Summer 2014 | VOLUME 56, NO. 2
Initial Lessons Learned from the MH370 Disappearance How Air Traffic Control Factors into the Current Situation
Plus
• Computational Red Teaming and Big Data • Enhanced Low Visibility Operations • Awareness of Distractions in ATC Operational Quarters
www.atca.org
Contents
ANTON BALAHZ / SHUTTERSTOCK
Summer 2014 | Vol. 56, No. 2
ATCA members and subscribers have access to the online edition of The Journal of Air Traffic Control. Visit lesterfiles.com/ pubs/ATCA. Password: ATCAPubs (case sensitive).
Published for: Air Traffic Control Association 1101 King Street, Suite 300 Alexandria, VA 22314 Phone: 703-299-2430 Fax: 703-299-2437 info@atca.org www.atca.org Published by:
140 Broadway, 46th Floor New York, NY 10005 Toll-free phone: 866-953-2189 Toll-free fax: 877-565-8557 www.lesterpublications.com President, Jeff Lester Vice-President & Publisher, Sean Davis EDITORIAL Editorial Director, Jill Harris Managing Editor, Kristy Rydz Editorial Assistant, Andrew Harris DESIGN & LAYOUT Art Director, Myles O’Reilly Senior Graphic Designer, John Lyttle Graphic Designer, Crystal Carrette Graphic Designer, Jessica Landry Graphic Designer, Gayl Punzalan ADVERTISING Sales Director, Danny Macaluso | 866-954-8168 Quinn Bogusky | 888-953-2198 Louise Peterson | 866-953-2183 Walter Lytwyn | 866-953-2196 DISTRIBUTION Jennifer Holmes | 866-953-2189
© 2014 Air Traffic Control Association, Inc. All rights reserved. The contents of this publication may not be reproduced by any means, in whole or in part, without the prior written consent of the ATCA. Disclaimer: The opinions expressed by the authors of the editorial articles contained in this publication are those of the respective authors and do not necessarily represent the opinion of the ATCA. Printed in Canada. Please recycle where facilities exist.
Cover image by Gwoeii.
Cover Story
36
Initial Lessons Learned from the MH370 Disappearance
By Steve Winter
How Air Traffic Control Factors into the Current Situation
Articles
12
The Computational Air Traffic Control Brain
By Hussein A. Abbass, Jiagniun Tang, Rubai Amin, Mohammed Ellejmi, and Stephen Kirby
19
24
29
40
Computational Red Teaming and Big Data for Real-Time Seamless Brain-Traffic Integration
Low Cost, High Impact
Enhanced Low Visibility Operations By David Hughes
Foundations of Professionalism
A Proactive and Collaborative Effort to Bring Awareness of Distractions in the Air Traffic Control Operational Quarters By Patrick Forrey
The Next After NextGen
The Continuous Process of Updates to Infrastructure and Technology By Gary Church
Department of Defense Domestic & International Engagement in the Aviation Enterprise How the PBFA is Working to Improve the National Airspace System
By David M. Heron
46
NextGen for All
By John Dobriansky
54
Taking it to the Next Level
Integration of UAVs into Air Traffic Systems How UAVs are being Brought into the National Airspace By GP Capt Parvez Mahmood
Departments 3 From the President 5 From the Editor’s Desk 9 Member Benefits
59
Directory of Member Organizations
60
Index to Advertisers
& Application
In the Spring 2014 Journal issue, Edward L. Bolton’s title was referenced incorrectly. He is the Assistant Administrator for NextGen, FAA. In the Winter 2013 issue, “Avoiding Turbulence” by Sharman, Trier, and Lane was missing several paragraphs of text. The article is posted in its entirety at www.atca.org/journal. The Journal of Air Traffic Control
1
FPO - Hewlett packard
FROM THE PRESIDENT
PHOTOGRAPHY: MO_SES PREMIUM / SHUTTERSTOCK.COM
The Value of Technology MY LAST MESSAGE FOR THE JOURNAL of Air Traffic Control was written in preparation for World ATM Congress (WATMC) in Madrid in March. That event was a resounding success – meeting and exceeding our expectations with a record-breaking 6,265 attendees. Truly a global representation of the aviation industry, attendees traveled from 120 countries (listed at worldatmcongress. org/2014-Visiting-Countries) and represented Air Navigation Service Providers (ANSPs), academia, airlines, and industry. I was pleasantly surprised to see so many ATCA members in Madrid. We had some very productive discussions. I also received valuable feedback on our planning and execution of WATMC as well as future ATCA events in the U.S. After returning from Madrid, I testified at a U.S. Congressional field hearing before the Committee on Transportation and Infrastructure, Subcommittee on Aviation. The topic was “Moving NextGen Forward: Leveraging the Assets of the William J. Hughes Technical Center.”
