Presentation to the First Intelligent Systems Technical Conference

Chicago, Illinois, September 20-23, 2004.

By Jim Hall

Slide 1: Opening the Skies

For centuries, man sought to fly in a machine. Now, a century after the first flight at Kitty Hawk, I can't help but see the irony in our discussion of finding ways to fly a machine without man.

NASA has been successfully flying uninhabited vehicles for years, but with the benefit of operating in deep space. Let's face it, there isn't a great likelihood of the Voyager 2 probe having to maneuver to avoid an errant Cessna.

Slide 2: Blank Slate

But we are moving into a period of rapid expansion in the use of uninhabited aerial vehicles, a period that will challenge our existing air traffic systems which are geared toward the transit and safety of manned aircraft.

I believe it is important that the industry take a proactive approach, seeking to develop regulations governing unmanned aircraft before a serious incident involving a UAV occurs. If, God forbid, there were to be a serious accident, there would be calls for strict regulation of UAVs. Those calls would be backed with political pressure should there be a significant loss of life.

At the moment, I do not believe the NTSB has investigated any UAV incidents, despite the occurrence of many non-fatal accidents involving these craft. It has yet to develop a specific plan for dealing with such accidents and would likely handle UAVs on a case-by-base basis.

Slide 3: Opportunity Knocks

This presents an opportunity for UAV researchers and industry leaders.

By acting now, the industry has an opportunity to work in concert with the NTSB, FAA, NASA and DOD in a deliberate, organized manner to develop regulations that ensure safety, while not compromising innovation. NASA's Access 5 project is moving us in that direction, and I believe it is critical for this overall effort to come to a successful conclusion before a serious UAV incident occurs.

Developing an appropriate regulatory framework is also important for potential customers of and investors in UAV technology, as uncertainty over where, when and how UAVs will be allowed to enter the national airspace is not conducive to continued development.

And given current restrictions on UAV flights, it can take days, weeks or even months of planning before a UAV flight is approved, dramatically reducing their utility. We must move toward file-and-fly for UAV operators before they will achieve full commercial viability.

Slide 4: Safety First

This cannot be achieved, however, without demonstrating that UAVs can be operated safely in the normal air traffic environment.

Currently, even the most basic manned aircraft carries the most advanced computer system in the world: the human brain. While fly-by-wire has been able to surpass the brain in some areas, turning otherwise unstable aircraft into nimble flyers, humans are still better able than computers to react to a wider diversity of situations.

Given this, and given the overall awareness handicaps presented to even the best ground-based UAV pilot, I believe there are some basic elements that should be a part of any requirements placed on UAVs to better ensure safety.

Slide 5: Safety First -- Recommendations

First, UAVs should be required to carry flight data recorders to provide controllers with data that can be used to identify problems before they cause serious malfunctions, or, to be used by investigators should an accident occur. Alternatively, for smaller UAVs, equivalent data from them should be recorded remotely at the ground control station. This is an area where NTSB and FAA should work together to develop flight data recorder requirements.

Second, UAVs should be required to be equipped with modern transponders and use special UAV transponder codes to allow for easy identification. This is important for safety, but it is also critical in the post-9/11 environment where any unidentified aircraft is viewed as a potential security threat, with fighters often launched to intercept the potential hostile. The last thing you want to have happen to your million-dollar unmanned aircraft is for it to be shot down by a multimillion-dollar manned fighter aircraft.

In conjunction with a transponder, I submit that UAVs should be equipped with modern traffic alert and collision avoidance systems and automatic collision-avoidance flight control functions to prevent mid-air collisions between UAVs and other transponder-equipped aircraft. This is of critical importance when UAVs are operating in the proximity of other aircraft. The cameras on even the best UAVs do not offer the controller the overall vision available to the pilot of a manned aircraft, making automatic collision-avoidance essential.

Along those same lines, non-military or law enforcement UAVs should be easy to spot visually, with bright colors on the exterior being used to draw attention. Now, I admit, outside of my home state of Tennessee, orange and white aren't quite as popular, but bright, visible colors would help them be spotted by pilots of manned aircraft and by other operators.

Finally, the possibility of loss of ground control must be considered, and appropriate fail-safe systems should be in place to ensure the safe recovery of the vehicle, or, if that is not possible, the bringing down of the vehicle in a manner that does not harm people or property on the ground. For some UAVs it is conceivable that the fail-safe could take the form of a simple parachute that is activated after it powers down due to loss of signal. This wouldn't be practical on some larger UAVs, however, necessitating "fly-home" software or other emergency procedures that the vehicle could implement automatically if the control signal is lost. This area presents perhaps one of the greatest challenges, given that it places the UAV into an uncertain situation and forces it to make decisions without the benefit of its human controller.

Slide 6: Innovative Cooperation

It is possible to integrate uninhabited aircraft into our national airspace while still preserving our current level of safety, but it will require innovation, cooperation and a truly collaborative effort before any adverse incidents occur.

Who knows? Maybe in another century, our successors will be talking about how to get humans back into aircraft.

Monday, September 27, 2004