Opinion: Training New Pilots In The Age Of Automation

pilot at flight deck controls
Credit: David Massey/Embry-Riddle

Twenty years ago, aeronautical science students did not arrive at Embry-Riddle Aeronautical University with hundreds of hours of flight time on a PC-based simulator. Nor did they own unmanned aerial vehicles equipped with remotely operated cameras and autoland control capability. Today’s “internet of wings” continues to advance, bringing new capabilities as well as vulnerabilities to all types of aircraft and, equally importantly, their operators.

As aircraft, computers and automation merge at a rapid pace, my workday priorities remain constant: to help make our students as successful as possible and to provide the aviation industry with critical research and a workforce prepared to innovate and adapt to technological advances in aviation. While COVID-19 has temporarily slowed our industry, we cannot stop preparing for the future of aviation.

In today’s environment, preparing students to become pilots and industry leaders goes beyond stick-and-rudder training. The future will be full of “flying computers.” As commercial airliners and airports become increasingly automated, aviation leaders must keep expanding their awareness of how current human-in-the-loop (HIL) research will revolutionize “business as usual” tomorrow.

Embry-Riddle has always prioritized safety, which has made us early adopters of simulation and virtual and augmented reality—critical tools in advancing our knowledge of human-machine interfaces. Our philosophy is that we do not train pilots; we prepare decision-makers. Our flight-training programs strive to incorporate an understanding of human factors to ensure that pilots of highly automated aircraft can respond swiftly in a crisis. Through our research programs, we continue to explore how aircraft systems can benefit from artificial intelligence and automation without causing a concomitant degradation of pilots’ stick-and-rudder skills. For example, we have ongoing research into whether onboard, real-time diagnostics help pilots maintain situational awareness despite increased automation on the flight deck. 

The convergence in technology drives a convergence of disciplines. As educators, we must integrate expertise from multiple disciplines. A traditional silo mentality restricts our field of view. To get the highest performance from humans and their flying machines requires cross-disciplinary research and education. 

In response to the merging of highly automated aircraft and their human operators, we have challenged our five colleges to create more cross-disciplinary initiatives and shared research, informed by industry advisory boards. We have convened aviation experts focused on flight, design, manufacturing and human factors to advise us in developing robust cyber- and big-data education, for example. We are also undertaking collaborative research with multiple airframe manufacturers—companies with different philosophies on the “allocation of control authority” between pilots and aircraft. 

Our goal is to infuse all of our educational programs with a comprehensive perspective on human and machine performance in aviation. That requires competencies that will transform the industry, such as data analytics, cyber-resilience and advanced manufacturing. We also want graduates of our technical programs to have a solid business foundation so they will recognize how to monetize new opportunities.

This broadening of old boundaries is critical to aviation safety, efficiency and the economy. Designers need to understand pilots so their designs reflect human manual and cognitive skills. Pilots need to understand the aircraft’s architecture and logic so they do not sacrifice situational awareness and become complacent or overly reliant on automation. Scientists need to understand entrepreneurship to speed the evolution of ideas into products.

Driving technological innovation and research in higher education is vital to the future success of our industry. Discovery-driven education is equally important. Providing real-world, applied research opportunities for students will ease their entry into tomorrow’s aerospace workforce and elevate their value to employers. 

Figuring out how best to keep “humans in the loop” as we develop more autonomous and intelligent systems is a defining challenge too complex for any single discipline or institution to address alone. Figuring out how best to keep “students in the loop” so they graduate with a strong foundation backed by a varied skill set is a defining challenge for aviation education.

P. Barry Butler is president of Embry-Riddle Aeronautical University.

P. Barry Butler

P. Barry Butler is president of Embry-Riddle Aeronautical University.

Comments

8 Comments
When we began training Glass Cockpits at FlightSafety's Gulfstream Facility in Savannah, GA we developed and utilized a "Desk Top Trainer" that mimicked the actual cockpit before the students ever advanced into the FTD or simulator. The menus of the Flight Management System, System Synoptic Displays, System Malfunction and the entire flight could all be trained and managed in the classroom. Although we still had "Death by PowerPoint" lessons, the students could create their own scenario while the teaching was in progress. Each student had two large interactive displays, an iPad and their iPhone, along with two very large classroom displays. Every student was in their ADD or ADHD element. The FlightSafety software engineers of our Simulator System's Division were (and still are) fantastic in creating these devices.
This incremental approach to teaching and learning Automation has been very successful in training and checking outstanding corporate pilots.

Tom Owens
GV Series Pilot Examiner (Ret)
FlightSafety International
The number one issue is pilots must be able to fly the plane if automation messes up. AF447 was flown into the ocean because the pilots couldn't fly the aircraft safely.
Everything you had to say in the article about becoming a pilot is good. However, I think you missed what is most important about becoming a pilot. In my pilot training I learned how to recover safely from ALL the 'crazy' flying situations one expect to encounter. I soloed and flew a Piper Cub, T-28, and T-33 to receive my wings. Then flew B-47 aircraft in the US Air Force; T-33, F-84, and F-86H aircraft in Mass. ANG while working at Pratt & Whitney for ~40 years; and finally owning and flying a Cessna 182 aircraft... a total of 50+ years if flying. Simulator training is fine.... but learning to fly a 'real' airplane is essential, in my humble opinion!
I have known two Embry-Riddle graduates. The first had become the RL-10 field service representative at Pratt & Whitney, Florida and developed a training set of resources for a group of F-100 augmented turbofan engine future field representatives to F-16 NATO bases in 1980. The second was my sons room mate at the University of Michigan, North Campus. From an engineering degree at the U of M he was going on to work for a drug company. This was in 1987-89. My own direct relevance to this blurb here was developing the adversary digital pilot for the Navy "Quickturn" and then AIM-95 "AGILE" missile. Now when that missile is known as the AIM-9X and has thrust vectoring by nozzle vanes and a flat panel seeker as shown to me by Raytheon at "Tailhook 2010" I am assured that the digital work was not in vain.
It is not technology that is holding back air transports from only having one pilot in the cockpit. We are already there. However, most likely by the time that single-pilot transport ops would be approved, the industry will just leap to fully autonomous ops.
1
Automation has improved the safety of flight, but one of the tenets of pilot training should be to provide pilots with a foundation of skills both inside and beyond the normal envelope to ensure resilience in the event of failure.
Over reliance on automated systems and lack of basic flying skills or "rusty" flying skills seem to be the primary causes in many accidents. The accident reviewed in this issue is a classic example of reliance on autopilot functions and marginal stick and rudder skills. Amazing that with three pilots in the cockpit the airplane was flown into the ground.