Pilot Panic Buttons in Future Cockpits?

The Paris Air Show, which ended Sunday, typically finds aircraft manufacturers and suppliers introducing some of the new ideas and accompanying technologies that they have been working on. As you may remember, a few weeks ago Airbus discussed its Airbus 2050 concept aircraft that could possibly have a transparent fuselage and virtual reality play areas for passengers.

Concepts and technology for increasing cockpit safety have been a major emphasis this year at the Show. One reason, of course, has been the number of high-profile aircraft crashes the past few years, including the January 2009 ditching of US Airways Flight 1549 into the Hudson River in New York City; the February 2009 crash of Colgan Air Flight 3407 in Buffalo, New York, and; the July 2009 crash of Air France Flight 447 off the coast of Brazil. In each case, the interaction of the aircraft's automated flight management system with the pilots has come under scrutiny, especially in the latter two incidents where the pilots involved, when they found themselves in trouble, apparently took actions that in retrospect may have been questionable .

There were two nice stories in last week's Wall Street Journal examining what the airline industry is doing to try to increase safety by increasing pilot situational awareness in the cockpit. The first article looked at the renewed push by avionics manufacturers like Honeywell and Rockwell Collins to get more large commercial airlines to install the latest cockpit avionics that are already on business and some regional jets. These new cockpit avionics have advanced displays, the WSJ says, that allow pilots:

"... to see accurate replicas of specific mountains, man-made obstacles and other terrain features, even if they are flying at night or if the potential hazards are obscured by clouds..."

"On the ground, cockpit-safety experts are devising mapping systems that enable pilots - without relying on air-traffic controllers - to know the precise position of their planes on the tarmac and to receive alerts about collision threats posed by other aircraft. Other systems are intended to warn cockpit crews if they are lining up to takeoff from the wrong runway, and to issue alerts if planes are approaching a strip too fast or will land too far down to stop safely."

Business aircraft tend to have more advanced cockpit avionics than commercial carriers have because owners of the former turn over their aircraft every 4 to 5 years, while the latter only every 12 to 15 years. In addition, the WSJ notes, business aircraft tend to fly into more out-of-the-way places that have limited ground navigation systems, and having cockpit avionics that increase pilot situational awareness makes economic justification from a safety perspective easier.

The other story in the Journal concerns a new - and apparently controversial - avionics system capability that Rockwell Collins is developing. According to the Journal, Collins is developing a "digital parachute" that will take control of the aircraft's flight management system and return the aircraft to level flight (if possible) when a pilot hits a "panic button" during an emergency situation.

Collins demonstrated the concept - which was an outgrowth of its work on unmanned systems - back in 2009. According to the WSJ article, Collins is now ready to introduce the concept in the next year or two into its advanced "Pro Line" family of avionics for business jets, and the company hints that its digital parachute capability will appear in certain airliners in the near future, although the WSJ says the company won't say which one(s).

Already in some Airbus A320-family aircraft outfitted with Rockwell Collins avionics, if there is a cabin depressurization and the pilots down act quick enough, the flight management system will command a sudden decent, the WSJ says.

The Journal quotes Dr. David Vos, senior director of Rockwell Collins Control Technologies and UAS, as saying:

"This level of automation, redundancy and ability to recover will become ubiquitous...It's a lot closer than many people are willing to recognize."

And according to the Journal article, Dr. Vos might have added the words, "... or accept."

The WSJ says that the Rockwell Collins digital parachute idea has been met with skepticism or outright dismissal in some aviation and flight safety quarters. According to the Journal:

"The skepticism stems partly from the fact that cockpit automation, no matter how sophisticated, sometimes doesn't work precisely as anticipated. When pilots don't understand what computers are doing, 'automation can be the source for many difficulties,' cautions Patrick Goudou, Europe's chief aviation-safety regulator."

I must admit I find this view a bit puzzling. If I understand it correctly, a pilot would be pushing the panic button only in situations where he or she doesn't understand what the heck is happening, which may be because the automated flight management system has become "unintelligible" to the pilot. Unless I am missing something, I fail to see how pushing a panic button is going to make a by definition dangerous situation much worse.

According to this article, the Collins system might have benefited the pilots in the Colgan air crash, although it would not have helped in the US Airway Flight 1549 situation. I don't know whether it would have helped in the Air France Flight 447 crash.

Collins does put a caveat on its digital parachute use, however. As the WSJ describes it:

"The proposed systems won't prevent crashes when planes have lost lift or end up flying dangerously slowly close to the ground, Mr. Vos said. Rather, the aim is to have automation kick in to ensure that aircraft at higher altitudes can escape peril when pilots lose awareness of their situation or end up in deadly stalls or flight upsets, the most common categories of accidents for today's advanced jets."

In other words, when the automation paradox may start to kick in.

Although not quite the same thing as a panic button, last month IBM filed a patent for "smart" traffic lights that could remotely stop a vehicle say from going through a train crossing when a train is approaching by turning off the vehicle's engine, says this blog post at the LA Times. This technology, if ever put into place, could reduce the number of train-vehicle collisions, of which there are some 5,800 each year in the US alone. How it would have fared in the fatal Amtrak-truck collision in Nevada last Friday, I don't know.

And speaking of Nevada, last week the state legislature passed Assembly Bill Number 511 (PDF) setting up the framework for making autonomous cars legal. It will likely be years before fully autonomous cars are driving Nevada highways, but cars equipped with a "temporary autopilot" capability as demonstrated by Volkswagen may be hitting the road before you know it.

Advertisement

Risk Factor

IEEE Spectrum's risk analysis blog, featuring daily news, updates and analysis on computing and IT projects, software and systems failures, successes and innovations, security threats, and more.

Contributor
Willie D. Jones
 
Advertisement