By Peter F. Dumont, President & CEO, ATCA
With ATCA’s annual Technical Symposium in Atlantic City, this invitation to testify could not have come at a better time. Our association is a long-standing proponent of the Technical Center and the aviation progress being made in Atlantic City. I was able to not only stress the value to aviation and NextGen the Tech Center provides, but also to explain ATCA’s role and mission in the process. If you think about leveraging the assets of the Tech Center, the first thing that comes to mind is the benefit of the Center having one of every piece of equipment currently deployed and operational in the National Airspace System (NAS). This seems logical, but when you consider all the different components that make up today’s ATC system, the value becomes readily apparent. In my testimony, I encouraged all senior FAA management to visit the Tech Center to understand its full capabilities. However, with the approaching ATCA Tech Symposium in May, all attendees – industry, government, academia, and more – will have this opportunity. Day one of the event – “Tech
Center Tuesday” – returns this year with exhibits and exclusive tours for attendees. I trust it will raise the awareness of the Tech Center’s resources and give professionals a chance to leverage its assets. Not many in aviation have readily available access to the facility, so take advantage of it now. The rest of the week boasts diverse and effective leaders from both industry and government. The FAA’s Edward L. Bolton, Assistant Administrator for NextGen, will participate in an interview-style presentation with the audience, an interactive conversation attendees always enjoy. A full schedule of events is available at atca.org/AgendaACY. Enjoy this issue of The Journal. I hope it spurs you on to increased knowledge and thought in continually contributing to the advancement of air traffic. And I look forward to seeing you in Atlantic City, where we will do the same.
Peter F. Dumont, President & CEO Air Traffic Control Association
The Journal of Air Traffic Control
3
FROM THE EDITOR’S DESK
ATCA
Air Traffic Control Association
Summer 2014 | Vol. 56, No.2 Air Traffic Control Association 1101 King Street, Suite 300 Alexandria, VA 22314 Phone: 703-299-2430 Fax: 703-299-2437 info@atca.org Air Traffic Control Association www.atca.org
By Steve Carver Editor-in-Chief, The Journal of Air Traffic Control
ATCA Formed in 1956 as a non-profit, professional membership association, ATCA represents the interests of all professionals in the air traffic control industry. Dedicated to the advancement of professionalism and technology of air traffic control, ATCA has grown to represent several thousand individuals and organizations managing and providing ATC services and equipment around the world. Editor-in-Chief: Steve Carver Publisher: Lester Publications, LLC
Officers and Board of Directors Chairman: James H. Washington, B3 Solutions Chairman-Elect: Neil Planzer, The Boeing Company President & CEO: Peter F. Dumont, Air Traffic Control Association Treasurer, Director-At-Large: Rachel Jackson, ASRC Research & Technology Solutions Secretary, East Area Director: Jeff Griffith, Washington Consulting Group North Central Area Director: Bill Ellis, Midwest ATC Northeast Area Director: Mike Headley, Apptis South Central Area Director: William Cotton Southeast Area Director: Robert Coulson, Harris Corporation Western Area Director: Mike Lewis, Jeppesen Canada, Caribbean, Central and South America, Mexico Area Director: John Crichton, NAV CANADA Europe, Africa, Middle East Area Director: Steve James Pacific, Asia, Australia Area Director: Bob Gardiner, ACMAT Consultants Directors-At-Large: Rick Day, CSC Charlie Keegan, Raytheon Sandra Samuel, Lockheed Martin
A New Era of Aviation AS I WRITE THIS EDITORIAL, THERE is no news on the location of Malaysian flight 370. The aviation community has been experiencing a period of uncertainty over the past weeks. A 777 is gone and with it, her passengers and crew. The primary mission of the profession we have chosen is to provide safe passage, but once again it has shown vulnerabilities. As I look out my window over Washington Reagan Airport, it seems that taxing aircraft are rotating slower than usual, paying tribute to one of their family members. Let’s remember those on flight 370 and their loved ones. We can hope the “black box” will be found and the world will learn what took this plane out of the air. We must know. With knowledge comes truth and hopefully the mitigation that will eliminate any fault. This issue of The Journal features an insightful and well-sourced article by Steve Winter covering the lessons learned from the MH370 investigation. Regardless of the unknown factors that
could be discovered and the flight’s final outcome, these tenants hold true and are highly applicable to future similar incidents. During the search for flight 370, many points of data were required to try and get the search close to the vicinity of the plane. It is interesting and a large attribution to the FAA that they are moving quickly toward providing the aviation industry, academia, and other stakeholders data on the National Airspace System (NAS). Of course, challenges continue around the FAA policy 1200.22 C, External Request for National Airspace System Data, or commonly called the “information release policy.” Outside the challenges of releasing aircraft positional information, the far reaching effect of providing NAS data for R&D, safety evaluation, search and rescue, and even to the general public where applications could be created to assist in future operational ideas, has far-reaching advantages and challenges. As an example, throughout the world, people Continued on next page
