Fall 2012
Thursday, Oct. 11 and Friday, Oct. 12

Draft Agenda and Abstracts

Thursday, Oct. 11 – room 33-206
12:30 - 1:00 PM Sign in
1:00 – 2:30 PM MIT Research Activities - Prof. R. John Hansman

• Anomaly Detection in Flight Data Recorder Data of Airline Operations
  Lishuai Li
• Design and Optimization of a General Aviation Conflict Alerting System
  Fabrice Kunzi
• Preliminary Observations of Attentional Filtering Techniques Applied in Current Electronic Cockpit Displays
  Abhizna Butchibabu

2:30 - 3:00 PM BREAK
3:00 - 4:30 PM MIT Research Activities - Prof. R. John Hansman

• Influence of Runway Occupancy Time and Wake Vortex on Runway Throughput
  Tamas Kolos-Lakatos 
• Investigating the Impact of Air Transportation on Economic Productivity
  Justin Stilwell

6:30 - 9:00 PM Dinner at Local Restaurant

Friday, Oct. 12 - Room 33-206
8:30 AM - 9:00 AM Coffee and pastries
9:00 - 10:30 AM Ohio University Research Activities - Prof. Wouter Pelgrum

• Flight test evaluation of eDME performance enhancements
  Kuangmin Li
• Simulator Architecture for L-Band Terrestrial Radionavigation
  Adam Naab-Levy
• Advanced Display Concepts in Support of NextGen
  Tony Adami

10:30 – 12:00 PM Princeton University Research Activities – Prof. Luigi Martinelli

• Data-Based gPC for Uncertainty Quantification in Wing Icing
  Anthony DeGennaro
• Sensitivity analysis in perturbed leading edge wing simulations [for icing prediction]
  Mark Lohry
• Modeling of unsteady flow over a wing at high angles of attack
  Carla Bahari

12:00 – 12:30 PM Principal Investigators’ Meeting

---

Massachusetts Institute of Technology

Anomaly Detection in Flight Data Recorder Data of Airline Operations
Lishuai Li
Because airline operations are exceedingly safe, future hazards are likely to emerge from new emergent sources. The objective of this study is to develop data analysis methods that can detect anomalous flights from Digital Flight Data Recorder (DFDR) data without the need to specify specific exceedance parameters or anomalies in advance.  The method under development detects anomalous flights based on a hyperspace cluster analysis, which identifies clusters representing nominal flight behaviors and allows identification of unusual flights that do not correspond to a normal cluster.  The anomalous flights could then be reviewed by subject matter expert pilots to determine if they are benign, or may be precursors of emerging risks.  This data-driven approach provides a way to look for unknown safety issues.

Design and Optimization of a General Aviation Conflict Alerting System
Fabrice Kunzi
Due to high nuisance alarm rates, current traffic alerting systems have limited usability in the airport environment where a majority of mid-air collisions occur. As part of NextGen, ADS-B will become the primary surveillance source in the National Airspace System. Using the higher quality surveillance information available via ADS-B, the Traffic Situation Awareness with Alerting Application (TSAA) will be the next generation of traffic alerting for General Aviation. TSAA will provide timely alerts to the flight crew in order to increase their traffic situation awareness.

Current work involves the design and evaluation of a new conflict alerting algorithm and it's performance under real-world conditions. A study of the sources of uncertainty associated with the ADS-B data that is used by the algorithm to predict conflicts was conducted. In order to evaluate the algorithm's performance with track data from aircraft encounters, a fast-time simulation environment was created that accurately degrades the "truth data" with uncertainties representative of what would be observed in the real world. Initial results have been generated and are presented.

Preliminary Observations of Attentional Filtering Techniques Applied in Current Electronic Cockpit Displays
Abhizna Butchibabu
The rapid emergence of cockpit installed electronic displays (e.g., MFDs, PFDs) and portable displays (e.g., iPADs, iPhones), and the rapid adoption by pilots have introduced a need for the design and development guidance for these displays.  As a result, there is a need to understand the Attentional Filtering Techniques used in these displays to efficiently filter and prioritize the high levels of information that are being presented to the pilots. In order to identify these techniques, and understand common ways in which they are implemented, a review of 15 currently available electronic displays from 13 avionics manufacturers was conducted, and observed attentional filtering techniques are recorded.

Influence of Runway Occupancy Time and Wake Vortex on Runway Throughput
Tamas Kolos-Lakatos 
Runway occupancy times and inter-arrival separations were measured at Boston Logan, La Guardia, Philadelphia, and at Newark to determine the potential benefits that could be gained by introducing wake vortex mitigation processes. Inter-arrival separations between landing aircraft were compared in visual and in instrumental meteorological conditions that serve as the basis for fast time computer simulations determining airport throughput.

