Automated & Connected

How has automation transformed aviation?

Over the past year, the aviation industry has consistently been in the news over unfortunate circumstances. One topic that has been highlighted as a potential failure point is automated systems in airplanes, which may negatively impact the development of these systems for future aircraft.  In order to gain further insight into how automated systems have been integrated into planes over the last century, I reached out to a pilot in the industry who has spent years flying for multiple airlines, operating various airplane types from multiple manufacturers. 

When was autopilot first implemented?  What did it control?

Autopilot has been around almost as long as aviation, with the earliest examples dating back to 1914.

During flight, three axes of rotation are used to control the flight path, pitch, roll, and yaw. Pitch controls the nose of the airplane up or down and is controlled by the elevator. Roll is the bank of the aircraft and is how an airplane turns and is controlled by the ailerons. Yaw is controlled by the rudder on the tail and is mainly there to help keep the airplane straight through various maneuvers, such as turning.

The capabilities of the autopilot are determined by how integrated it is to the airplane navigation system. This first system used gyroscopes to maintain a preset tolerance in which the hydraulically powered flight controls (elevator and rudder) would be used to maintain that tolerance set on the gyroscope.

How has autopilot advanced over the years?

In the early days, autopilot was simply based off a gyro holding its position in space. Over the years, features were added to broaden the abilities of an autopilot to help reduce pilot workload and increase the safety of air travel. One feature that was added was altitude hold, which enabled the pilot to select a particular altitude for the airplane to maintain. Another added feature was heading select, where instead of just pointing an airplane in a certain direction and then turning on the autopilot until a correction needed to be made, the pilot could select a particular heading to fly.

Other advancements have included integration with navigation sources. These earlier navigation sources were radio wave-based in which onboard receivers could communicate to the autopilot. The autopilot would then maintain the particular direction the pilot selected along the navigation route. If winds changed along the route, the autopilot would steer the airplane to maintain its course. Radio-based navigation is still widely used today, such as the VHF Omni-direction Range (VOR) for long-range navigation. As GPS has become more prevalent, accurate, and reliable for civilian use, it has made its way into the aviation industry. GPS is the primary source for long-range navigation today.

When the weather at your destination has some low clouds or poor visibility, pilots use what are called approaches to safely navigate to the runway. An autopilot can be coupled to an approach to fly low to the runway. Approaches have a lower tolerance for deviations and have increasingly tighter parameters as the aircraft gets closer to the runway. 

Today’s commercial airliners have highly integrated autopilot systems that do a very good job of flying the route and approaches as designated by the pilot. The Flight Management System (FMS) takes in navigation data, verifies its accuracy, and communicates to the autopilot to fly the correct path. 

What is autopilot like in the latest planes, such as the Airbus A220 and Embraer E2?

These aircraft have the “latest and greatest” autopilot systems because they have the benefit of being the newest models of commercial aircraft to date. These aircraft are able to take advantage of better computing power and larger data storage of the FMS than planes developed in the 1990s and 2000s. The pilot has the ability to fine-tune different aspects, such as the descent from cruise altitude to the approach environment, by making an input to the FMS. It can provide comfort for the passengers by adding pilot-insight to the descent to foresee changes that might occur based on traffic or weather.

What other forms of automation are on modern airplanes?

For airliners with wing-mounted engines, many have a system called Thrust Compensation. The basic premise is that because the engines are mounted beneath the wings, if large amounts of power are introduced, the nose of the aircraft will pitch up without any input from the elevator. So, this system will maintain your current degrees of pitch during power application using automatic inputs through the elevator.

Engine starting has come a long way over the years. Jet engines, such as those used on passenger airliners, use air to start rotation. All airliners have a small jet engine in the tail of the airplane called the Auxiliary Power Unit (APU). This one is small enough to be started by electric power (internal or external) or the battery. Once started, it provides pneumatics (air conditioning) and electrical power. For a main engine to start, we must use pneumatic and electric power from the APU.

