How much do pilots actually really rely on automation?
This post contains references to products from one or more of our advertisers. We may receive compensation when you click on links to those products. Terms apply to the offers listed on this page. For an explanation of our Advertising Policy, visit this page.
Last week it was reported that a Boeing 777 taking off from Dubai came within metres of hitting buildings at the end of the runway after getting airborne. The exact details of the incident are unclear and the investigation that is in progress will no doubt clear these up.
However, it seems most likely that there was a breakdown in harmony between the pilots and the automation of the aircraft. Modern jets, such as the 777, are largely automated but this does not mean that the pilots have any less control of the aircraft.
A large part of pilot training, both initial and recurrent, is based on the use of automation both in normal and non-normal situations. Contrary to common belief, it’s not simply a question of the pilots pressing a button and putting their feet up until they reach their destination.
The History of Aircraft Automation
Today’s airliners and the airlines that fly them have come a long way since the first transport category aircraft over 100 years ago. Hot off the back of the Wright brothers first flight at Kitty Hawk in 1903, the potential for the commercial use of these flying machines was clear to see.
The first company to use fixed-wing aircraft for commercial service was St. Petersburg – Tampa Airboat Line, who in 1914 first flew the 23 mile trip between the two Floridian towns. From here, commercial air travel began to grow rapidly. Of the big names still in operation today, KLM was founded in 1919, QANTAS in 1920 and Aeroflot in 1923.
The aircraft used at these times were basic and were very much hand-flown by the pilot. That first commercial flight between St. Petersburg and Tampa in 1914 used a Benoist XIV, an open-air biplane. By the time KLM made their first flight, aircraft design had improved somewhat in those few years, launching service with a Fokker F.II.
As technology improved, so did the range of the aircraft and the experience of those who flew them. Despite the first non-stop transatlantic crossing taking place in 1927, it took another decade before the first flight carrying passengers achieved the same feat. Using a Focke-Wulf Fw 200, Lufthansa flew the 3,728mile trip from Berlin to New York in 1938, taking just over 25 hours. This was still very much a hand flown aircraft and would have been incredibly tiring for the team of pilots on board.
However, even in the late 1930s, the idea of an automatic pilot which could help lower the pilot’s workload wasn’t a work of science fiction. It had actually been invented by American Lawrence Sperry just a few years after the Wright brother’s first flight.
Lawrence Sperry – the father of automatic flight
As the son of famous inventor Elmer Sperry, Chicago born Sperry was destined to achieve great things. Along with his brother, he built his own glider at home managing to fly it 500ft in the air and land it safely, despite having no proper training. He is responsible for many modern-day aviation staples such as retractable landing gear, the turn bank indicator and of course the autopilot.
As part of his career, Sperry Snr. had invented a gyroscopic compass for use on battleships. Thinking how this could translate to aviation, Lawrence wondered if gyroscopes could be used to control the pitch and roll of an aircraft in the same way. Depending on the original setting of the gyroscope, any change in direction caused by wind or passenger movement on the aircraft could be corrected back to the original course.
Sometime later, Sperry’s invention was ready to show off to the public at the Concours de la Securité en Aéroplane in Paris in 1914. Here, numerous aviation pioneers were displaying their inventions, all vying for the top prize of 50,000 French Francs.
Along with his assistant Emil Cachin, Sperry made a number of passes in front of the judges in his single-engine, open-air aircraft to demonstrate his automatic pilot, each pass more daring than the last.
As if flying with his hands above his head wasn’t enough, for the final pass of the judges, both he and Chacin climbed out onto the wings, leaving the cockpit completely empty, demonstrating the true ability of his automatic pilot. As stunned faces from the crowd looked on at this incredible feat, it was unsurprising that Sperry won the first prize and achieved instant fame for his invention.
How Today’s Pilots Use Automation
Fast forward 100 years and the autopilot plays a central role in the operation of a modern airliner. On any given flight, the autopilot is engaged for around 98% of the time that the aircraft is airborne. However, the systems in modern aircraft have come a long way from simply keeping the wings level and the nose on the horizon. Nowadays the automatics of the aircraft are referred to as the ‘autoflight’ system and comprise of several different parts.
The Flight Management Computer
The Flight Management Computer (FMC) is the brain of the navigation system. Before each flight, we download the flight plan from the company server and this details the route between the departure point and the destination. We then add a specific departure runway and route and arrival route and runway from the FMC database to complete the flight plan. This will also detail the altitudes and speed that we plan to fly at each stage of the flight.
Any subsequent changes to the route, planned altitudes and speeds are made via the FMC. For example, if our Oceanic Clearance is to cross the pond at 43,000ft, we will enter this into the FMC. It is used primarily to change future segments of the flight path.
Mode Control Panel
Along with the glare shield, between the two pilots, is the Mode Control Panel (MCP). This panel of switches is primarily used to make changes to the immediate flight path. For example, if ATC instructs us to climb to 39,000ft, we use the MCP to change the altitude and climb the aircraft to the new altitude.
The MCP has an interface to control the primary flight parameters of speed, heading and altitude. It also has buttons to engage some of the ‘automatic’ systems fed from the FMC such as lateral and vertical navigation.
The autothrottle is a system that controls the engine thrust to meet the demand for a speed either dictated by the FMC or set in the MCP. Of all the automatic systems, this is the one that is normally engaged for the entirety of the flight on aircraft such as the 777 and 787.