Staff Marion Brophy, Director, Communications Ken Carlisle, Director, Meetings and Expositions Jonathan Fath, Manager, New Media Christine Oster, Chief Financial Officer Paul Planzer, Manager, ATC Programs Claire Rusk, Vice President of Operations Mindy Soranno, Office Manager Rugger Smith, Senior Account Manager Sandra Strickland, Events and Exhibits Coordinator Tim Wagner, Membership Manager
The Journal of Air Traffic Control (ISSN 0021-8650) is published quarterly by the Air Traffic Control Association, Inc. Periodical postage paid at Alexandria, VA and additional entries. EDITORIAL, SUBSCRIPTION & ADVERTISING OFFICES at ATCA Headquarters: 1101 King Street, Suite 300, Alexandria, Virginia 22314. Telephone: (703) 299-2430, Fax: (703) 299-2437, Email: info@atca.org, Website: www.atca. org. POSTMASTER: Send address changes to The Journal of Air Traffic Control, 1101 King Street, Suite 300, Alexandria, Virginia 22314. © Air Traffic Control Association, Inc., 2014 Membership in the Air Traffic Control Association including subscriptions to the Journal and ATCA Bulletin: Professional, $130 a year; Professional Military Senior Enlisted (E6–E9) Officer, $130 a year; Professional Military Junior Enlisted (E1–E5), $26 a year; Retired fee $60 a year applies to those who are ATCA Members at the time of retirement; Corporate Member, $500–5,000 a year, depending on category. Journal subscription rates to non-members: U.S., its territories, and possessions—$78 a year; other countries, including Canada and Mexico—$88 a year (via air mail). Back issue single copy $10, other countries, including Canada and Mexico, $15 (via air mail). Contributors express their personal points of view and opinions that are not necessarily those of their employers or the Air Traffic Control Association. Therefore The Journal of Air Traffic Control does not assume responsibility for statements made and opinions expressed. It does accept responsibility for giving contributors an opportunity to express such views and opinions. Articles may be edited as necessary without changing their meaning.
The Journal of Air Traffic Control
5
have been asked to help search for Flight 370. Using the Tomnod website, (www.tomnod.com/nod), individuals have scanned satellite pictures for wreckage, and volunteers have become part of the search effort. Who knows what the future may hold? Of course, sharing of NAS data has its potential challenges. Think about the pressure placed on the FAA if the public develops reasonable analyses showing a safety concern. How would you handle such input? We are moving into another new era in aviation with the access to aviation big data. Be prepared and keep your options open; things are on the move.
Steve Carver, Editor-in-Chief, The Journal of Air Traffic Control
PHOTOGRAPHY: AHHMED FAISEL YAHYA / SHUTTERSTOCK.COM
FROM THE EDITOR’S DESK
C
C
M B
When the industry’s biggest names rely on you, it’s easy to pull ahead. With the ability to detect aircraft and targets at all levels while avoiding wind turbine clutter, innovative LightWave Radar from C Speed is taking radar to places it has never been before. The technology is the first to provide air traffic controllers with a clear picture over wind farms – making it a big breakthrough for the industry from one powerful little company. For more information, visit www.lightwaveradar.com
6
Summer 2014
A
POWERING TODAY’S FAA MISSION
WITH TOMORROW’S NEXTGEN TECHNOLOGY Cloud. Cybersecurity.
As a leader in next-generation technologies, CSC draws on our commercial best practices and proven solutions to deliver scalable, more cost-effective IT solutions to government agencies seeking efficiency and value through innovation. With pride, CSC has steadfastly supported FAA’s critical
Mobility. Big Data Analytics.
mission for more than 40 years. FAA’s trust in our expertise and domain knowledge strengthens our unwavering commitment to its ongoing success.
And More.
From legacy ATC systems to the infusion of disruptive technology and new operational concepts, CSC continues to deliver innovative solutions supporting the modernization of the National Airspace System. Our current focus is the integration of our operator-centric solutions within a safe and secure cloud-based SOA infrastructure. We are also conducting research into future operational improvements and applying innovation to prepare the aviation stakeholders for upcoming challenges. FAA and CSC: A partnership that has stood the test of time welcomes the challenges of tomorrow — by preparing for them today. csc.com/government
Committed to Providing Proven, Cost-Efficient
Training Solutions for the
FAA
CACI is a leading provider of a wide range of aviation and aerospace offerings, with core competencies in: n
n n
Enterprise training Advanced simulation Remote and classroom education
With experience managing and delivering training services to over 40 government organizations in the United States and around the world, CACI assures the FAA of high-quality, cost-efficient solutions that produce results.
