Global strategy envisions training all air carrier pilots in airplane upset prevention and recovery.
By Wayne Rosenkrans | From Orlando
AeroSafety World, June 2012
In a hard-won consensus, about 80 international specialists from 45 organizations have identified “critical” knowledge, skills and attitudes that professional pilots must have to prevent airplane upsets — their primary goal — and to recover from an inadvertent upset. In laying out a rational plan for pilot training, their evidence-based work and expertise have been a stabilizing influence in the wake of high-profile loss of control–in flight (LOC-I) accidents, several representatives told the World Aviation Training Conference and Tradeshow (WATS 2012) in April in Orlando, Florida, U.S.
This work group — the International Committee for Aviation Training in Extended Envelopes (ICATEE) of the Royal Aeronautical Society — currently is completing the last of several near-term deliverables to the global aviation community, specific civil aviation authorities and the air transport industry, according to Bryan Burks, a captain for Alaska Airlines and ICATEE member. He also is the vice chairman of the National Training Council of Air Line Pilots Association, International (ALPA) and ALPA representative to the Royal Aeronautical Society’s Flight Simulation Training Device International Working Group.
“For too long … we [as an industry] assumed that when we hire pilots with an ATP [airline transport pilot] certificate, they will come with the requisite knowledge and skills when it comes to aerodynamics,” Burks said. “ICATEE [identified] a training gap; that that is not the case [see, “Not Worth Being Upset” and “Pilot Project”]. Hopefully, licensing requirements in the future will assure that an ATP license means something more. But in the meantime, the operator should probably [address] that deficit.”
ICATEE has developed a strategy for graduated — that is, one step at a time in a building block approach — implementation of enhanced upset prevention and recovery training (UPRT) that can be supported by the existing pilot training infrastructure.
“Enhanced, integrated UPRT contains three primary elements: academics, on-aircraft training at the [commercial pilot] licensing level [and] the appropriate use of flight simulation training devices [FSTDs],” he said. “[On-aircraft training] would be in an all-attitude, all-envelope, aerobatic-capable aircraft with a trained instructor early in a professional license scheme. … ICATEE also has identified opportunities to enhance FSTDs to provide UPRT.” The on-aircraft element would be a UPRT endorsement to a commercial pilot certificate.
For FSTDs, the work group advocates and recommends enhanced aerodynamic (or aero) models beyond the normal envelope, new/improved tools for feedback to instructors and pilots in post-flight briefing, and significantly improved UPRT motion and buffet cues. “These will happen in the future … in a way that the industry can adopt in an organized fashion — and control the quality and, most importantly, the instructor qualifications,” Burks said. “We’ve made some strong recommendations on how instructors should be qualified for the on-aircraft training aspect and for flight simulators. … [We also advocate] a gradual implementation of these requirements.”
Exposure of pilots to the actual threat environment helps to develop habitual responses to incipient conditions and confidence in their ability to respond correctly to upset situations, said Sunjoo Advani, chairman of ICATEE. “There is no single tool for providing the optimum solution; we must integrate several tools,” he said. “If pilots have the knowledge, if they have the capability, that is one thing. But being put into that threatening environment is very important.”
Specifically, ICATEE concluded that the inadequate training environment has been based on several assumptions, which in turn became limitations to how the industry provided training to prevent upsets and respond to LOC-I. “[The industry] had assumed that the aircraft is in a normal operational envelope in a non-agitated flight condition,” Advani said. “We also had assumed that situational awareness and information can be accurately correlated by the pilot with respect to the observed flight condition. … And we assumed that the handling skills that are taught during licensing are suitable and adequate to resolve the [potential upset] situation.”
In the academic arena, ICATEE members have been updating, augmenting and adapting to current instructional media the Airplane Upset Recovery Training Aid, Revision 2.1 “We wanted to refresh the [training aid in October 2012] by looking at its limitations,” Advani said. “Our new training manual, based on the [training aid], will include sections for pilots, instructors, training providers and regulators [that will be] very usable and user-friendly when implemented into training programs.”
