Projektarchiv

MOSES - More Operational Flight Security through increased Situation Awareness



Situation awareness is a commonly used term in the context of flight safety.

 

Within the framework of the DLR-internal project MOSES these answers have been the focus of research.

Engineers, Psychologists and Medical Scientists joined in an interdisciplinary team representing the major human factors expertise of the three DLR departments of Human Factors (FL, Brunswick), Flight Physiology (ME KP, Cologne) and Aviation and Space Psychology (ME HH, Hamburg).

The project focussed on measuring situation awareness of pilots. Therefore a metrological approach to record the phenomena has been proved in tests with more than 80 pilots in the DLR-Institute of Flight Guidance simulator cockpit. 

 

 

Content:

 

 

Overview

Situation awareness is considered an important factor for successful and safe acting in complex environments like the cockpit. Therefore adequate measuring techniques for situation awareness are required to evaluate impact and improvement of assistance systems. 

The project focussed on measuring situation awareness of pilots. Therefore a metrological approach to record the phenomena has been proved in tests with more than 80 pilots in the DLR-Institute of Flight Guidance simulator cockpit. 

 

At the same time possible enhancement of situation awareness has been investigated by use of the following novel assistance systems, which have been applied under experimental control for the first time:

  • the interactive navigation display A-HMI (Airborne Human Machine Interface, project ASSIST)
  • the advanced visual vision system ADVISE (Advanced Visual System for Situation Awareness Enhancement) and
  • the taxi and ramp management display TARMAC-AS (Taxi And Ramp Management And Control – Airborne System)
A_hmi
A-HMI: planning the flight in advance via the interactive navigation display.
TARMAC-AS: supports the real-time surveillance during taxiing phase.
ÁDVISE: supports guidance of the aircrafft during take-off and landing.

[ back ]

Situation Awareness: operational definition

 

graphic representation of Enddyley's model of situation awareness, Endsley, 1988
As a construct for defining situation awareness (SA) the definition by Endsley (1998) was of central importance. The definition follows the idea of information processing and can be divided in three levels:

 

  • Level 1: perception, the „serial” perception of single elements of the environment,
  • Level 2: comprehension of perceived elements (a “global picture”)
  • Level 3: projection, resulting from levels 1 and 2, looking ahead into the near future.

 

Endsleys model differentiates situation awareness from decision making and performance, the execution of an action.

[ back ]

 

Multimodal measurement battery for capturing situation awareness

 

The multimodal measurement battery which has been assembled for the project MOSES, follows strictly the idea of Endsley’s above explained definition of situation awareness.

 

Inputs to be made by the operator are increasingly restricted in a highly automated environment like the cockpit. In contrast the monitoring task is becoming predominant. It is therefor necessary to study the way pilots gather their information. i.e. their eye movement behavior as well as their physiological state parameters. Furthermore psychological data of self and outside assessment should be used and set into context to the specific conditions of the scenario, the requirements of the system and the system state.

 

In combining the specific competencies of the three departments of Human Factors, Flight Physiology and Aviation and Apace Psychology a corresponding measurement battery could be composed, which allows the gathering of eye movement behavior and EEG data at the same time.

 

The Elements of the MOSES measurement battery are mapped onto the graphic representation of Endley's model.

The components in detail:

  • Measurement of eye movement with head mounted system iViewX by SMI (Berlin) (resolution: < 0,1 degrees [pupil]; < 0,5 degrees [compensated for movements]) (sample rate: 50 Hz)
  • Recording of pilots input into the system and flight attitude data of the simulator (sample rate: 20 Hz)
  • Questionnaires, amongst others newly developed Questionnaire SAP (Self Assessment of Pilots) for measuring situation awareness during flight and taxiing phase
  • Recording of physiological parameter as influencing factors for situation awareness: EG, EKG, EOG, Respiration, Cortisol (stress hormone soluble in saliva), pupil diameter
  • Gathering of self assessment in terms of situation awareness: mental work load (NASA TLX), fatigue (FAT), mental state (MDBF
  • Eligibility data as influencing factor for situation awareness (mental capacity, personality disposition)

Depending on the specific task the multimodal measurement battery can be adjusted for future tests.

 

Measurement of eye movement: the fixation of the semipermeable mirror and the infra red light source at the shown measurement facility allows the additional applikation of electrodes to capture EEG.

Electrode's positions for measuring the EEG, EOG and EKG.

Example of a eye movement video: the focus of attention (orange circle) lays on the artifical horizon display in this example. Upper lieft corner shows the number of captured frames, lower right corner displays the time.

 

Self assessment after the scenarios via notebooks with touch screens.