Having the FMC or MCP provide data for flight is of no use unless it is represented to the pilots in a way that is easy to understand. To do this, a set of crosshairs can be displayed on the Primary Flight Display (PFD) in front of both pilots. This directs them whether to pitch the nose up or down and whether to turn the aircraft to the left or right.
Crucially, if for whatever reason the pilots do not agree with the instructions that the flight directors are giving them, they can be switched off. If this is done, the massive 777 just becomes like any other small aircraft. Point the nose up, and the aircraft will climb. Roll the aircraft to the right and it will turn to the right. This is the most basic and fundamental rule of automatic flight.
If the aircraft is not doing what you want it to be doing, turn off the automatics and fly it like a big Cessna.
With the flight directors turned on, the aircraft can still be flown by hand (manually). This is exactly what we do during the initial climb away from the runway on departure and for the last stages of the approach to land. We use the FMC or MCP generated data to provide us with guidance via the flight directors to fly the aircraft away from the ground or down towards the runway.
All the autopilot does is instruct the aircraft to follow the flight director guidance. This is another key point. The autopilot is only as effective as the information which we, the pilots, feed it via the FMC or MCP. If we have programmed either to fly at 3,000ft in a certain direction, the autopilot will do this perfectly – until it flies the aircraft into a hill.
As a result, the pilots are always in control of what the autopilot is doing. Either immediately via the MCP or later on via the FMC.
Why use automatics?
The great benefit of these automatic systems is that once programmed correctly, it gives us the ability to take a step back from the capacity sapping manual flying of the aircraft and allow us to take in the bigger picture of what is going on around us. Flying an aircraft is a complex task. At times it feels like we are operating within a game of 3D chess with one wrong move risking disaster.
Not only must we keep a good mental picture of what is going on with our aircraft but also what other aircraft around us are doing, all whilst communicating with ATC. We also have the comfort and safety of our passengers and crew to bear in mind – and this is all just during a normal flight. Operating the aircraft in a non-normal or emergency situation, the automatics can really help us achieve a safe outcome – but only if used in the right way.
The key to using automation on a modern airliner is to know exactly how the system works. How each different mode will affect the way the aircraft flies. If one system fails, what other system will it revert to? Is the mode that you’re using the best mode for the current situation? If left in that mode, are there any potential pitfalls?
Not only is it the responsibility of the pilot to understand every detail of the automatic systems on their aircraft, airlines and manufacturers also have a responsibility to ensure that pilots are provided with effective training to help them do so.
Ultimately, the golden rule always applies. If the automatics are not doing what you want them to do, turn them off and fly the aircraft as a big Cessna.
The Future of Automation
As technology continues to evolve, one can’t help but wonder how automation will change in the flight deck. The big question here is if automation will ever replace pilots completely.
Last year, Cathay Pacific and Airbus announced that they were looking into the feasibility of single-pilot operations on the A350 aircraft during long-haul flights. This would entail a normal crew of 3 of 4 pilots being reduced to just 2, for flights up to, and maybe over, 17 hours. The thought process is that both pilots will be in the operating seats for take-off and then once airborne, they will then take it in turns to sleep in the rest facilities. They will both then return to the flight deck for landing.
From the outside, this may seem like a good idea. The workload during the cruise is much less than during take-off and landing. Why not use this time for some quality sleep so that they can be rested for landing? However, it’s here that the problem lies. Sleeping on an aircraft is never nearly as good as in your own bed. Modern aircraft do have half-decent rest facilities for pilots and crew as the area is dark and well air-conditioned.
However, the ‘beds’ are little more than a thin mattress on the ground, the bedding can be a thin blanket and the noise from the cabin service still permeates into the rest area. Then there’s turbulence.
Some people say they like the feeling of being rocked to sleep but I can assure you that isn’t for everyone. Numerous times I have either been woken up by being rocked around in the rest facility or unable to get to sleep at all. Regularly I’ll return from my break feeling like I have slept, but don’t feel particularly rested. I’m sure many readers will have had that experience after a bad night in bed at home.
As a result, inflight breaks can not be relied upon to ensure that the pilots are well rested for landing. The extra pilot on longer flights also brings another benefit – an extra pair of eyes and ears at safety-critical moments.
When the crew have been on duty for 17 hours and quite possibly awake for even longer, having a 3rd person in the flight deck to monitor what is going on is of huge benefit to flight safety. It provides another layer of backup to the flight safety net. Should both operating pilots miss something, be it an altitude clearance from ATC or become confused about the proximity to terrain, the 3rd pilot is quite often able to resolve the situation by offering their views.
All critical aircraft systems have backups. For the most important systems, there is a backup for the backup. Why not use the same philosophy for the pilots themselves?
The bigger the threat to our personal safety, the more we want to have a fellow human performing the actions. We all have daily struggles with technology. Our phones won’t send messages, the satellite TV stops working and our laptops give us the “green screen of death.” How many set a “backup” alarm, just in case the original one on your phone doesn’t work when you have to be at work on time for an important meeting?
If the power fails to an automated train, friction and gravity will bring it to a natural, safe stop. If the same were to happen to an aircraft, the same outcome can not be assured.
Until passengers have the confidence to step into a fully automated machine and be carried 7 miles above the ground at 550mph, there will always be the need for a well trained, well-rested team of pilots. How this will evolve remains to be seen. What is clear is that the connection between pilot and aircraft needs to be focused on constantly. Recurrent training needs to equip pilots with the skills they need to know when to let the aircraft automation do its thing and when to step in and take control.
Featured Image – Getty Images
Welcome to The Points Guy!