For more information:
Suzan Zimmerman Senior Vice President szimmerman@caci.com 703-725-7671 www.caci.com
CACI is a sponsor of the 2014 ATCA Tech Symposium
A Fortune World’s Most Admired Company I N F O R M AT I O N D E P L O Y E D . Š CACI 2014 A360_1404
S O L U T I O N S A D VA N C E D .
MISSIONS ACCOMPLISHED.
Referred by:___________________
Phone: __________________________
Email: ________________________
Professional ATCA Individual Membership Application Air Traffic Control Association
❑ Professional Individual Member – Industry/Academic/Association/Government Staff - $130.00 ❑ Military Staff E1 – E6 Enlisted rank - $26.00 please indicate branch of service: ❑ Army ❑ Air Force ❑ Marines ❑ Navy ❑ Air National Guard ❑ Military Staff E7 Enlisted - Officer rank - $130.00 please indicate branch of service: ❑ Army ❑ Air Force ❑ Marines ❑ Navy ❑ Air National Guard ❑ Professional FAA Employee Dues Withholding** Member $5.00 per biweekly pay period ($130 per year) ❑ Young Aviation Professional: Must have less than 5 years of experience in the industry – $75.00 per year Name: Mr./Ms./Other ( ______ ) _________________________________________________________________________________ (First)
(Middle)
(Last)
Title/Grade of Present Position: ________________________________________________________________________ Organization/Facility: __________________________________________________________________________________ Mailing Address: ______________________________________________________________________________________ (No.)
Contact:
(Street)
(City, State, Zip/Country Code)
Work ( _______ ) ____________________ ( ______ ) Area/Country Code
Fax ( ______ ) _________________________
Ext
Home ( ______ ) _____________________________
Country
Area/Country Code
Email _________________________________
Area/Country Code
❑ I want to “Opt Out” from having my information printed in an ATCA Membership Directory. Method of Payment: (Please indicate one) ❑ Bank Wire Transfer: Please call 703-299-2430 or email info@atca.org for bank information. ❑ Check or Money Order in U.S. Funds drawn on U.S. Bank made payable to Air Traffic Control Association – Send with application to ATCA ❑ **FAA Dues Withholding (please call 703-299-2430 to have an SF1187 form sent to you. – Please complete and return to ATCA with your application.) $5.00 per biweekly pay period. ❑ VISA ❑ MasterCard ❑ American Express Credit Card Number
CVV
Expiry
M M
Signature of Credit Card Holder
Y Y
Date
Annual membership dues include $20.00 for subscription to The Journal of Air Traffic Control. I would like to participate on the following ATCA Committees: ❑ Publicity & Public Relations Committee ❑ Aviation Education Committee ❑ Air Traffic Control Committee ❑ ATC Engineering & Development Committee Note: Dues, Contributions, Gifts or other fees paid to AIR TRAFFIC CONTROL ASSOCIATION, INC. may be deductible to members for federal income tax purposes as ordinary business expenses. Dues, Contributions, Gifts or other fees paid to AIR TRAFFIC CONTROL ASSOCIATION, INC. are not deductible as charitable contributions. Please consult your tax advisor for individual assistance in specific situations.
Air Traffic Control Association | 1101 King Street, Suite 300, Alexandria, VA 22314 | T: 703.299.2430 | F: 703.299.2437 www.atca.org
The Computational Air Traffic Control Brain Computational Red Teaming and Big Data for Real-Time Seamless Brain-Traffic Integration By Hussein A. Abbass, Jiagniun Tang, Rubai Amin, University of New South Wales, Canberra Campus, Australia
and Mohammed Ellejmi and Stephen Kirby Eurocontrol Experimental Centre, Brétigny, France
AS SESAR, NEXTGEN, AND THE WORLD MOVE TOWARD the automation of air traffic control (ATC) tasks to safely accommodate future growth in air traffic movements, it is clear that the air traffic controllers’ (ATCO) role will need to change significantly. We believe controller workload will remain a key factor in future systems, but with a change in focus from detailed repetitive activities to higher level monitoring and assurance of system-level safety. Therefore, new controller workload indices will need to be developed or adapted to be truly effective. To address this need, our research models the future division of ATC roles between the human and automation on the human brain’s two hemispheres. The left hemisphere is responsible for arithmetic, logic, and looks at the details; the right hemisphere is responsible for human qualities such as emotion, spatial awareness, and the capacity to judge holistically. We call our integrated model of decision-making within the future ATC system the “computational ATC brain” (CAB); it consists of automation (left) and controller (right) components that will deliver ATC services in the future. For example, calculating time to closest point of approach of two aircraft to estimate time of conflict is a typical systematic brain’s left hemisphere problem. As a classical problem for automation, we assign it to the left hemisphere of the CAB. The brain’s right hemisphere qualities – such as intuition and spatial awareness – that make airspace safe
are not well suited for automation and are assigned to the controller. Until now, most controller workload indices developed have focused on: Static, not adaptive, models that work for many operators, in many traffic situations resulting in “fixed” indices. Although dynamic data can be used in an index – dynamic density, for example – a static index does not adapt to changes in the environment. Little research has been done on adaptive workload indices that adapt the impact of a variable, such as the number of aircraft in a sector, from one situation to another. As in much of life, one size does not fit all. Providing a situational awareness only based on history without factoring in potential future actions. This works well for analyzing what happens but they are of limited use for developing systems that can avert potential overload situations. That is, once a situation is assessed to be of high Continued on page 15
12
Summer 2014
THE COMPUTATIONAL AIR TRAFFIC CONTROL BRAIN
Better anticipatory cognitive complexity metrics are needed to support air traffic controllers in real-time environments.