 

 

 

 

 

 

 

 

[ back ]

 

Design of Scenarios that make insufficient situation awareness observable

Within MOSES no measurement techniques for SA that directly grab information (SAGAT, SPAM) have been used as one assumed that these techniques disturb the flow of the procedure of the simulation too much and therefore diminish the explorative power of the results. Instead the project followed an approach that is called “implicit measurement” (Durso, 1999) that starts with the design of the simulation scenarios.

Visualization of the flight phases which have been identified and investigated for the MOSES tests of TARMAC-AS display (initial approach, intermediate approach, final approach, taxiing on runway, taxiing on exit, taxiing on apron).

high-res JPEG 

Basically simulations are designed in a way that during the simulation events occur that ask for unambiguous and highly trained reactions from the pilot. In case these reactions do not come up, one can assume a loss of situation awareness. If there is i.e. an object on the runway and the pilot is not cancelling the landing procedure to avoid a collision or is not breaking than this behavior can be rated as an indicator for insufficient SA, which means that critical occurrences have not or to late been remarked by the pilot.

 

A second method to gain insights into the grade of SA through the scenarios design is the systematical variation of SA-limiting circumstances. These are among others:

  • Different wind, e.g. view conditions up to CAT II

  • Different traffic conditions

  • Cooperative, uncooperative obstacles whilst taxiing (aircraft/ further vehicles like fueling vehicles.

These conditions harden the extraction of flight guidance relevant information for keeping up SA.

 

An article regarding scenario design can be found in [14].

[ back ]

 

Testing

Prior to the start of a scenario: check of flight plan (left) and calibration of the eye movement tracking (right).

Four campaigns with more than eighty test persons took place, including pilot students as well as experienced pilots. 

 

 

The study took place in the GECO (Generic Experimental Cockpit of the Institute of Flight Guidance). This fixed based simulator with collimated outside view is used to assess and demonstrate flight guidance functions.

 

A synchronized record of simulator, eye movement and physiological data during the study is guaranteed.

 

 

 

 

Pre-tests with Head-p Display (HUD).

GECO and measurement unit for capturing eye movement data (right).

Human-Factors_medium
Measurement of eye movement during the use of TARMAC-AS.

 

[ back ]

 

Data interpretation and analysis

 

First results show thaat the head down version of TARMAC-AS does not cause an attentional capture of the pilot.
Within MOSES an automated interpretation of eye movement behavior has been developed that calculates cumulated dwells times and attention frequency for defined areas of interest per segment. This procedure is much more effective than the often used “manual” counting of (gaze) targets.

Afterwards the eye movements can be mapped onto the simulator- and physiological data.

 

 

 

 

 

[ back ]

Proposals for design and realization

Guidelines for selection of display- and input alternatives should be derived from the experimental results. To realize a human centered automation assistance systems are measured by their suitability regarding human factors.

 

Use and assessment of Head-Up Displays in the GECO.

Testing of different symbols on a Head-Mounted-Display (HMD).
 
 

 

[ back ]

Literature

[1] Biella, M. (2001). Mehr operationelle Flugsicherheit durch Erhöhung des Situationsbewusstseins (Projektplan MOSES). DLR-IB, 112-2001/19, (2001)

[2] Biella, M.: (2002). Messung von Situationsbewusstsein. CCG-Seminar TV 3.02 "Moderne Unterstützungssysteme für den Piloten", Braunschweig, 7.-9.10.2002, Carl-Cranz-Gesellschaft, (2002)

[3] Biella, M.: (2003). Situationsbewusstsein messen und verbessern. VC Info, 03-04/2003, (2003), S. 29-30,

[4] Biella, M.: (2003). Situationsbewusstsein im Cockpit: Erfassung in kritischen Flugphasen. 5. Berliner Werkstatt Mensch-Maschine-Systeme, Berlin, 08.-10.10.2003, Zentrum Mensch Maschine Systeme, (2003)

[5] Biella, M.: (2004). Human Factors im Cockpit: Situationsbewusstsein in kritischen Flugphasen. 5. Braunschweiger Symposium "Automatisierungs- und Assistenzsysteme für Transportmittel", Braunschweig, 18.02.2004, Gesamtzentrum für Verkehr Braunschweig (GZVB), (2004)

[6] Biella, M.: (2004). Situationsbewusstsein messen und verbessern: Das DLR-Projekt MOSES. Institutsüberprüfung, 20. Januar 2004, DLR-Institut für Flugführung, (2004)

[7] Biella, M. (2008). Pilot gaze performance in critical flight phases and during taxiing: Results from the DLR-Project MOSES. ISAP'7 Proceedings (in press)

