Slashdot Mirror
Designing Tomorrow's Air Traffic Control Systems
aarondubrow writes According to FAA estimates, increasing congestion in the air transportation system of the United States, if unaddressed, will cost the American economy $22 billion annually in lost economic activity by 2022. MIT researcher Hamsa Balakrishnan and her team are making air traffic control systems more efficient through a combination of better models and new embedded technologies. Testing their algorithms at Logan Airport in Boston, they showed that by holding aircraft back for 4.5 minutes, they could improve flow on the runways and save nearly 100 pounds of fuel for each aircraft.
Make sure the tower doesn't have software written by a company that went out of business, but still managed to get an update that can bring the whole thing down, and employ versioning in your datacenter backups...but if that doesn't work, ensure that copies of the software are uploaded to the planes themselves so that, in the event there's a group of socially awkward geniuses that can drive a Ferrari down a landing strip, they can download the software via an Ethernet cable and save everyone.
anybody have a non-PAYWALL link to the Sciencedirect article?
This article is about Airport Surface Traffic Control, ie: having 20 planes lined up on the taxi-way to take off is not efficient. Instead, delay aircraft pushback from the gate so wait time on the taxiway is minimized (hence saving fuel).
building in some redundant backups so that what happened in Chicago last month doesn't happen again?
No thank you...
“He’s not deformed, he’s just drunk!”
Instead of having only a few super-airports that all airlines use for connecting flights, why not reduce traffic to those airports by having more hubs?
improve flow on the runways and save nearly 100 pounds of fuel for each aircraft.
100 pounds of fuel, or 100 pounds of fuel?
systemd is Roko's Basilisk.
Does this include economic activity lost for the FAA being too shortsighted or too belligerent to have a workable permit system for commercial use of drones in 2014?
1st this was done 4 years ago. The future is now.
But mostly...
Why were none of the tests more than 4 hours long? What happens after the test period, do they need to recover or something?
If this is so efficient why are we talking about a 4 year old test instead of the implementation 3 years ago?!?
The Economist has a great article on free-routing. Not only does this save time & fuel, but a "continuous descent approach" is also quieter at airports.
When the choice us between saving 100 lbs of fuel (possibly savings for some other airline) or declaring an on-time departure because you've pulled back from the gate, which do you think the airlines are going to choose? Hint: they have to pay flight crew from the time the door closes today and they have been choosing to pay their employees more rather than sit longer at the gate.
If the FAA was a private corporation this would never be an issue.
They’ve been spending money on an Operational continuity plan (OCP) for 13 years.
The FAA can only spend money on paperwork, not on productive products.
FAA garbage doc 1
FAA garbage doc 2
Do a keyword search for OCP or BCP to see whats going on.
That's because Boston isn't utilizing decent surface management software. Busier airports like Atlanta and JFK do, so if you want to do a real study, study an airport that has already addressed this issue. http://www.airtrafficmanagemen... https://www.youtube.com/watch?...
Isn't this one of Larry Page's Google Y's ambitions also?
Great, now we are going to have ramp metering for airplanes.
People often forget that the first A in NASA stands for aeronautics. They have been hard at work at fully computerizing air traffic control. However as you can imagine this requires lots of testing given the potentially fatal consequences. NASA has held several competition rounds among contractors for the next air traffic control system, providing feedback to all candidates for the next round.
Here is one such paper. There are many others, from various academia and industry consortia. The work they are doing is rather cool.
I cannot find much detail on this, but it sounds suspiciously like well known techniques for avoiding congestion in complex systems that I learned in queuing theory over 40 years ago.
This story frustrates the hell out of me. Once again, the government refuses to realize that the entire air traffic control system is out of date and no longer needed. Instead of improving their algorithms, they should be dismantling all the ARTCCs and TRACONs, leaving only the tower flowers and ground controllers in the major airports. Before the Slashdot community gets it collective panties in collective bunches, please realize this is a simple problem of government seizing control over something at a time when nobody had any better ideas and refusing to give control back to the people after ATC was no longer needed.
In 1956, a United Airlines DC-7 and a Trans World Airlines L-1049 Super Connie collided over the Grand Canyon. Back in the 50's the skies were nearly empty and such a disaster was inconceivable. Obviously, something needed to be done to prevent repeat disasters. At the time there was no techology that gave pilots the ability to see other planes through bad weather or over great distances. So, the government stepped in and blanketed the entire country with radar stations and Air Route Traffic Control Centers (ARTCC). This system has been in place ever since. As a result, you cannot fly ANYWHERE without the government knowing about it and controlling your every move.