The Journal of Air Traffic Control
13
14
Summer 2014
PHOTOGRAPHY: OLIVER SVED / SHUTTERSTOCK.COM
workload, it is too late to do anything about it without a major disruption of traffic. This paper will address part of the wider research we have done in evaluating the true effectiveness of workload models, and will provide an overview of a prototype environment for evaluating adaptive and predictive workload models in the real time management of complexity. We will present results from a November 2013 experiment, aimed at evaluating a proof of concept for comparing mental workload assessed directly from electroencephalograph (EEG) data obtained from the human brain and a classical “fixed” workload model while closing the automation loop, performed by the authors at Eurocontrol Brétigny in France. Background Since our paper uses some terminologies not usually found in the ATC lexicon, we provide some background information below. The Big Data Challenge Big data is an opportunity that we cannot miss out on this century. The classical definition of big data relies on the 5V model: volume, velocity, variety, veracity, and value. The air traffic domain and the work in this paper are exemplars of the 5Vs. Volume is about the size of the data. Repeating the analysis presented in this paper in a single command and
control center with 20 operators will generate close to 10 terabytes (TB) of psychophysiological data alone every day. In our Australian ATC research lab based at the University of New South Wales, Canberra Campus, we developed the TOP-LAT (Trajectory Optimization and Prediction of Live Air Traffic) system; a passive ground watchdog system of air traffic. The system operated from 2008 to 2013 receiving the real-time traffic of the two Australian flight information regions – processing trajectories, estimating dif-
erogeneous (variety); in this paper, traffic and psychophysiological data needed to be aligned and fused together. The data contains noise (veracity) that needs to be dealt with, from artifacts in the EEG signal, noise in radar, and possible inaccuracy of ADS-B in certain geographic regions. With the complexity of the future air traffic system, it is hard to deny the value that will generate from properly utilizing big data within the air traffic domain. The era of big data is an era of opportunities for improving ATC worldwide.
The right hemisphere of the human brain needs to continue to operate as the right hemisphere of the CAB in the future. ferent emissions, calculating different task complexity metrics, estimating risk and analyzing the traffic network, to name a few. In this system, a few TB of data can be easily generated on a daily basis. We only saved raw data that enables us to recalculate what we may need in the future and system-level statistics as we did not have the storage capacity to save all the data being generated. The above examples capture velocity, whereby the situation is highly dynamic (for example, human mental load), causing continuous changes in the data. It is clear that the data is het-
The Computational ATC Brain (CAB) A brain with two hemispheres: the left hemisphere is made of silicon automating complex ATC tasks including tactical controller tools, midterm conflict detection and resolution, complex real-time adaptive tactical flow management tools, dynamic airspace sectorization tools, data links, and satellite navigation capabilities. The right hemisphere of the CAB is made of humans including ATCOs, pilots, supervisors, and different stakeholders. The human’s right hemisphere brain functionalities represent those The Journal of Air Traffic Control
15
THE COMPUTATIONAL AIR TRAFFIC qualities that no machine will replace, at least for years to come. The right hemisphere of the human brain needs to continue to operate as the right hemisphere of the CAB in the future. The role of ATCOs will change to emphasize right hemisphere functions. The integration of the computations done in Silico in the left hemisphere of the CAB and the computations done by humans in the right hemisphere is a complex Brain-Traffic Integration problem[1]. This problem sees the controller’s brain as an influential component “inside” and an integral part of the automation loop rather than an external recipient of those decisions made by the automation loop. But how to achieve seamless integration between the left and right hemispheres – the silicon occupying the left hemisphere and the humans occupying the right hemisphere – of the CAB remains to be a scientific challenge. The benefits from NextGen and SESAR will not reach their maximum potentials until we think anew of this interface to achieve seamless integration between the human and silicon brain that will manage the complexity of the future air traffic environment. This Brain-Traffic integration problem requires out-of-the-box thinking. The massive amount of data (the Big Data challenge) that will exist at the interface of the human brain and the air traffic environment require efficient mechanisms to analyze these data, extract the right cues at the right time for the right context, and seamlessly exchange cues between the left and right hemispheres of the computational
CONTROL BRAIN ATC brain. Computational red teaming (CRT) is possibly one scientific endeavor to study this interface in the form of a challenge from a risk lens to disentangle the uncertainties occurring at the interface and their impact on the environment and different objectives of the participants. Computational Red Teaming Red teaming (RT) is an ancient military concept to play “devil’s advocate” against our own plans, concepts, and systems to discover vulnerabilities, risk assess our thinking and systems, and design innovative mitigation strategies to secure these systems. The difference between RT and other test and evaluation approaches lies in its reliance on two concepts: “deliberate challenge” where the process assesses system boundaries and “deliberately” pushes a situation beyond these boundaries to stress-test the system, and “risk thinking” which emphasizes continuous innovation, uncertainty analysis in time and space, and analyses focused on systems’ objectives. CRT, an innovation of this research group[3], attempts to design artificial intelligence systems to assume the role of, or support, the human red team. CRT relies on an integrated architecture with optimization, data mining, and simulation components to design an intelligent system that is able to “challenge” and risk-assess a concept, technology, or a human. We emphasize the concept of “challenge” in CRT as opposed to a classical decision support system, which aims to “aid” or “support” humans.
