Human performance

Measuring Pilot Knowledge in Training: The Pathfinder Network Scaling Technique

Assessing Knowledge Acquisition in Academy Flight Screening (AFS)

Exercise Pacific Link 2: Distributed Training for Air Battle Managers

Birds of Prey: Training Solutions to Human Factors Issues

PACERS: Platoon Aid for Collective Employment of Robotic Systems

Soldier-Centered Interface Design for Armoured Vehicles:  Lessons and Guidelines for Acquisition Project Staff

Profiling is Politically ‘Correct’: Agent-Based Modeling of Ethno-Political Conflict

Modeling Human Perception of Situation Awareness During Constructive Experimentation

Automated Scenario-Based Training Management: Exploring the Possibilities

A Quantitative Measurement of Presence in Flight Simulators

The  Appropriate  Loudness  for  Flight  Simulator  Voice  Communications

 

 

Measuring Pilot Knowledge in Training: The Pathfinder Network Scaling Technique

2007 Paper No. 7293

 

Leah J. Rowe

L-3 Communications

Mesa, AZ

 

Roger W. Schvaneveldt

Arizona State University Mesa, AZ

 

Researchers have attempted to measure pilot knowledge and changes in knowledge, in both simulated and live-fly events. However, measurement in these training environments has been more successful in measuring overall flight performance outcomes rather than on underlying changes in knowledge. Research to assess changes in pilots’ knowledge as a result of training is underway at the Air Force Research Laboratory (AFRL) in Mesa, Arizona, using the Pathfinder Network Scaling technique. The Pathfinder method uses individual judgments of the relationships between concepts/constructs in a domain as a basis to develop an empirically derived representation of knowledge about the concepts/constructs. These representations can be compared and changes in representation can be quantified to assess the impact of an intervention on knowledge.  Previous research has demonstrated the value of Pathfinder for assessing the impact of both education and training interventions in domains such as computer programming. At AFRL, pilots, as part of a week-long 4-ship F-16 Distributed Mission Operations (DMO) training research program, participated in a Pathfinder study to asses F-16 pilot understanding of complex combat mission constructs/concepts critical to mission performance. The objective was to assess training effects that are more fundamental and process-orientated. This paper will report findings from a sample of 71 F-16 pilots who vary in experience level. Our results will be discussed both in terms of practical utility of the Pathfinder technique as a measurement methodology and in terms of knowledge measurement as a criterion for evaluating training.

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Assessing Knowledge Acquisition in Academy Flight Screening (AFS)

2007 Paper No. 7009

 

Lt Abby Barger, Lt Kamille Kemp, Capt Chad Tossell

Department of Behavioral Sciences & Leadership

USAF Academy, CO

 

The Air Force relies on the Academy Flight Screening (AFS) program at the United States Air Force Academy (USAFA) to screen, motivate, and prepare pilot candidates for entry into Undergraduate Pilot Training (UPT).  In this training, cadets are taught principles and techniques used in basic flying operations.  In order to assess the effects of this training, we compare knowledge structures obtained from individuals that have completed AFS with knowledge structures obtained from individuals that have not graduated from AFS.  Each subject was presented 435 pairs of basic aviation concepts and asked to rate their relatedness.  These ratings were used to develop pathfinder networks and measures of internal consistency.  Results show differences in cadets’ understanding of important flying concepts.  In addition, networks obtained from AFS graduates are more similar to one another than the non-graduates.  Analysis of knowledge structures using pathfinder was also shown to be a useful adjunct to performance-based methods of assessing training already employed in AFS.    

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Exercise Pacific Link 2: Distributed Training for Air Battle Managers

2007 Paper No. 7104

 

Craig Eidman, Dr Peter Crane, 1Lt Clinton Kam, Dr Sara Elizabeth Gehr, Mitch Zamba

Air Force Research Laboratory

Mesa, AZ

 

Dr Michael Skinner, Christopher Best, Sam Hasenbosch, Eleanore Burchat, Melanie Finch, Christopher Shanahan

Air Operations Division, Defence Science and Technology Organisation,

Melbourne, Victoria

 

