Aerobatics versus Upset Prevention and Recovery Training

What’s the Difference, and Why is it Important?

Every year Boeing releases a statistical summary of worldwide commercial jet accidents covering the preceding ten years. For several years now, Loss of Control In-flight (LOC-I) has led controlled flight into terrain as the number one cause of both accidents and fatalities[1]. The trend is similar with regard to business aircraft[2]. Clearly, LOC-I is a major flight safety issue. Despite this fact there are currently no defined standards for the delivery of upset recovery training. Now that we understand why this conversation is important, let’s consider some ways we can alleviate the problem.

Original Article in CAT Magazine

There are two important avenues to reducing this hazard: technology and training. With respect to technology, the most promising area is the envelope protection provided by fly-by-wire (FBW) flight controls. Data shows that there is a significantly lower rate of LOC-I among fleets of FBW aircraft[3]. In non-FBW aircraft, advances in autopilot technology now offer some elements of envelope protection previously seen only in FBW aircraft. However, across the design spectrum there are examples of LOC-I accidents and incidents involving failures in flight control system defenses.

Despite these and future hi-tech advances, thousands of classically controlled aircraft without envelope protection will be carrying the majority of the world’s passengers for decades into the future. The last line of defense against upsets will continue to depend, as it does today, on the awareness and capabilities of pilots.

As we consider the training side of the equation, we will begin with simulators since that is where the vast majority of pilots receive their operational training. Flight simulators are unquestionably the greatest tool for aviation safety ever devised. Unfortunately, standard full motion simulators have three limiting areas with regard to providing initial Upset Prevention & Recovery Training (UPRT). First, the prolonged forces associated with the accelerations experienced in flight during dynamic maneuvering cannot be provided.

Secondly, simulators are only as good as the data they are programmed with. Currently, that data ends well before the boundaries of what an aircraft could encounter in a recoverable upset situation. It simply is not safe for a transport category aircraft to intentionally gather flight test data in the areas of the envelope that might be unintentionally encountered in an upset. Training in simulators past the limits of valid data invites inappropriate pilot response based on unrealistic simulator behavior[4].

The third area is not a technology related limitation. Even if a simulator could provide accurate cockpit g (some can), and had perfect aerodynamic fidelity, at some level the pilot would still be aware they were in, well…a simulator. It is not unusual to see pilots emerge sweat-soaked from the dark recesses of the simulator after facing the various gremlins and demons that they must be prepared to face in flight.  It is not that simulators cannot induce nervousness, surprise, anxiety, and even fear, but only to a degree.

This psycho/physiological dimension is not a trivial aspect of the upset recovery dynamic. A pilot may know with perfect academic clarity the correct response or flight control inputs to make while in the classroom, but unfortunately that knowledge often does not translate into the immediate, proportionate control inputs (often quite different from those used in normal operations) that may be required in an emergency. The practiced ability to suppress the startle response is best achieved through skill development in an environment providing all the inputs encountered in flight, including the perception of risk and the threat of consequences that are present in an actual upset event.

If a pilot encounters an engine failure in flight, or has to deal with a system failure, they have probably seen it and dealt with it in the simulator first. They have familiarity with the situation because they have virtually been there. Not so in the case of an upset event, where a pilot may be experiencing an attitude in an aircraft that they have never seen before, real or simulated.

In a study of several LOC-I accidents, “Defining Commercial Transport Loss-of-Control: A Quantitative Approach”[5], it was shown that it takes on average less than 10 seconds for accident aircraft to progress from an upset condition to loss of control.  With less than 10 seconds to fix a time critical problem that could rapidly become life threatening, pilots must respond in a nearly instinctive manner. Expecting pilots to safely accomplish perilous tasks that they have not been trained for, on their first attempt, is not realistic. Viewed in this light, the current LOC-I accident record is not surprising.

This inability to fully leverage existing flight simulators in the delivery of UPRT is truly unfortunate. With current flight simulation training devices we are unable to provide pilots with training in the domain that accident data shows they need it the most; in the skills required for safe recovery from in-flight upsets. If the simulator cannot provide the environment to practice in, and such maneuvering cannot be accomplished in the transport aircraft in which such a situation could be encountered, an appropriately capable surrogate training aircraft becomes the best option.

Now that we are discussing the use of an aircraft as our training resource, let’s get back to the original question: what is the difference between aerobatics and UPRT? Part of the confusion over aerobatics and UPRT comes from the fact that there are elements that they both share. Both types of training encompass operation through the entire range of possible attitudes and cover the entire flight envelope, including those portions never encountered in normal flight operations. This all-attitude/all-envelope (A3E) training is essential to prepare pilots for unexpected upsets, but the application of that training through UPRT can look very different from the familiar maneuvers seen at airshows.

Classic aerobatics can provide pilots with improved capabilities in precision maneuvering and aircraft handling skills. While the enhanced understanding and capabilities that pilots can receive from aerobatic training has been proven to be beneficial in situations that require upset recovery skills[6], that is not the primary objective of aerobatic training; it is a secondary benefit. Unlike aerobatics, the entire focus of UPRT is the recognition and avoidance of situations that increase the probability of an upset event, and in providing the skills necessary for recovery if it becomes required. This leads to an entirely different approach to training.