Notably, the training manual also will furnish examples of negative training to help airlines and other simulator training providers anticipate FSTD limitations as they implement UPRT scenarios. Every UPRT event recommended for initial and recurrent pilot training will have a dedicated instructor manual, the presenters said.
A substantial number of the pilot-track session attendees raised their hands when Advani asked if they were familiar with the current training aid and had used it in their training programs. ICATEE also has concentrated on breaking content into parts that are easier to absorb and is seeking to officially incorporate the manual into standards and recommended practices endorsed and/or required by the International Civil Aviation Organization (ICAO).
The idea of conducting UPRT training for the current population of airline pilots in small all-attitude, all-envelope airplanes has been controversial. In the 2012 update on its work, ICATEE has been more specific and realistic about targeting this element to generations of pilots coming into the profession.
“The airplane is really the place where we can provide the psychological component, the physiological component, g-awareness [actual gravitational acceleration] and an accurate recovery environment,” Advani said. “If pilots haven’t been exposed to it, and they encounter an upset — even though it may be rare — they may end up applying the wrong control strategies and make the situation worse. … We realize we cannot take a transport category aircraft and start doing training on upsets. On a voluntary basis, [airlines also could provide UPRT flight training in an all-attitude, all-envelope airplane], and I think that improves pilot skills. However, we have to concentrate on the future. … So we need integration of the use of aerobatic-capable aircraft, qualified [UPRT] instruction [and] appropriate [and] better use of today’s FSTDs.”
The industry can enhance feedback through use of better instructional tools and information in FSTDs. “In the future, we can look toward improving simulation fidelity through better aero models and … feedback tools, such as informing the pilots where they are with respect to the validated envelope,” Advani said. “If they exceed the envelope, the instructor should have the ability to tell the pilot, ‘We have gone beyond the bounds of what is known.’
“We also want to see if the pilots have exceeded the structural limitations of the [real] aircraft [because] incorrect control inputs can be devastating.” Methods tested as effective include displaying color-coded aerodynamic diagrams in the instructor operating station alongside replays of the pilot’s control inputs with animation software. The instructional tools described have been designed to provide instructors more accurate situational awareness and a “powerful new way of providing UPRT feedback to pilots while avoiding negative training,” he said.
The list of ICATEE deliverables comprises tasks being accomplished, several with 2012 target dates. So far, ICATEE has presented its recommendations to the U.S. Federal Aviation Administration (FAA), the FAA-Industry Stall–Stick Pusher Working Group,2 the Adverse Weather Working Group and the Loss of Control Aviation Rulemaking Advisory Committee.
Proposed language was delivered in January in an executive level recommendation to ICAO for an amendment stating that UPRT “shall be conducted.” “In [ICAO’s] Procedures for Air Navigation–Training document, there will be references [submitted in October 2011] that refer to our training manual, which is scheduled for delivery to ICAO later this year, as well as the UPRT component for simulator documents such as [ICAO Doc 9625],” Advani said.3 Anticipated products include a report in mid-2012 from ICATEE’s research and technology group to the Royal Aeronautical Society and a revision to the International Air Transport Association’s FSTD data document.
Toward FSTD Stall Realism
ICATEE now considers the prospects of expanding the aero model used in simulators to be favorable for several reasons and will continue to pursue that objective, Burks said. He cited recent demonstration by Boeing Commercial Airplanes of a prototype enhanced aero model that would help commercial aviation to conduct aerodynamic stall training.
