Can A CFIT App Help Pilots Stay Alive?


NASA – which always has a cool setup at Oshkosh, space suits and all – quietly conducted research last month under the radar of the world’s biggest airshow. In the basement of EAA’s Air Academy Lodge, a team of NASA staffers had pilots fly an X-Plane sim to test a cellphone app designed to avoid controlled flight into terrain. (See a video of it here.) Whether or not you think we need more apps in the cockpit, you’ve got to admit this sounds interesting. And NASA’s work in CFIT prevention up to this point is a good example of how a lot of behind-the-scenes data crunching and engineering we never see end up as everyday tools we take for granted.

What’s now called iGCAS (improved Ground Collision Avoidance System) isn’t a new idea for NASA. Since the 1980s, the agency’s Armstrong Flight Research Center and the Air Force have been designing and testing this concept. The result is Automatic Collision-Avoidance Technology for the F-16. The big reduction in accidents for the fighters led to NASA taking the project into the realm of light GA aircraft, including a NASA Cirrus that flies for a variety of studies. Now in the form of a smartphone app, iGCAS can tell researchers if GA pilots would benefit from the technology. With CFIT persisting in its pattern of fatal accidents, this is worth examining. (And AirVenture was a fun, inexpensive way for NASA to bring pilots into the study.)

The app, hooked up with the phone’s display to X-Plane’s navigation data, is simple to use. Green arrows are good; wherever they point indicates safe passage above or past terrain you see from the cockpit. Now, watch the rising hills looming ahead (if you can see them; some CFIT crashes occur at night or in bad visibility). The phone issues a verbal warning: “Terrain near.” Yellow arrows tell you what the viable avoidance options are, as long as you’re timely in your maneuvering. Any direction that’s out of the question no longer has an arrow, just a red X. If you continue approaching the terrain and collision is imminent, the phone flashes more yellow; a voice and a big arrow tell you to apply full throttle and climb. If you do so immediately, you’ll clear the terrain. The format brought to mind angle-of-attack indicators, which use a similar color-coding system to tell the pilot when he needs to do something soon, followed by he needs to do something now.

It would be fascinating to see the data from the show when the researchers sift through each and every move the pilots made during their sessions. Anecdotally, project manager Mark Skoog told me that all the test subjects, regardless of flight experience, were able to react to the app’s indications in time and no one came to a simulated bad end while at the controls. Naturally, they liked this concept; a common follow-up question was: “When can I get this app?” First things first, Skoog says. Among other things, they still have work to do. AirVenture was just a week of recording data; now they’ll go back and analyze how pilots reacted to the visual and aural cues provided by the software. Each pilot also gave personal feedback in an interview; all this will be taken into account. Then they’ll make any necessary improvements and do more testing. Even when they arrive at what would make a nice app, NASA isn’t in that business – that’s for the private sector. After seeing how it worked, it was easy to imagine pilots some day downloading, say, a “Steer Clear” app from GooglePlay and iTunes.

Could such an app reduce the CFIT accident numbers? Like any cockpit tool, certified or not, it must be used properly to reap the benefits, lest it bite you. Technology can never take the place of decent judgment or skill; it was never meant to. Maps, annunciators and gauges have been known to distract pilots from the task of flying, sometimes with tragic results. It seems that the more displays we cram into the panel, the more distracted we are, but that’s for another discussion. The iGCAS study does consider the various human factors to the extent possible, and the test pilots’ interviews will shed a lot of light on that. For example, Skoog notes that eliminating “nuisance warnings” is a big focus of the study, as pilots often ignore signals they believe are erroneous or not useful. And the app was deliberately designed for hand flying, as GA autopilots can’t make abrupt avoidance maneuvers. Bottom line: This is a tool to help pilots stay alive if, for whatever reason, they failed to take preventive measures before that mountain filled the windshield. So to help this tool help them prevent accidents, pilots must treat it for what it is.

If this piqued your interest from a safety or software standpoint, it’d be worth watching NASA’s progress on iGCAS. Software developers might very well want to take this to the marketplace, the real testbed for effectiveness and popularity. The concept could be a viable candidate for integration into certified panels. If anything, it will keep the conversations rolling on CFIT and the digitization of cockpits. I, for one, will keep an eye on it.