Distributed simulation provides warfighters with training to enhance their team and inter-team skills with greater frequency and at lower cost than range training exercises. Distributed simulation training for small groups of warfighters such as a formation of four fighters working with an Air Battle Manager can be focused on specific skills such as beyond-visual range, dissimilar air combat tactics using constructive simulations as adversary forces.  Training for command and control teams, however, requires interactions among blue force entities, particularly voice communication, that cannot be supported using only constructive simulations.  One solution is to conduct large scale virtual training events such as VIRTUAL FLAG exercises.  Another solution is to combine human-in-the-loop virtual simulators with white-force role-players who provide responsive verbal communications for constructive entities.  Both of these approaches require participation from a significant number of warfighters or subject matter experts which increases training cost and decreases ability to focus training on command and control teams such as Air Battle Managers.  To overcome these difficulties, Australian and US researchers conducted Exercise Pacific Link 2 in which an Air Battle Manager team in Melbourne, Australia was networked with a four-ship of F-16 simulators and a constructive forces simulator in Mesa, Arizona.  Using a novel approach to scenario design combined with an improved constructive entity generator, a small team of pilots and engineers provided five, fully interactive four-aircraft formations of F-16s which engaged multiple waves of adversary aircraft over a one-hour vulnerability period.  Evaluations from the Air Battle Management team…

 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Birds of Prey: Training Solutions to Human Factors Issues

2007 Paper No. 7133

 

Robert T. Nullmeyer 

Lt Col Robert Herz

Air Force Research Laboratory

Mesa, AZ

Gregg A. Montijo

Crew Training International

Memphis, TN

 

Robert Leonik

CAE-USA

Shalimar, FL

 

 

The use of unmanned air vehicles (UAVs) in military operations is expanding rapidly, and this trend will likely continue given increases in funding for UAV development from $3 billion in the 1990s to over $12 billion for 2004-2009. High UAV mishap numbers have generated multiple reviews of unmanned operations in the past few years, but even within common platforms, different analysts attributed these mishaps to differing causes.  Thirty Air Force Predator Class A mishaps (more than $1 million damage) occurred from the introduction of this system into the Air Force inventory in 1995 through the end of FY 2006. Reports were reviewed to identify trends.  Substantial changes over time were observed regarding annual mishap rates, annual mishap counts, and causal factors. Mishap rates across the past three years dropped to less than one half the rate across earlier years. Mishap counts, however, steadily increased, as did Predator flying hours. Early mishap reports typically cited mechanical problems and operator station design issues. Mechanical problems were much less frequently cited in the last three years.  Rather, 80% of recent mishaps cited causal human error factors. Equipment interface problems were still cited as causal or major contributing factors in almost half of recent mishaps. Recent mishap reports often cited shortfalls in skill and knowledge (checklist error, task misprioritization, lack of training for task attempted, and inadequate system knowledge), situation awareness (channelized attention), and crew coordination. These trends come in a period characterized by a rapidly growing crew force and highlight the need to revisit both individual and team Predator training objectives and consider alternative training interventions that focus on the practice and improvement of these key operator skill areas. Predator team coordination and situation awareness training objectives are also addressed for the command and control personnel with whom Predator crews interact. 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

PACERS: Platoon Aid for Collective Employment of Robotic Systems

2007 Paper No. 7110

 

Paula J. Durlach

U.S. Army Research Institute

Orlando, Florida

 

In the future, Army platoons will be equipped with small aerial and ground robotic systems, which will provide the unit with real time information about the immediate surroundings; what’s over the next hill, around the next corner, or on the roof of a building. The Army currently is supplying platoons with both prototype and commercially available unmanned systems to evaluate military utility. The training provided prior to these “experiments” focuses almost entirely on individual operator training; but, training on how to integrate the system into unit operations is also required if the system is going to be properly exploited. While operators need training on the details of each system, the rest of the unit needs training on more general aspects involving coordination and communication. In addition, the employment of robotic systems puts new responsibilities on the platoon leader in terms of planning, delegation, resource allocation, coordination and workload, which he or she must learn to handle. Many of these unit level training objectives are common across different systems, both aerial and ground. The purpose of this paper is to (1) lay out these system-general aspects and (2) suggest a list of activities that trainers and leaders could focus on in order to help train unmanned system employment at the unit level. For each activity, associated observations and after-action review questions are suggested. Trainers could use these to help assess proficiency and coach the unit in system employment. One benefit of this system-general approach is that trainers don’t need to stay conversant with the details of every system a unit may bring to a training event. The benefits that this kind of training could provide will be illustrated with examples from the Micro Aerial Vehicle Advanced Concept Technology Demonstration 2006 Soldier Experiment. 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Soldier-Centered Interface Design for Armoured Vehicles:  Lessons and Guidelines for Acquisition Project Staff

2007 Paper No. 7468

 

Captain Erik Esselaar

Department of National Defence

Ottawa, Ontario, Canada

Joe Armstrong

CAE Professional Services

Ottawa, Ontario, Canada

 