“[Today’s model] is very good up to approach to stall, to the critical angle-of-attack,” he said. “After that, there are [not enough] flight test data from the manufacturers that provide a good model to do training in the device. … An aerodynamic stall — for a swept-wing, transport category jet — is a place pilots don’t want to be. Unfortunately, if a crew gets to the aerodynamic stall in most simulators today, it is a very benign representation. It does not look very much different than the approach to stall. So if they haven’t actually stalled an airplane since they were in a Cessna 152 25 years ago, pilots have this false or benign sense of the aircraft performance. In an approach to stall, the aircraft still has airflow attached to the wings, and it is still somewhat controllable; it is in a decayed state, it has less margin, but it is controllable.”
Advani noted, “What we have to teach is not the actual flight dynamics in that stall — how to fly in that region — but how to immediately recognize [the situation] and recover. The most important thing [is] how to avoid it and, if [they go] there, to get out as quickly as possible. … We’re looking at how we can incorporate today’s high-fidelity models that go up to the top of the [lift curve slope] with, perhaps, lower fidelity or representative models that simply teach the skills necessary for the recovery from upsets.”
From experience supporting military FSTDs, aircraft manufacturers have a wealth of knowledge and can deliver very accurate engineering data, Burks said. “We are excited because they’re going to bring that capability into the civil market now,” he said. “The bottom line is we hope to have a good platform to introduce aerodynamic stall training to pilots and show them the marked difference between approach to stall and aerodynamic stall. This is going to enhance pilots’ … upset prevention through avoidance, recognition and awareness.”
UPRT is not a one-time inoculation. “These are perishable skill sets,” Burks said. “[At Alaska Airlines,] we believe that we need to revisit [UPRT] on an annual basis. After the skill sets are developed, we want to measure the effectiveness of the prevention strategies. So, eventually, after we gain exposure and develop the skill sets in the maneuver-based training, we want to validate that training by giving our pilots these events in a true surprise scenario.” The objective is to apply prevention skills, not recovery skills, in those events.
One desirable type of display for instructor operating stations shows that the airplane is being operated outside of a green-bounded area that represents the certified airplane flight envelope on an alpha beta graph (i.e., angle-of-attack on the Y axis and angle of side slip on the X axis). “If the flight crew exceeds those bounds, the instructor should get a flag, or give a flag to the pilot, and say, ‘Look, you were in an area [of the flight envelope] that could have caused problems, so we have to be very careful.’”
ICATEE also has focused on flagging negative training, such as the inappropriate use of controls. “We want to provide situational awareness through the instruments,” Advani said.
Alaska Airlines last year chose “managing angle-of-attack” as the theme for its 2012 UPRT training event, implemented under the flexibility the FAA allows airlines operating with an advanced qualification program for pilots. The company asked internal and external airplane upset specialists to validate the event’s content, method and learning value. The event also included approach to stall, which is within the valid aerodynamic model of the full flight simulators, Burks said.
This 30-minute simulator training event positions a Boeing 737-800 flight crew in level flight at 15,000 ft with autopilot and autothrottle disengaged and the flight path vector displayed. The captain has an angle-of-attack indicator on the head-up display (HUD), and the first officer does not have a HUD but can compare the flight path vector with the pitch attitude of the aircraft for an indirect indication of angle-of-attack on the primary flight display.
As pilot flying, the captain “rolls into a bank that starts off at about 3 degrees angle-of-attack, and as he rolls into the bank, maintaining altitude, the angle-of-attack increases as he uses the vertical and lateral components of lift,” he said. As the first officer reads out the increasing angle-of-attack, the captain exactly correlates that data on the angle-of-attack indicator on the HUD. Another demonstration involves managing angle-of-attack using the speed tape at up to 60 degrees of bank.
“At that point, they are right in the buffer bars on the speed tape, [and] we have the pilots demonstrate a rolling pullout,” Burks said. “We want them to recover from the 60-degree bank while holding altitude by using aileron and spoiler. But as they do that, they can see that … they are still in [the buffer bars], and it is not until they roll out slowly through the maneuver that they are out of that threat area. Then we have them demonstrate [managing] angle-of-attack by the only direct control the pilot has: the control column and elevator. They’re in that bank, they’re in the buffet, and as soon as they reduce angle-of-attack with the control yoke — boom! — the speed tape gives them almost instant validation of the effectiveness of managing angle-of-attack.”