Figure 1. A schematic diagram of the experimental environment.
ATWIT Tagging
EEG States
Low Level Signal Processing
Pilot
EEG Driver ATC
Traffic High-order States
Traffic Low-order States
Traffic Information
Advisory Pilot
Shadow Simulation Rule-based System
16
Summer 2014
Optimization
A challenge is a pro-active process whereby the situation is continuously evaluated, risk assessed, and actions are dynamically introduced to counteract a user, system, or environmental effects. To design a system that is able to challenge something, it needs to be able to 1) continuously receive information, extract patterns and trends, and assess the risk associated with the situation; 2) have available models or mechanisms to mimic the behavior
We need to understand causes and controls that enable us to influence the system under consideration when a measure indicates a state of risk. of the system it is trying to challenge and be able to evaluate the impact of a potential challenge; and 3) to exhibit the ability to search for optimal actions that will influence the system it is trying to challenge. CRT systems are designed to either discover vulnerabilities by proactively discovering holes (exposure of critical elements in a system to hazards or threats), or support humans’ blind-spots by proactively and continuously anticipating threats in the environment and providing counter-actions to help the human operator overcoming these challenges. In this paper, an undesired change in the complexity of the traffic in the environment is the hazard that needs to be continuously anticipated, assessed, and mitigated. CRT has been applied to different application domains. Within ATC, it has been used to evaluate mid-term conflict detection tools[4], ground-air interaction in the terminal maneuvering area[8], and correction of air traf-
THE COMPUTATIONAL AIR TRAFFIC CONTROL BRAIN maneuvers and actions required to steer the system back to a safe operating envelop, it is then a clear indication that the solution is incomplete. To put it succinctly, it is no longer sufficient to measure; we need to understand causes and controls that enable us to influence the system under consideration when a measure indicates a state of risk.
Figure 2. The EEG Cap during a pre-experiment testing on one of the authors. The diagram also shows the ATWIT system displayed on the right screen, the traffic scenario in the middle screen, and a communication panel on the left screen.
fic events[5], to name a few. In our previous work on CRT, we used it to discover vulnerabilities in systems. In this paper, we demonstrate, for the first time, its use to support humans. In this paper, CRT is used to demonstrate the two workload challenges discussed in the abstract, and to close the loop between measuring workload and intervening within the environment to manage workload. We share the first experiment of its kind to demonstrate that seamless integration of human brain data and air traffic data is possible, and that humans and machines can work in harmony. Real-Time Seamless Integration The objective of this work is twofold. From an academic perspective, it demonstrates that real-time, dynamic, and seamless integration of cues from human mental processes within automation improves the effectiveness of the decision support environment. From an applied ATC perspective, it reveals the advantages of using CRT to assess workload metrics. First, by acting in real-time on the workload information obtained through different metrics, we can assess the suitability of the metric for a real-time environment. For example, the dynamic density metric we used in the experiment reflected nicely the complexity of the traffic situation based on subjective metrics and informal discussions we had with the ATCOs. However, by the time the load is detected to be high, it was too late to do anything about it. In fact, we found that the interference of automation at this late stage increased ATCOs’ workload. Therefore, this metric is not suitable as an anticipatory metric for ATC complexity. Second, by assessing psychophysiological metrics obtained from sensors such as EEG, we can calibrate whether or not the traffic-dependent metric can truly assess mental load. By closing the automation loop through feeding EEG cues back into the decision support environment, CRT provides an objective and effective test for any workload metric. CRT couples the development of the metric with the “act” on the metric. If a metric measures or even anticipates workload, identifies that the system is outside its safe operating envelope, but we have no way of knowing what
