Return to Back Issues main page

Return to Current Issue's Home Page

NASA’s FutureFlight Central uses virtual reality to simulate airport environments so designers can test new configurations without having to build anything

By Danek Kaus

At Dallas/Fort Worth Airport (DFW) in Texas, flight controllers are testing a new set of perimeter taxiways to see how they would improve airliner traffic. They hope the new taxiways will prevent incursions of taxiing planes into the path of those taking off and landing on the runways. The controllers are just getting used to the new taxiways, and the chance of error runs higher than normal. A mishap could prove deadly.

But in this case, it doesn’t matter because the taxiways are not there. Neither are the airport or the planes. The flight controllers are real and they do work at DFW, but on this day they toil in a dark room in Silicon Valley, testing a proposed taxiway at FutureFlight Central (FFC) at the NASA Ames Research Center in Mountain View, California.

According to Mike Madson, deputy facility manager and project manager, FutureFlight’s mission is to provide a world class simulation research facility to improve the safety, efficiency, and cost effectiveness of airport procedures, designs and technologies. To that end, they have created the world’s first full-scale simulated airport control tower with all the equipment controllers would normally use.

To depict many major airports around the country, the center has 3-D databases created from Digital Terrain Elevation Data, airport AutoCAD files, high resolution aerial survey photography, and close-up digital photos. FFC also has a 3-D library of over 100 different aircraft, including airline colors and insignia. Each 360-degree database depicts every runway, taxiway, building, gate, ground vehicle and surrounding terrain. A simulation can include up to 250 planes, all doing different things at the same time.

The FFC tower has 12 wraparound “windows” looking out onto the airport that are actually images created from rear screen projectors. Not only do controllers see their airport, they look at it from the same perspective they would in real life. For example, if the height of the tower gives them a 30 degree line-of-site to a runway, the simulation recreates that. “It’s like you’re really at that particular airport, looking out of the control tower,” Madson says.

Another key element of the simulation is that everything happens in real time. Aircraft taxi, take off, and land, all controlled by pseudo-pilots, former pilots and controllers, who sit at displays in another room. Each pseudo-pilot is responsible for a section of the airport and usually controls several aircraft at once. When a plane leaves that pilot’s area of responsibility, controllers pass the plane on to the pilot of the adjoining section. Each pseudo-pilot can see all the aircraft. and communicate with other pilots and the control tower, using the same frequencies that each airport uses. “Everybody is involved in the simulation at the same time,” Madson says. “The pilots and controllers are interacting like they would in real life. It’s really fantastic.”

With this scenario, FFC offers the chance to test support tools for controllers in real time in a controlled environment. Testing them in a real control tower, with planes landing and taking off, can jeopardize safety and efficiency because controllers are distracted by the new equipment while still trying to move traffic. This means the equipment may not get a fair assessment because the controllers are too busy to use many of its features. But in a real time simulation, controllers can concentrate more on evaluating and using new technologies. “If someone happens to make a mistake, no one is hurt,” Madson says. They can better assess the technology and provide feedback to the developer about which aspects were useful and which were not.

FFC can collect measurements such as statistics on taxi time, departure rates, runway occupancy time, and non-movement area time. They can also make digital audio recordings of communications between pilots and controllers for analysis. Video cameras monitor controllers to study such human elements as how many times a controller looks up and down and other physical actions. The simulations and audio and video files can be played back for study and training. As another benefit, findings from simulations can allow airport officials to reduce aircraft taxi times, which can save millions of dollars in jet fuel.

Airport Traffic Grows

In managing ever increasing traffic, airports know they need to expand or modify to accommodate more aircraft safely and efficiently. However, just adding more runways or taxiways, thereby creating new traffic patterns, can create more problems than it solves. In many cases, airports have used Fast Time Simulation software to see the possible affects of changes. “But this software lacks the human element,” Madson says.

DFW Airport had considered building new taxiways because almost every arrival has to cross a runway. Each such crossing is a potential runway incursion, which also has a ripple effect of slowing down departures and arrivals. They had studied this issue for almost 12 years when they decided to have FFC help them test the idea. “The airport had wanted it, but they preferred to have the concurrence of the controllers and pilots to get funding,” Madson says. “We were able to provide the experience. They were uniformly excited and enthused about the concept and felt it would improve both safety and efficiency.” The numbers also bore it out. Total taxi time was reduced, as was communication levels between pilots and controllers, which reduces workloads.