Armoured Vehicles (AVs) have become sophisticated pieces of modern machinery. Today’s soldier is faced with expanded roles in AVs that greatly increase workload, tasks, and training requirements to operate these modern vehicles.  Soldiers must now manipulate several complex systems such as fire control, command, navigation, defensive aids and automotive systems simultaneously as well as manage the distribution of crew tasks and maintain overall situational awareness.  Exacerbating the problem in current AVs is that each vehicle sub-system has separate interfaces that were developed independently by sub-system contractors.  It is clear that in order to support the continued development of future AVs and harness the full capabilities of these complex systems, we must consider the soldier as an integral part of the vehicle system, and design system interfaces that reflect their requirements.  This soldier-centered interface development process will be traced through the outcomes and experiences of the currently running Multi-Role Combat Vehicles Technology Demonstration Project (MRCV TDP) within Defence Research and Development Canada.  This project focuses on future requirements for the Canadian Light Armoured Vehicle fleet and is directly linked to major capital acquisition programs funded by the Canadian Forces.  The first phases of MRCV TDP encompassed a series of international distributed simulation experiments using high-fidelity ground AV simulators, with highly configurable operator-machine interfaces.  These AV simulators were linked to virtual and constructive ground and air asset simulations, from both Canadian and U.S. research facilities.  This paper examines soldier-centered interface design, and recommends an interface development process for use in future military vehicle procurement programs – from a Project Manager’s point-of-view.  This process of soldier-centered interface design is crucial for the future implementation of AVs, allowing our soldiers to effectively control these complex systems, strike with precision, and have the …

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

 

Profiling is Politically ‘Correct’: Agent-Based Modeling of Ethno-Political Conflict

2007 Paper No. 7015

 

Barry G. Silverman, Gnana K. Bharathy, Benjamin Nye 

Electrical and Systems Engineering Department University of Pennsylvania,

Philadelphia, PA

 

A holy grail for military, diplomatic, and intelligence analysis is a valid set of software agent models that act as the desired ethno-political factions so that one can test the effects that may arise from alternative courses of action in different lands. This article enumerates the challenges of such a testbed and describes best-of-breed leader and follower profiling models implemented to improve the realism and validity of the agent. Realistic, ‘descriptive’ agents are contrasted to rational actor theory in terms of the different equilibria one would expect to emerge in conflict games. These predictions are examined in two real world cases (Iraq and SE Asia) where the agent models are subjected to validity tests and a policy experiment is then run. We conclude by arguing that substantial effort on game realism, best-of-breed social science models, and agent validation efforts is essential if analytic experiments are to effectively explore conflicts and alternative ways to influence outcomes. Such efforts are likely to improve behavioral game theory as well. 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Modeling Human Perception of Situation Awareness During Constructive Experimentation

2007 Paper No. 7448

 

Philip Colon

M Toyon Research Corporation

Goleta, CA

John Tran, Ke-Thia Yao

Informaton Sciences Institute

Marina del Rey, CA

 

Michael Anhalt, Jacqueline M. Curiel

Alion Science and Technology

Marina del Rey, El Cajon, CA

 

Highly advanced sensor technologies give our military commanders a significant command and control (C2) advantage over our enemies during conflicts, particularly with respect to situation awareness (SA). The use of advanced sensor technology models in synthetic battlespace gives war fighters parallel advantages. Two accepted simulation methodologies for analyzing the impact of sensor technologies are through Human-in-the-Loop (HITL) experiments, such as Joint Urban Operations (JUO), which utilize sensor capabilities to assist human participants during the experiments, and Monte Carlo Constructive (MCC) simulations, which can be used to model human performance. In HITL experiments using Joint Semi-Automated Forces (JSAF), participants describe their SA using Situation Awareness Objects (SAOs) which then can be reconstructed using Endsley’s (1995) three levels of SA (perception, comprehension, and prediction). MCC experiments, which are dominated by algorithmically determined behaviors, can be used to model SA. Sensor measurements currently can be fused to perceive individual entities, but do not have the capability to recognize groupings of entities, resulting only in partial perceptual SA. Furthermore, current sensor data fusion models do not produce the second and third levels of SA, comprehension and prediction.