Later in the session, the instructor pilot slews the airplane to 37,000 ft for a high-altitude repetition of elements of the previous manual flying scenario. “We want to embed this [immediate response of reducing angle-of-attack] by practical exposure … to an upset at altitude,” Burks said. “We felt it was necessary for pilots to have a little bit of understanding of how the aircraft actually flies and feels at 37,000 ft. … We turned off the yaw damper at 15,000 ft and at 37,000 ft so that they could see some of the dynamic instabilities, the lack of damping qualities.”
At 37,000 ft, the flight crew is flying “very close to the buffer bars and not too far away from intermittent stick shaker, the red bar at the bottom [of the speed tape],” he said. “We impress upon them that they are in a thrust-limited condition — with less than 10 percent of the thrust they have at sea level.”
The session also includes approach-to-stall scenarios with the airplane flying behind the power curve at the two altitudes. “We have them get into the buffer bar because as angle-of-attack increases, drag increases, [and] the pilot runs out of excess thrust,” Burks said. Operating at high altitude at the back of the power curve in the approach to stall, a very small amount of bank is required to trigger the stall warning, and airspeed continues to degrade despite the selection of maximum continuous thrust, he said.
Out of Bounds
One time, a 737 training captain expressed pride that while exploring the boundaries of simulator fidelity, company pilots “recovered fine” from being rolled 180 degrees with a 40-degree pitch toward the ground, according to another presenter affiliated with ICATEE. Paul Kolisch, a captain; a supervisor, flight operations training, Pinnacle Airlines; and member of the FAA-Industry Stall–Stick Pusher Working Group recalled, “Well, maybe they did in that simulator, but there is no guarantee that that is going to happen in an airplane. So, when they are giving training, [instructor pilots] need to operate within reasonable bounds of the envelope for which the simulator is designed — typically, 25 degrees up, 10 degrees down and 50 to 60 degrees [of bank] left and right. If pilots start wandering around at 110 degrees [of bank], there is no telling if they are getting an adequate response or not. You may be just misleading your pilots.”
In addition to incorporating startle into UPRT scenarios in FSTDs, several other factors enhance skill retention and confidence, he added. “The reason we need to do good stall training for recognition and recovery is that we have not been able to figure out how to teach pilots to never make a mistake,” Kolisch said. “I was pleased to see [a presentation] that said [regulators and airlines] should not mandate a predetermined loss of altitude [in UPRT], particularly if [pilots] practice these at high altitude. If you are operating a jet, and you don’t do stall training at the maximum operational altitude, you are not doing the training. … If you aren’t practicing at practical altitudes [such as approach to landing], you’re not training. And if you’re not doing it on the autopilot, you’re not training.”
- A supplement providing guidance on operations, unintentional slowdowns and recoveries in the high-altitude environment was added during the October 2008 revision of the training aid, first released in 1998. “While [previous versions] specifically addressed airplanes with 100 seats or greater, the information in this supplement is directly applicable to most jet airplanes that routinely operate in this environment,” the 2008 version said.
- A public comment period ended Jan. 12, 2012, for FAA draft Advisory Circular 120-STALL titled “Stall and Stick Pusher Training.” ICATEE was among organizations that helped develop the 39-page draft, which contains issue background information; stall and stick pusher training philosophy; ground school/academic training; concepts for building demonstrations, scenarios and recovery techniques; a stall-recovery template; and appendixes with sample FSTD demonstrations, sample FSTD training event scenarios and FSTD considerations.
- ICAO. Manual of Criteria for the Qualification of Flight Simulation Training Devices (Doc 9625), “Volume 1 — Aeroplanes,” 3rd edition, 2009. This edition currently is undergoing revision with input from ICATEE.