On my aircraft, the MD-88, I have to get the engine the ingredients to start and then monitor the start to prevent engine damage. I shut off the air conditioning to have the proper amount of air to spin the engine shown on the pressure gauge in the cockpit. Next, I verify the ignition is on, and then I hold down the starter. The engine begins to spin, and I make sure that the engine gets to the proper RPM, internal engine temperature if it is already warm from a previous flight, oil pressure, and hydraulic pressure. Once these parameters are met, then I move the fuel level to introduce fuel. Temperatures start to rise quickly, and I must shut off the fuel level if it looks like the engine will become too hot. After I start the engine, I re-introduce air conditioning and make sure the engine generator is producing electricity. Then, I repeat the process for engine #2. Airplanes that have been developed in the last 20 years have varying levels of automation with the start, with some to all of the start sequence handled by the engine computer, which includes some engine parameter protections.

The pressurization system is also an automatic system. Instead of pilots flying the airplane and trying to make sure everyone has enough air pressure to breath, engineers decided to automate the pressurization system. It has a built-in schedule to stay within the design pressure limits between the inhospitable low pressure outside the aircraft and the air pressure inside the airplane. Some airplanes need to have the landing field elevation input into the pressurization controller to know when to equalize the pressure between the ambient pressure and that of the airplane, and other controllers get that information from the FMS.

How does autopilot / automation improve your abilities as a pilot?

Autopilot enables pilots to free up their focus to perform other demanding tasks. It really is helpful if there is a medical event with a passenger, mechanical issues, or weather to decide how to get around and formulate contingencies/Plan C or D if weather is rapidly changing or not forecast. (We always have a Plan B.) The autopilot system helps to reduce fatigue through the day. Automation helps us to fly more efficiently by helping us calculate a descent point that allows us to stay higher longer before needing to descend into an airport. We are able to fly faster and burn less fuel the higher we are.

Are there any downsides for increased automation in airplanes?

Automation in airplanes does reduce some mental math skills that pilots have had to perform in years past. It is most prevalent in descent planning. The FMS can calculate better descent points, but we aren’t calculating it out ourselves as much -- although we always verify the descent point to see if it makes sense.

The current level of automation I feel is at a good level; it is enough to enhance flying and efficiency, but not too much in which the pilot is a monitor, not a pilot. People are working on increasing the level of automation where one would be able to “press a button and go,” as some people think is the case now, possibly reducing a flight to have only one pilot and maybe be autonomous one day. Progress is always good to strive for, but flying is one activity in which two pilots that have years of experience using that skill and judgement need to be fully in command of the flight.

Myriad decisions and variables need to be sifted through on a normal flight, let alone when things aren’t going well. Pilots not only deal with medical events, weather, and mechanical issues, but also have to figure out the best way to handle air traffic control delay programs that we learn about after pushing back from the gate. Some questions and thoughts we sift through include: whether we need to shut down the engines to save fuel, where do we then park because it takes a couple minutes to start the engines and we can’t block certain taxiways, passenger comfort is a big deal, how many supplies do we have onboard (water, snacks), should we just go back to the gate, is one available, what is our fuel situation, is there a different route we have to take, do we have enough fuel for that route, how is the weather over the new route, will there be holding in the air and do we have fuel for that, and is our alternate airport still a good option now that the weather has changed?

How can automation in airplanes be improved?

I feel that it is at a good level currently. The aircraft that are coming off the assembly line now have the latest model FMS that complement the pilot nicely. Opportunities may exist to integrate weather models to help with flight planning, but that is all I can think of.

 

The pilot interviewed has elected to remain anonymous; the views expressed are his own.

 

Learn more

Purchase the latest book on the aviation industry, “Fundamentals of Electric Aircraft” by Pascal Thalin.

Visit SAE.org/standards/development for more information on developing aerospace standards at SAE.

Attend AeroTech Europe for more information on the latest commercial vehicle research and advancements.

 

Matthew Borst is a content editor at SAE International in the Global Products Group. Previously, he worked as a technical writer at Polaris Industries and was responsible for writing service manuals for various powersports products. He graduated from Minnesota State University, Mankato with a degree in Automotive Engineering.  His interests include the latest automotive industry news, movies, hockey, and anything that keeps his two kids entertained.

Contact him regarding any article or collaboration ideas by e-mail at matthew.borst@sae.org.