The Study The hypothesis of the experiment was that it is possible to continuously, simultaneously, and seamlessly monitor and analyze data from the traffic and human mental/brain activities to control the complexity of the traffic in real-time in order to 1) seamlessly manage workload and 2) compare task and mental load indicators in real-time scenarios. The system (Figure 1 on the previous page) is designed to extract cues in real-time from both traffic and mental data to trigger CRT to balance complexity. Subjective assessment of traffic complexity through the ATC Workload Input Technique (ATWIT) developed by the FAA[7] is performed every two minutes. Brain signals are measured continuously from the ATCO in real-time using a Nexus-32™ EEG cab with 19 channels (Figure 2). These signals are analyzed and cognitive cues on the mental states of the ATCO are extracted. Simultaneously, traffic information is analyzed in real-time and task cues are extracted to capture the complexity of the traffic in play. One or both types of cues is used to make a decision based on the scenario being used. Changes in these two sources of complexity are continuously being assessed. When an undesired situation arises, the optimization component of CRT gets triggered to assess which maneuver or action in the environment is best to counteract the undesired state. It uses the simulation environment for impact and look-ahead analysis. Once this action is identified through the CRT process, it gets displayed on the advisory screen and the process continues. The tagging position has software that serves four purposes: 1) keeping track of the experimental protocol ensuring that every step has been completed; 2) recording subjective observations during the session; 3) sending event-markers to other software components, such as the start of session event; and 4) time-stamped synchronization messages to all other software components. The experiments provided a wealth of data that can be used to improve the models we built. For example, Figure 3 shows a picture of the areas activated in the brain for one of the controllers during one of the tasks. Red color indicates a high level of mental processing. This type of analysis provides us with deeper insight when evaluating workload. The four participants were previous ATCOs who are currently working within the Eurocontrol experimental center. Each air traffic scenario lasted for 50 minutes. During the first 25 minutes, the aim for CRT was to identify maneuvers that will increase complexity, while in the last 25 minutes the aim was to identify maneuvers to decrease complexity. Four different trials were conducted for each subject: one with CRT turned off, one with CRT relying on task complexity (using the index in [6]) alone to make decisions, one with CRT relying on cognitive complexity alone to make decisions, and one with CRT relying on both task and cogThe Journal of Air Traffic Control
17
THE COMPUTATIONAL AIR TRAFFIC CONTROL BRAIN
Figure 3. An example of brain activities for one of the controllers during a task. The two lines in the north symbolise the nose, while the two curves in the east and west depict ears.
nitive (based on brain data) complexity working together. CRT worked best when both task and cognitive complexity were used. Subjective assessments using ATWIT was used to assess ATCO’s perception of workload every two minutes. It revealed that the reliance on task complexity alone increased complexity in the last 25 minutes instead of decreasing it. The reason is that classical workload metrics detect complexity after they happen. The belated automated advice increased the complexity of the situation when the operator was already attempting to manage the complexity in the space. Cognitive complexity metrics acted, however, as lead indicators; thus, the advices were given on time for controllers to act. Discussions and Conclusion A recent human-based experiment conducted at the Eurocontrol Experimental Centre at Brétigny, France used computational red teaming (CRT) as a proof of concept for a test and evaluation environment for cognitive load and workload metrics. The results demonstrate the feasibility of using cognitive and task indicators to help ATCOs to manage complexity in airspace. Cues extracted from EEG brain signals acted as lead indicators that allowed the system to manage complexity. The situation reversed with workload indicators. Better anticipatory cognitive complexity metrics are needed to support air traffic controllers in real-time environments. Future research of this work is twofold. First, we will explore the concept of anticipatory adaptive workload indices where the metric for workload changes during one session based on circumstances. Secondly, we will continue to enhance our computational red teaming (CRT) environment to better test and evaluate the ability of a proposed metric to adapt to human mental processes and traffic situations, and potentially be used in real-time to inform better decision-making. Acknowledgements The authors are obliged to Mr. Edward Stevens for his thorough comments that tremendously helped in improving the readability of this paper, and to Mr. Steven Carver and Mrs. Marion Brophy for their professionalism and hard work in handling this submission. We thank Miss Shen Ren for her support during data collection, Dr. Jon Hegg for training the team on the hygienic and fast fitting of EEG caps and Dr. Anastasia Lampropoulou for support with Nexus 32. A special thank you goes to all participants who were very profes18
Summer 2014
sional, patient and supportive; and for Eurocontrol for being a forward-looking organization. This work was done with human ethics approval HC13251 under the title “Computational Red Teaming for Air Traffic Control.” References