While academics associated with learning aerobatics are generally limited to what is necessary for the completion of a particular maneuver, for UPRT academic fundamentals provide the informed basis for the instinctive understanding required in a time critical upset event. An aerobatic maneuver like an aileron roll might be used in teaching UPRT principles. In UPRT, however, the focus is not on the performance of a precise maneuver, as is the case in aerobatics, but instead on the ability to rapidly orient the aircraft’s lift vector in order to minimize altitude loss. While a full flight simulator provides little or no help in teaching aerobatics, a comprehensive UPRT program can utilize a simulator, within its valid training envelope, to apply the principles introduced in a surrogate aircraft to the specific aircraft systems and avionics of a transport aircraft. A fundamental precept of UPRT is that the most appropriate resources should be used to administer the required elements of training.

The table below helps to differentiate between these two methods of providing A3E flight training.


A fundamental distinction between aerobatics and UPRT is that while most aerobatic instruction only teaches mastery of the specific aircraft type used in training, in UPRT the airframe used for training is a stand-in or proxy platform for introducing concepts representative of the recovery techniques required for a wide range of aircraft.  Much like basic instrument skills, which can be applied to flying a vast array of aircraft, the majority of flight skills and techniques required for upset recovery are not aircraft specific. Such phenomena as lateral control instability at high angles of attack, lift vector orientation, and accelerated stalls apply to all fixed wing aircraft regardless of size or performance.

Just as basic instrument skills learned in lighter and lower performing aircraft are applied to more advanced aircraft, introducing basic upset recovery techniques early in a pilot’s training provides lessons that will remain with pilots throughout their entire career.

There is some skepticism that a light aerobatic aircraft can teach anyone anything about flying a swept-winged jet transport. The flight training of a true UPRT program strives to remain within the structural envelope of the non-aerobatic subject aircraft. For a transport category aircraft that would be a 2.5 g limit load and a 3.75 g ultimate load. Likewise, while it is unsafe to flirt with the area at or beyond stall in an aircraft not certified for spins, that threat is not present in an appropriate training aircraft recoverable from autorotation. The excess structural capability and high angle of attack/spin recovery ability of the surrogate training aircraft merely provides a safety margin for the exceedances of the trainee; a safety margin not present in the transport category aircraft in which an upset could be encountered or in a non-aerobatic training aircraft.

It is not the type of training aircraft platform used that matters. What is important are the lessons the aircraft is used to deliver and how they are conveyed. Unfortunately, today too many of these important lessons are not being delivered at all.

While the primary focus of a comprehensive UPRT program is the avoidance and safe recovery from upsets, secondary benefits in the development of manual handling skills, confidence, and airmanship are significant.


We would not conceive of putting pilots on the flight deck unprepared to deal with engine failures, or failures of fuel, electrical, hydraulic, or pressurization systems, yet we continue to inadequately prepare pilots to face the real threat of LOC-I. Accident data shows that among commercial jet transports worldwide there are more accidents and fatalities from the category of LOC-I than from all powerplant or other systems related accidents combined. Introduction of pilots to the fundamentals of all-attitude/all-envelope flight through a comprehensive program involving underlying  academic principles, practical skill development in appropriate surrogate aircraft, and aircraft specific characteristics in a full flight simulator provides the most thorough approach to the reduction of the LOC-I accident rate.


The author would like to thank Keoki Gray and Ed Longoria for their contributions to this article.


About the author:

Randall Brooks is the Senior Director of Flight Training for Opinicus Corporation and serves as the President of the Upset Prevention and Recovery Training Association (UPRTA). Mr. Brooks is the Training Matrix subgroup leader for the International Committee for Aviation Training in Extended Envelopes (ICATEE) and before joining Opinicus, was the Director of Customer Training and Manager of Emergency Situation Training for the Eclipse Aircraft Corporation. Mr. Brooks has provided all-attitude/all-envelope flight instruction for over 25 years in more than 15 different jet, piston, and sailplane aircraft types. Mr. Brooks holds a B.S. in Aerospace Engineering from the University of Colorado.

[1] “Statistical Summary of Commercial Jet Airplane Accidents-Worldwide Operations, 2002-2009”, Airplane Safety

Engineering, p. 23, Boeing Commercial Airplane Group, Seattle, WA, USA, 2010

[2] Veillette, Patrick,”The Why and When Behind Loss of Control”, Aviation Week & Space Technology, May 6, 2009

[3] Bateman, Don, “Some Thoughts on Reducing the Risk of Aircraft Loss of Control”, p.5, EASS Flight Safety Paper, March 2011

[4] “In-Flight Separation of Vertical Stabilizer, American Airlines Flight 587, Airbus Industrie A300-605R, N14053, Belle Harbor, New York”, Aircraft Accident Report NTSB/AAR-04/04, p.157, NTSB, 2004

[5] Wilborn, J.E. and Foster, J.V., “Defining Commercial Transport Loss-of-Control: A Quantitative Approach”, p. 9,

American Institute of Aeronautics and Astronautics Atmospheric Flight Mechanics Conference and Exhibition,

Providence, Rhode Island, 2004.

[6] Gawron, Valerie, NASA Airplane Upset Training Evaluation Report, Veridian Engineering, Buffalo, NY, 2002.

[7] “Statistical Summary of Commercial Jet Airplane Accidents-Worldwide Operations, 2002-2009”, Airplane Safety

Engineering, p. 23, Boeing Commercial Airplane Group, Seattle, WA, USA, 2010


This entry was posted in Key Messages, Training, UPRT Resources. Bookmark the permalink.

Leave a Reply

Your email address will not be published. Required fields are marked *


You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>