Los Angeles International Airport (LAX) has also studied an expansion plan. They looked at adding a taxiway between two parallel runways on the south side of the airport as a way to reduce or even eliminate runway incursions. The FFC team adapted the LAX database to add the proposed taxiway. Then LAX controllers spent three days at the simulation tower studying possible procedures in real time, looking at various ways of managing the runways. “They walked away with good ideas from a safety and efficiency standpoint,” Madson says. “You can’t do this in a real airport without laying down concrete.”

When San Francisco International Airport opened a new ramp control tower, sort of a mini tower near the gate, they sent their new ramp operators to FCC for training. “The safety of conducting training in a real time simulation is very attractive. Trainees are going to make mistakes. This way they can relax and focus on learning. If a situation is shaping up, the trainer can freeze the scene and point out how to change it in real life,” Madson says.

A Team Effort

Madson joined FFC as a simulation engineer. While graphics specialists create the airplanes and airport, the simulation engineers give the aircraft the ability to move. This is done by identifying discreet points and connecting each point with others. “Basically, it’s like connecting the dots,” Madson says.

The simulation engineer also creates the scripts that controllers and pseudo-pilots use for each simulation, including what they will say to each other and which gates and runways they will use. The sim engineer will insert the various scenarios into simulations clients want to test. “You have to build the exercise you are going to run. That includes setting up the stream of arrivals, having departures at the gates and have some aircraft being spread around as part of the problem,” Madson says.

Test Engineers Claudine Herbelin and Chris Murphy make sure all the systems are working and interacting with each other properly before and after each simulation. During the simulation, they monitor the event, with Murphy in the pseudo-pilots room. In another location, Herbelin watches her monitors and interjects problems such as storms or gate closures. “We can create whatever events the customer wants,” Herbelin says.

Boris Rabin, visualization development lead, joined FFC when the project was still in the design stage. His background is in modeling, animation, and graphics, and he has experience working with architects. He currently works on every part of the visual aspects of the system, including creating the databases. “From the beginning it was a challenging task because the facility was supposed to have a unique set of capabilities. It had to accommodate up to 10 controllers and allow them work comfortably at the same time,” Rabin says.

Another challenge was creating realistic images as seen by the controllers in such a way as to not distract them, so they could focus on the simulations. FFC also had to put together a system powerful enough to handle the complex geography of each airport with lots of moving, realistic models. “The system had to handle multiple photo textures in real time, and the resolution had to be as high as possible,” Rabin says.

It was a complex puzzle which required the best optical screens that could transfer light to fit the configuration of the tower. FFC couldn’t use projectors inside the tower because they would generate heat and noise. so they opted for rear projection systems that use folded mirrors.

A critical component was the image generators, which would have to handle 360 degrees in real time, which meant a multi-channel system. “The question was how many channels this generator had to be capable of displaying. On the one hand, the more channels, the smoother the transition. On the other hand, fewer channels would be less expensive,” Rabin says. The choice ranged from six to 16 channels. They opted for 12 channels, powered by six SGI “Reality Monsters.”

To make the simulation experience as realistic as possible, they installed separators between the 12 “windows” of the tower to recreate partially obstructed views. “The controllers have to move around to see around them, so we had to recreate that,” Rabin says.

A key requirement was to provide full visualization of the air traffic without any jerky movements or artifacts, so FFC decided to have the images move at 30 frames per second, the standard for TV, Rabin says. “If it was less than 30, say 10, you would see all kinds of jerky movements, which would destroy the realism of the installation,” he says.

Looking to the Future

The FFC crew sees many more applications for their simulation capabilities. For starters, they have begun to explore space applications. They used photographs from the Spirit and Opportunity rovers to create 360-degree 3-D panoramas of the Martian surface. They are also considering developing “synthetic vision” for low-vision situations such as night or fog. The technology could also help bolster security, as airport towers are especially vulnerable to terrorist attack.

Rabin enjoys his work at FFC and looks forward to such new roles for the simulator. “It’s a very exciting facility. We deal with real world projects.” Madson says an entire new virtual airport has been proposed that could be built to determine design and efficiency issues. This would epitomize the effectiveness of simulating airport environments. “You can solve design flaws before they lay the concrete. You can avoid potential expensive mistakes.”

For more information on FutureFlight Central, visit http://ffc.arc.nasa.gov

A freelance writer and publicist in San Jose, California, Danek Kaus is co-author of the book "Power Persuasion," due out this fall. Contact him at dkaus@sbcglobal.net

Progressive Engineer
Editor: Tom Gibson
2049 Crossroads Drive, Lewisburg, PA 17837
570-568-8444 * progress@jdweb.com
©2004 Progressive Engineer