[1.] Abbass H.A., Tucek D., Kirby S., and Ellejmi M. (2013) Brain Traffic Integration, Air Traffic Technology International, 34-39. http://www. ukipme.com/mag_airtraffic.htm [2.] Abbass H.A., W.M. Mount, D. Tucek and JP Pinheiro (2011) Towards a Code of Best Practice for Evaluating Air Traffic control Interfaces, Australian Transport Research Forum, Adelaide, Australia. [3.] Abbass H.A., Bender A., Gaidow S., and Whitbread P. (2011) Computational Red Teaming: Past, Present and Future, IEEE Computational Intelligence Magazine, IEEE Press, Vol.6, 30-42. [4.] Alam, S., Abbass, H.A., Lokan C.J. (2009) Computational Red Teaming to Investigate Failure Patterns in Medium Term Conflict Detection, 8th Eurocontrol Innovation Research Workshop and Conference, Eurocontrol Experiment Research Center, Paris, France. [5.] Amin R., Tang J., Ellejmi M., Kirby S. and Abbass H.A. (2013) Computational Red Teaming for Correction of Traffic Events in Real Time Human Performance Studies, USA/Europe ATM R&D Seminar, Chicago, IL, USA. [6.] Sridhar, B., Sheth, K.S., & Grabbe, S., (1998) Airspace Complexity and its Application in Air Traffic Management, 2nd USA/Europe Air Traffic Management R&D Seminar, Orlando, Florida. [7.] E.S. Stein (1985). Air traffic controller workload: an examination of a workload probe, US Department of Transportation, Atlantic City Airport, N.J. 08405, Federal Aviation Administration Technical Center. (Report Number: DOT/FAA/CT-TN84/24). [8.] Zhao W., Alam S. and Abbass H.A. (2013) Evaluating Ground-Air Network Vulnerability in an Integrated Terminal Maneuvering Area Using Co-evolutionary Computational Red Teaming, Transportation Research Part C, Elsevier, 29(4), 32-54.
ELVO
Low Cost, High Impact Enhanced Low Visibility Operations
PHOTOGRAPHY: LARS LINBLAD / SHUTTERSTOCK.COM
By David Hughes, Federal Aviation Administration Overview The FAA is making it possible for aircraft to land in lower visibility at little or no added cost using avionics already installed, an innovation that keeps traffic moving during adverse weather conditions at many airports, including some likely to have to shut down operations. The FAA’s Flight Standards Service came up with the idea in 2005 of allowing Instrument Landing System (ILS) approaches in lower visibility based on the insight that avionics already on board enable pilots to fly more accurately than ever before. A Flight Standards team proved some ILS approaches can be flown safely in lower visibility than previously thought or with much less expensive runway lighting systems. Once this was established, the FAA began updating hundreds of ILS procedures and developing new types. NextGen now funds the Enhanced Low Visibility Operations (ELVO) program that is expanding these capabilities to more airports. The payoff for passengers may be getting home for the holidays. For example, at Portland (PDX), weather would have shut down ILS approaches on December 24, 2009, without the new lower visibility requirement. The Details The FAA’s effort to improve U.S. airport access during low visibility is allowing air carrier aircraft to take off and land during extremely low visibility by providing operational credit for aircraft equipped with advanced avionics such as autopilot, flight director, head-up display (HUD), and autoland. As long as the operator is approved by the FAA and incorporates the appropriate training into their program, these aircraft can fly to sites that utilize lower minimums for landing than standard Category (CAT) I or standard CAT II. Departures can be to as low as 500 feet versus the current
1,600 feet, clearly providing an operational advantage. The impact of low visibility due to weather systems or fog can cause flights to be diverted, delayed, or canceled at a high cost to airlines, passengers, and air cargo or small package carriers and their customers. Disruption of air traffic at even one of the busiest 30 U.S. airports in the morning can cause ripple effects throughout the National Airspace System (NAS) all day. When major weather events cause visibility to go below minimums at several major airports in one region, it can spread trouble like tumbling dominoes. Getting airlines’ schedules back to normal can take days. Efforts led by the FAA Flight Standards Service have resulted in significant benefits to flight operations that allow operations to continue in certain low visibility situations. Fewer flights are delayed, diverted, or canceled and can move through the NAS as scheduled. Standard CAT I, CAT II, and CAT III ILS procedures use visibility minimums based on measurements of the Runway Visual Range (RVR) system. This ground-based system near the runway measures visibility in feet, background luminance, and runway light intensity to determine the distance a pilot should be able to see down the runway. The ILS, lighting, and RVR are all required ground components. RVR is critical for determining what the ILS minimums will be for each landing category because of its accuracy in showing realistic runway visibility. Historically, more airfield lighting equated to lower landing minimums. The FAA has required high-intensity approach lighting systems, centerline runway lighting and touchdown zone lighting for the lowest landing minimums. Such lighting equipment is expensive to install and maintain. Flight Standards proved that some expensive ground lighting requirements can be mitigated by advanced avionics common in modern aircraft. This has been extremely The Journal of Air Traffic Control
19