Investigating the Impact of Air Transportation on Economic Productivity
Justin Stilwell
As part of an FAA-sponsored project investigating the impacts of air transportation on U.S. economic productivity, we have focused on identifying the mechanisms of such impacts, while addressing both intermediate and final use of air transportation. We have gathered data from a range of sources, including the BEA’s Input-Output (I-O) accounts, the BLS’s productivity statistics, and the BTS’s Schedule P12 data to help inform and suggest revisions to the mechanisms. Preliminary analyses of the I-O accounts and correlations between air transportation use and labor productivity appear to suggest that industry productivity (particularly in the goods-producing industries) might not be as dependent on air transportation as previously hypothesized. Air transportation appears to be more important for services-producing industries and for personal travel. This presentation highlights recent work investigating industry location preferences based on air service availability. This research has supported the hypothesis that company headquarters are more likely to be located near air service than most other industries or the general population. Examples of recent headquarters relocations have also been investigated with regards for the impact of such moves on improving access to air transportation.

Ohio University

Flight test evaluation of eDME performance enhancements
Kuangmin Li
Enhanced distance measuring equipment (eDME) is proposed as a non-GNSS alternate Positioning, Navigation and Timing (PNT) solution for NextGen aviation.  Cm-level displacement accuracy is demonstrated with DME carrier phase. This, combined with DME pseudoranging, allows for various methods that increase the accuracy and integrity of the Positioning, Velocity, and Timing (PVT) solution. Carrier Smoothed Pseudorange (CSP) significantly reduces pseudorange noise and multipath. Next, variations in Pseudorange Minus Carrier (PMC) are used to bound the potential pseudorange bias. Finally, the Pseudorange Noise MultiPath (PNMP) algorithm combines CSP and PMC to achieve meter-level ranging accuracy with integrity, which is demonstrated using recent flight-test results.

Simulator Architecture for L-Band Terrestrial Radionavigation
Adam Naab-Levy
Aviation requires an alternate to GNSS Positioning, Navigation, and Timing, which most likely will be an L-band terrestrial radionavigation solution. An efficient development and detailed performance assessment of these potential future systems requires a high-fidelity simulation environment. This presentation presents a simulator architecture currently under development by Ohio University which will be able to accurately simulate the various error sources of ground and air equipment as well as L-band propagation. 

Advanced Display Concepts in Support of NextGen
Tony Adami
In support of NextGen Air Transportation System goals, FAA-funded research into advanced cockpit display concepts is underway at Ohio University. In partnership with recognized experts in the field from TU Delft in The Netherlands, implications of NextGen are being studied from the the pilots' perspective to identify display concepts that will fill information gaps existing in the current NAS system. A broad literature search has been performed identifying the main theoretical and experimental results on cockpit display-related issues over the past decade, and an analysis is underway to help bring the current study into focus. A part-task simulator is being assembled at OU that will be used for pilot studies of concepts such as Conflict Probing and the Energy Management Primary Flight Display, and the current status of construction is reported.

Princeton University

Data-Based gPC for Uncertainty Quantification in Wing Icing
Anthony DeGennaro
The formation and accretion of ice on an airplane - in particular, on the wings - is a phenomenon which is problematic for aircraft performance. Icing can significantly alter airflow over the wing by inducing separation earlier, which could affect the lift curve , lower the aerodynamic efficiency, or lead to stall at lower angles of attack. This work aims to quantify the uncertainties in various aircraft performance metrics (ex.-- stability derivatives) which arise due to uncertainty in wing icing conditions/parameters. This is achieved using a data-based generalized Polynomial Chaos (gPC) framework. The first step in this methodology is to approximate the cumulative distribution function (CDF) which describes the uncertainty in a particular aircraft performance metric arising due to uncertainty in some wing icing parameters. This can be achieved in a variety of ways (ex.-- Monte Carlo simulations). Once this has been done, the CDF is orthogonally projected onto the span of a complete orthogonal basis set consisting of polynomials of a stochastic variable (ex.-- Hermite polynomials). This projection is truncated at a finite number of modes. Once this procedure has been repeated for a wide range of input icing uncertainty conditions, linear regression and interpolation may be performed to attain a single matrix which maps from some arbitrary input uncertainty to a finite number of modes which approximate the CDF/PDF of the resulting output uncertainty.

Sensitivity analysis in perturbed leading edge wing simulations [for icing prediction]
Mark Lohry
In order to begin our study of the aerodynamic effects of wing icing, we assess the capabilities of computational tools to predict separated flows over perturbed wing geometriesand the sensitivity of the simulation to the perturbation.

Modeling of unsteady flow over a wing at high angles of attack
Carla Bahari
The unsteady flow over an infinite wing at high angles of attack is studied.  Through Reynolds-Averaged Navier-Stokes simulations, sinusoidal and pitch-up maneuvers were conducted. The response in lift was used to construct linear reduced order models.  In addition, longitudinal dynamics were investigated, in particular stability of the short period and phugoid modes. Future directions includes more maneuvers, further development of reduced-order models and investigations of stability in lateral-directional modes of motion.

 

For additional information, please contact:
Sally Chapman-
email-sallyc@mit.edu
Phone (617) 253-4926 • Fax (617) 253-4196