This paper will report research efforts to utilize both methodologies to expand the use of SAOs beyond player declarations to the automatic generation of SAOs.  We develop a method to organize events drawn from…

 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

Automated Scenario-Based Training Management: Exploring the Possibilities

2007 Paper No. 7233

 

Bruce Perrin, Barbara Buck, Brandt Dargue, Elizabeth Biddle, Troy Stull, Curtis Armstrong

The Boeing Company

St. Louis, MO USA

 

Despite the prevalence of software applications that exploit user information to individualize the experience, personalized training systems are still relatively rare.  This paper describes changes in technology and standards that may alter this trend.  Utilizing these advances, we have developed a standards-based learner model that is updated dynamically during training and that controls content sequencing.  We have established the impact of this technology on learning through training effectiveness research.  With this core learner modeling capability established, we have subsequently started exploratory studies into ways it might be used to manage scenario-based, simulation training.  Specifically, we describe two prototype systems that use this core modeling capability, but that use the information it provides in distinctly different ways.  Because of the complexity of simulation training, the root cause of performance issues is seldom apparent.  The first prototype addresses this issue by using the learner model to select follow-on scenarios that help to build skill while distinguishing among competing learning needs hypotheses.  The second prototype addresses the issue of maximizing learning opportunities within a scenario.  It uses the core learner model to modify a scenario during execution in order to provide additional opportunities to achieve specific learning objectives or to adjust the challenge of an exercise.  Directions for future research for both efforts are described.   

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

 

A Quantitative Measurement of Presence in Flight Simulators

2007 Paper No. 7031

 

Javier (Jeff) Covelli

Northrop Grumman Corporation (NGC) CREOL

Orlando, Florida

Peter Hancock

Institute for Simulation & Training (IST)

Orlando, Florida

 

Jannick Rolland

College of Optics and Photonics, University of Central Florida (UCF)

Orlando, Florida

 

Virtual Reality (VR) and Augmented Reality (AR) Head Worn Display (HWD) technology is being considered as a novel alternative for low cost, wide Field of Regard (FOR), deployable simulators.  There are inherent differences in display characteristics among different HWDs and between HWDs in general with conventional displays.  For example, the effective Field of View (FOV) in most HWDs is no more than 60º horizontal and 45º vertical, which is far narrower than the human eye’s 200º horizontal and 135º vertical FOV.  Developing a HWD with a wide 200º horizontal FOV is expensive.  Current HWD flight simulator implementations provide limited effective FOV that reduces the pilot’s visual stimulus, perception, sense of presence and overall training effectiveness.  To successfully utilize a VR or AR HWD in a simulator, we hypothesize the user must have the same or even a higher mental immersion experience as compared with the conventional simulator experience with unrestricted FOV. Attempts to measure mental immersion or presence from VR simulations were normally conducted with questionnaires.  Although more convenient, the validity of measuring the continuous experience of presence with post experience questionnaires has been challenged (e.g., Slater, 2004). Here, a quantitative approach to measure presence in relation to mental image processing and performance is proposed.  This paper presents experimental methods involving measurement and analysis of normal head and eye movement patterns of experienced pilots while accomplishing specific tasks in a conventional flight simulator with a 170° horizontal x 75° vertical FOR.  The paper outlines metrics taken using head and eye tracking equipment, and results of pilot…

 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.

 

The Appropriate Loudness for Flight Simulator Voice Communications

2007 Paper No. 7150

 

Nathaniel M. Napoletano, Thomas Babyak

Lockheed Martin, STS

Akron, Ohio

 

Voice communications and warning tones are vital elements of virtually every flight-training scenario. Before and during missions, instructors and pilots perform coordination intercommunications and simulated radio communications across digital networks between flight stations and other training assets. These networks may connect facilities around the globe during U.S. Air Force (USAF) distributed mission training.  In the requirements analysis phase leading to the development of each training device, questions arise about how loud the voice communications and warning tones should be played in the instructor’s and pilot’s headsets. The problem is more involved than matching the levels used in the equipment being simulated because of several complicating factors. These factors include the facts that the pilots in the field often wear hearing protection (earplugs) under their helmets, the ambient noise is lower in the simulator and the training participants may be using a variety of equipment.  This study also takes hearing safety concerns related to noise induced sensorineural hearing loss into account. The lower criticality of voice intelligibility in a simulated mission compared to combat alters the balance of factors considered while driving audio safety specifications toward the more conservative.  This paper defines a common set of audio alignment parameters and methods for use by the networked fighter trainer community. There is little existing published data establishing optimal loudness levels particular to radio network communications taxonomies used in flight simulators. In order to develop these parameters the authors combined publicly available data with their new study: “Acoustical Comfort Levels in an Immersive Training Environment.” These methods and parameters are now being used to align pilot and instructor voice communications equipment at our facility and customer sites.

 

 

This paper is available on the 2007 I/ITSEC CD ROM.

Order the paper from I/ITSEC’s Website.