Some decades passed and the Traffic Collision Avoidance System (TCAS) was developed along with weather radar, both of which are required on every high-altitude and commercial plane. In fact, so much time has passed that TCAS is becoming outdated in favor of TCAS II, TCAS III, and TCAS IV. Additionally, groundwork is being laid to create an entirely new air traffic avoidance system based on the existing AIS and S-AIS systems.
After working out some cultural and training issues involving the proper prioritizing of TCAS and ATC instructions (en.wikipedia.org/wiki/Überlingen_mid-air_collision), I don't believe there's been a single mid-air collision between aircraft equipped with TCAS where both TCAS devices were operating.
The simple fact of the matter is that pilots are now equipped with airborne technology that lets them see and avoid each other. We don't need an air traffic control system or government control over every flight any longer. In fact, in Crandall v. Nevada, 73 U.S. 35 (1868) the Supreme Court ruled it is illegal and unconstitutional for the government to demand to know our whereabouts and our travel plans or to otherwise interfere with our freedom to travel as we please.
What's needed is a minimum set of "rules of the road" that determine which aircraft must yield right of way in any situation, and the FAA has already established these. In fact, they are currently coded into every TCAS unit. The government's active participation in and control over every flight should be eliminated in favor of the role we need government for the most: that of service, acting as referee and law enforcement agency. This is exactly how marine vessels and vehicles on our roads operate.
One of the more dangerous aspects of the ATC system as it exists now is our system of "highways in the sky". The sectional maps pilots use to navigate use a web of straight lines drawn between radio navigation beacons. ATC clearances always involve travelling along a route that traverses these "airways". Instead of allowing airplanes to fly wherever they want, the net effect of the FAA's airway system is to gather the airplanes together into these airways, bringing them closer together and greatly increasing the number of potential collisions. Once ATC is removed, planes will be spread out all over the sky, reducing overall fuel consumption and reducing the potential for collisions.
We need to resist these kinds of power plays the government always perpetrates on us. I have to admit that with such a left-leaning, narcissistic, big government President, the odds of seeing the right thing happen here are extremely low.
The US government should sell the air traffic control system to the highest bidder. That would probably be a consortium of the major airports and airlines. Any system run by bureaucratic management will necessarily be way behind the times technologically. The air traffic control system needs to be operated as a business, not a bureaucracy.
Ideally, the airlines themselves would pay for the service.
Air traffic control is the nanny state at its worst. Planes should be able to fly anywhere they goddam like, dancing a fandango if they want. If they crash into another plane then it's a simple matter of the lawyers working out whose fault it is. That will directly hit the bottom line and THE MARKET will sort it out like it does everything else.
--
roman_mir
That is all.
"If any question why we died, Tell them because our fathers lied."
I'm not talking AI here, just how difficult can it be to have the computer program do all this real time? There cant be more than a few thousand planes in the air at any time, on known vectors and in 3D space, what am I missing?
www.boznz.com Simple solutions to complex problems.
Yeah...
20+ years ago they "Upgraded" to new computerized systems.
Ran into a guy that used to talk about how his father worked on the new software and how F'ing retarded and horrifying the FAA bureaucracy is. Seems they had been designing a great system that would virtually eliminate the chances of screw ups and had to do a major re-write and make it a mess.
Why?
The FAA has a HUGE training system in place for the old system. They weren't willing to upgrade the training program, so the "New" software system had to work the same way the old crappy equipment worked. Serious design flaws and all.
Wow...
Right now, if I want to fly from LAX to JFK, I need to wait for ATC to slot me so I’ll likely be able to land without delay upon reaching JFK. Why do I still have to talk to ATC to do this? Why are humans even involved when computers could do this instantaneously? I should just be able to file my flight plan from a laptop or smartphone, and the system tells ME when I need to depart (i.e. I get an email stating “depart RWY 19R 1900Z to 1905Z). This would make ground control’s job a hell of a lot easier. Any aircraft not contacting ground twenty minutes before their assigned departure time goes to the back of the line.
Furthermore, anything in the air, whether a helicopter, drone, or 1950s taildragger should be required to have Mode S. If my TCAS is interrogating everything in a 360 degree sphere around me, and feeding that to my MFD and autopilot, there’s very little chance any of us will ever run into each other, and I don’t need to worry about visibility when flying VFR, or ATC while flying in IMC.
This is part of NextGen, a scheduled overhaul of FAA operations that has ran over time and financial budgets and could cost up to $150 billion by 2022. The idea is to reduce congestion by moving from ground-based radar to GPS-based technologies that increase the ground towers' and aircraft's awareness of each other.
It continues to see modest implementations with technologies like ADS-out and ADS-in, more advanced ground radar systems and improved flight paths, but has a long way to go. The FAA needs to take a stronger leadership role, and convince airliners to use these new paths. You might recognize JetBlue and Southwest's advertising campaigns for the use of special wingtips that increase fuel efficiency. This is all part of NextGen.