Automation and the pilot in the cockpit

With new technology allowing drones into US airspace, manned aircraft do still have a future.

By Marvin Cetron
Pres, Forecasting Intl

Singapore Airlines Deputy Chief Pilot A340/A380 Training B K Chin in the cockpit of an Airbus A380 at SIN (Changi, Singapore) could represent an endangered species. ATC systems now under development at NASA could ground human pilots to serve as backup for computers.

One question many of us have been pondering is how long it will be before technology allows bizjets to carry out all aspects of flight-from takeoff to landing-on their own, on a routine basis. Obviously, any such development will have implications for corporate pilots.

A state-of-the-art drone can take off, fly to its destination, and land without a human hand on the controls. A Northrop Grumman RQ4 Global Hawk observation drone leaving San Francisco CA could fly to Maine with up to 3000 lbs of internal payload, loiter for 24 hours while surveying an area 230 miles square, fly home again and touch down safely.

The $35-million aircraft is not completely autonomous. It requires a ground-based "launch and recovery element" to provide differential GPS information for takeoff and landing. Yet the Global Hawk makes it clear that the day of pilotless aircraft has arrived.

Thus far, drones are banned from flight in congested airspace. The National Oceanic and Atmospheric Administration (NOAA) is eager to launch small, remotely controlled aircraft from south Florida to monitor hurricanes.

However, at least for the 2008 storm season, NOAA has been forced to fly them out of Barbados. Even with a remote human pilot, FAA is concerned that remote piloted vehicles (RPVs) could endanger other aircraft because they are incapable of the traditional "see-and-avoid" approach to air safety.

A fully autonomous airplane would be an even greater concern. Yet many airliners and Part 135 aircraft already come close to the RQ4's near-independence. When using the autoland function, pilots select the approach and runway, operate flaps and landing gear, and perhaps deploy the reversers. Everything else is handled by the machine.

Although no Cat IIIc (zero ceiling, zero visibility) ILS approaches have been certified yet, today's autoland systems could handle them. Eliminate those last few human operations, and we could consider this the gold standard for passenger-carrying aircraft autonomy.

USAF Northrop Grumman RQ4 Global Hawk can fly 3000 miles, survey nearly 140,000 square miles for 24 hours, and fly home without human control.

Airbus has gone a step further by taking certain control decisions away from the pilot altogether. The company's fly-by-wire (FBW) models are hard-limited to prevent pitching up into a stall or more than 15degrees nose-down, banking more than 67degrees, or putting more than 2.5 G acceleration on the airplane.

Yank-and-bank an A320 to avoid a potential midair, and the control computers automatically apply full thrust, retract spoilers, and limit max climbout to 30degrees. The pilot is still more than an advisor, but the computers have final authority over what the plane will do.

Boeing, in contrast, allows the pilot to override the automated controls, so that the plane's maximum performance is available when it's needed. Which approach is better is a matter of considerable debate in the air transport and aircraft design communities.

Yet the principal is clear. If we could automate the last few operations left to the pilot in a Cat IIIc landing, today's passenger-carrying aircraft would differ from the Global Hawk only in their size and cargo.

If technology were the only requirement, it would not be long before avionics turned even the left-seater into a backup system, needed only on those rare occasions when the hardware goes belly-up.

Yet technological possibility is only one factor to consider in making a forecast-the others are economic feasibility and social acceptability. And when it comes to automating passenger-carrying aircraft, some of those concerns could be a lot harder to satisfy.

NextGen nears autonomy

Today's autoland systems have some significant drawbacks, especially for corporate aviation. As avionics specialist Glenn Connor, president of Discover Technologies Intl, points out, they are expensive to install and use.

Cyborg insects from DARPA's HI-MEMS project would turn moths into GPS-guided autonomous surveillance drones.

It can cost up to $3.5 million to equip a runway end for Cat III ILS use-as a result, there are only about 60 Cat III approaches in the US and 250 in the entire world. (Australia lacks any Category III approaches and bans the use of autoland except in declared emergencies.)

Pilots and hardware require recertification every 90 days, which adds to the cost of operation. And all this is to prepare for just 1.6% of the landings in the US and between 2.2 and 2.6% of the landings in Europe. "That kind of investment makes sense for the airlines, where the plane is going back and forth between the same few airports," says Connor.

"It doesn't make sense for bizjets. Corporate fleets seldom use even Cat II approaches-the only ones I know of who use Cat III are a few French companies." FAA's answer to improved flight operations is the Next Generation Air Transportation System (NextGen), which is being rolled out now and is scheduled to spread throughout the world by 2025.

While NextGen involves many new technologies and procedures, area navigation (RNAV) and required navigation performance (RNP) are the most critical for pilots. Area navigation starts with the familiar RNAV track line computer, which creates an artificial waypoint offset from a VORTAC, allowing planes to fly a straight line to their destination, rather than zigzagging from beacon to beacon.

Added to this, and perhaps eventually replacing it, are 2 relatively new technologies-WAAS and LAAS. Wide area augmentation system (WAAS) provides GPS-based horizontal and vertical navigation for departures, enroute and arrivals.

Fully internal, it frees the aircraft from VOR/DME and other conventional navigation aids. In fact, for most pilots GPS has already replaced VORTAC for enroute navigation. Even for the approach, GPS is as accurate as Cat I ILS.

FAA has published WAAS approaches for several major airports, with more in the pipeline. This offers to provide approaches to the 200-ft decision point (DH) for thousands of airports where ILS is unavailable.



1 | 2| 3| Next