[8] Biella, M., Hörmann, H.J., Meier, B.A., Radke, H., Samel, A., et al.: (2002). MOSES: Arbeitshypothesen und Methoden. DLR-IB, 112-2002/01, (2002)

[9] Biella, M., Schäfer, H. (2002). Situationsbewusstsein: Definitionen und messtheoretischer Zugang.. 44. Fachausschusssitzung Anthropotechnik "Situation Awareness in der Fahrzeug- und Prozessführung"; Langen, 22.-23.10.2002, Deutsche Gesellschaft für Luft- und Raumfahrt, Fachausschuss T5.4 Anthropotechnik, Situation Awareness in der Fahrzeug- und Prozessführung, S. 251-273

[10] Biella, M., Teegen, U.: (2002). Human Factors im Cockpit: Situationsbewusstsein messen und verbessern. 43. Kongress der Deutschen Gesellschaft für Psychologie; Humboldt-Universität zu Berlin; 22.-26. September 2002, Deutsche Gesellschaft für Psychologie

[11] Jakobi, J., Lorenz, B., Biella, M.: (2004). Evaluation of an Onboard Taxi Guidance System. Human performance, situation awareness and automation technology conference, HPSAA II, Daytona Beach, 22.-25.03.2004, Embry-Riddle Aeronautical University, Daytona Beach, Human Performance, Situation Awareness and Automation: Current Research and Trends, S. 143-149, Lawrence Erlbaum Associates, Inc., Publishers, (2004)

[12] Korn, B., Lorenz, B., Többen, H., Döhler, H.U., Hecker, P.: (2004). Radar PAPIs: Human Factor Issues of EVS Landing Aids. Enhanced and Synthetic Vision 2004, Orlando, Florida, USA, 12.04.2004, SPIE, Enhanced and Synthetic Vision 2004, S. 23-30, SPIE - The International Society for Optical Engineering, (2004)

[13] Lorenz, B., Biella, M., Jakobi, J.: (2004). Enhancing pilot situation awareness by using an onboard taxi guidance system: an empirical study. Enhanced and Synthetic Vision 2004, Orlando, FL, USA, 12.4.2004, SPIE, SPIE, Enhanced and Synthetic Vision 2004, S. 125-133, SPIE - The International Society for Optical Engineering, (2004)

[14] Lorenz, B. , Biella, M., Schmerwitz, S. , Többen, H.: (2004). Verlässlichkeit der Interaktion zwischen Pilot und Assistenzsystem: Erfahrungen mit seltenen kritischen Ereignissen in Simulationsszenarien. 46. Fachausschusssitzung Anthropotechnik, Warnemünde, 12.-13.10., 2004, Deutsche Gesellschaft für Luft- und Raumfahrt (DGLR) e.V., Verlässlichkeit der Mensch-Maschine-Interaktion, S. 271-294, (2004)

[15] Lorenz, B., Korn, B.: (2004). Crew coordination issues of EVS-approaches. Enhanced and Synthetic Vision 2004, Orlando, FL, USA, 12.4.2004, SPIE, SPIE, Enhanced and Synthetic Vision 2004, S. 89-97, SPIE - The International Society for Optical Engineering, (2004)

[16] Lorenz, B., Többen, H., Schmerwitz, S.: (2005). Human performance evaluation of a pathway HMD. SPIE Defense & Security Symposium, Orlando, FL, 28.03.-01.04.2005, The International Society for Optical Engineering, (2005)

[17] Rovira, E., Lorenz, B., Edinger, C., Becker, H., Kuenz, A., et al.: (2004). Pilot conflict resolution planning: effects of levels of automation and traffic density. Human Performance, Situation Awareness, and Automation Technology Conference (HPSAA II), Daytona Beach, 22. - 25. März, 2004, Embry-Riddle Aeronautical University, Daytona Beach, FL, USA, (2004)

[18] Schmerwitz, S.: Entwurf und Implementierung einer Mensch-Maschine Schnittstelle für ein Enhanced and Synthetic Vision System. DLR-IB, 112-2005/12, (2005)

[19] Többen H., Lorenz B., Schmerwitz S.: (2005). Design of a pathway display for a retinal scanning HMD. Defense & Security, Orlando, 28.03.2005, SPIE, (2005)

[ back ]


Contact
Marcus Biella
German Aerospace Center

Institute of Flight Guidance
, Human Factors
Tel: +49 531 295-2001

Fax: +49 531 295-2550

E-Mail: Marcus.Biella@dlr.de
URL for this article
http://www.dlr.de/fl/en/desktopdefault.aspx/tabid-1150/1741_read-3121/