload for pilots occurs right at the gate prior to departure. It’s here that the automation systems are fed the information they need in order to assist in navigating a particular flight. After powering up the aircraft and performing various systems checks, the flight crew enters information into an onboard computer called a flight management system, or FMS. The FMS links various information fed by the pilots, such as the route and altitude to be flown, to a flight director system. The flight director system in turn provides various visual cues via instrumentation for the pilots and/or autopilot to follow for proper aircraft guidance. Once in flight, various levels of automation can be engaged that will reduce pilot workload.
The most basic level of automation begins with what we call “raw data”. This involves manually flying the aircraft with the absence of the autopilot and all flight director cues. The highest level of automation involves complete aircraft control in terms of speed, altitude, and direction. Often pilots will use a combination of these levels, such as manually flying the aircraft while following the guidance of a flight director. Some situations require more automation, while others require less.
Regardless of the level of automation being used, it is vastly important for the pilots to monitor the state of the aircraft in all phases of flight. Past accidents, attributed to the use of automation, may have been prevented had pilots been aware of problems being masked by the autopilot. Case is point, many years ago, ice accumulation on the wings of a regional turboprop aircraft formed beyond the reach of the aircraft anti-ice systems. The ability for lift to be produced with such ice build-up was hindered dramatically. The autopilot did a fantastic job of maintaining the assigned altitude up until the point of an aerodynamic stall. The autopilot system is smart, yet it can be inept. The system is similar to a balloon being filled by an air tank; the tank makes filling a balloon much easier, but it doesn’t know when to stop before popping the balloon. It takes manual intervention at a certain point to control potential chaos. Manually flying the aircraft in these conditions would have revealed a sensation of abnormal control loading, which would have alerted the pilots of a problem. The autopilot, masking this high load on the flight controls, worked right up to the point that it could no longer hold the aircraft steady. This situation highlights the importance of pilot situational awareness and knowing when automation can become the enemy. Pilots are trained to keep a hand on the control yoke at lower altitudes during autopilot use to detect unusual forces that would otherwise go un-noticed.
In today’s airline environment, workload management is a key topic discussed during annual training. Concepts such as this all stem from the origins set forth decades ago by United Airlines through their Crew Resource Management concept, or CRM. The precedent set fourth here touted the concept that the captain was the end-all-be-all entity on the flight deck. CRM stresses the importance of flight crews working together in unison in all phases of flight, be it an emergency or not. Workload management is a tactic crews use to handle complex situations that can often arise simultaneously. For example, at airports that utilize multiple, closely spaced runways, a precise means of aircraft separation after takeoff is necessary. Aircraft are assigned departure routes that must be followed both laterally and vertically with extremely close tolerances. Coupling that demand with other post-takeoff tasks such as radio communication, aircraft configuration changes, and monitoring engine instrumentation, can quickly maximize the capacity of the pilots. Throw in an engine fire, and there isn’t much room to accommodate the additional workload. In this case, as in many others, the autopilot plays the role of a third crewmember by freeing up the hands of a capable pilot to manage other tasks.
The advances in aircraft automation have made it possible for aircraft to go as far as landing themselves. Alongside such strides has been the advancement in indication systems that alert pilots of a problem. It’s these warning systems that prompt the decision to increase or decrease the level of automation being used. However, a computer cannot replace the complexity and capability of the human brain to react to dynamic situations that only manual flying can handle. I’m sure the pilots who manually guided their highly automated Airbus into the Hudson River would agree!
I’m often asked about the all-mighty autopilot and how truly capable the system is at flying the aircraft. We’ve all heard the saying that “airplanes practically fly themselves these days”, but little is known about the pilot workload behind managing and monitoring the automation systems on board today’s commercial aircraft. The automation used is extremely accurate, even more so in many cases than the pilots hands themselves. However, an autopilot can do an incredibly precise job of maintaining an undesired state. Thorough interfacing by the flight crew and knowing when manual intervention is necessary is crucial.
Surprisingly, one of the highest phases of workload for pilots occurs right at the gate prior to departure. It’s here that the automation systems are fed the information they need in order to assist in navigating a particular flight. After powering up the aircraft and performing various systems checks, the flight crew enters information into an onboard computer called a flight management system, or FMS. The FMS links various information fed by the pilots, such as the route and altitude to be flown, to a flight director system. The flight director system in turn provides various visual cues via instrumentation for the pilots and/or autopilot to follow for proper aircraft guidance. Once in flight, various levels of automation can be engaged that will reduce pilot workload.
The most basic level of automation begins with what we call “raw data”. This involves manually flying the aircraft with the absence of the autopilot and all flight director cues. The highest level of automation involves complete aircraft control in terms of speed, altitude, and direction. Often pilots will use a combination of these levels, such as manually flying the aircraft while following the guidance of a flight director. Some situations require more automation, while others require less.
Regardless of the level of automation being used, it is vastly important for the pilots to monitor the state of the aircraft in all phases of flight. Past accidents, attributed to the use of automation, may have been prevented had pilots been aware of problems being masked by the autopilot. Case is point, many years ago, ice accumulation on the wings of a regional turboprop aircraft formed beyond the reach of the aircraft anti-ice systems. The ability for lift to be produced with such ice build-up was hindered dramatically. The autopilot did a fantastic job of maintaining the assigned altitude up until the point of an aerodynamic stall. The autopilot system is smart, yet it can be inept. The system is similar to a balloon being filled by an air tank; the tank makes filling a balloon much easier, but it doesn’t know when to stop before popping the balloon. It takes manual intervention at a certain point to control potential chaos. Manually flying the aircraft in these conditions would have revealed a sensation of abnormal control loading, which would have alerted the pilots of a problem. The autopilot, masking this high load on the flight controls, worked right up to the point that it could no longer hold the aircraft steady. This situation highlights the importance of pilot situational awareness and knowing when automation can become the enemy. Pilots are trained to keep a hand on the control yoke at lower altitudes during autopilot use to detect unusual forces that would otherwise go un-noticed.
In today’s airline environment, workload management is a key topic discussed during annual training. Concepts such as this all stem from the origins set forth decades ago by United Airlines through their Crew Resource Management concept, or CRM. The precedent set fourth here touted the concept that the captain was the end-all-be-all entity on the flight deck. CRM stresses the importance of flight crews working together in unison in all phases of flight, be it an emergency or not. Workload management is a tactic crews use to handle complex situations that can often arise simultaneously. For example, at airports that utilize multiple, closely spaced runways, a precise means of aircraft separation after takeoff is necessary. Aircraft are assigned departure routes that must be followed both laterally and vertically with extremely close tolerances. Coupling that demand with other post-takeoff tasks such as radio communication, aircraft configuration changes, and monitoring engine instrumentation, can quickly maximize the capacity of the pilots. Throw in an engine fire, and there isn’t much room to accommodate the additional workload. In this case, as in many others, the autopilot plays the role of a third crewmember by freeing up the hands of a capable pilot to manage other tasks.
The advances in aircraft automation have made it possible for aircraft to go as far as landing themselves. Alongside such strides has been the advancement in indication systems that alert pilots of a problem. It’s these warning systems that prompt the decision to increase or decrease the level of automation being used. However, a computer cannot replace the complexity and capability of the human brain to react to dynamic situations that only manual flying can handle. I’m sure the pilots who manually guided their highly automated Airbus into the Hudson River would agree!