No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Snap: A time critical decision-making framework for MOUT simulations
Abstract
Deliberative reasoning based on the rational analysis of various alternatives often requires too much information and may be too slow in time critical situations. In these situations, humans rely mainly on their intuitions rather than some structured decision-making processes. An important and challenging problem in Military Operations on Urban Terrain (MOUT) simulations is how to generate realistic tactical behaviors for the non-player characters (also known as bots), as these bots often need to make quick decisions in time-critical and uncertain situations. In this paper, we describe our work on Snap, a time critical decision-making framework for the bots in MOUT simulations. The novel features of Snap include case-based reasoning (CBR) and thin slicing. CBR is used to make quick decisions by comparing the current situation with past experience cases. Thin slicing is used to model human's ability to quickly form up situation awareness under uncertain and complex situations using key cues from partial information. To assess the effectiveness of Snap, we have integrated it into Twilight City, a virtual environment for MOUT simulations. Experimental results show that Snap is very effective in generating quick decisions during time critical situations for MOUT simulations. Copyright © 2008 John Wiley & Sons, Ltd.
... RPD is believed to be a more realistic account of how human make decisions in everyday situations (particularly, time-critical situations), which emphasizes the role of experiences and knowledge in decision making. Recently, RPD based models have been used in human-agent collaboration [Fan et al. 2005] and military simulations [Ting and Zhou 2008]. ...
As a collective and highly dynamic social group, the human crowd is a fascinating phenomenon that has been frequently studied by experts from various areas. Recently, computer-based modeling and simulation technologies have emerged to support investigation of the dynamics of crowds, such as a crowd’s behaviors under normal and emergent situations. This article assesses the major existing technologies for crowd modeling and simulation.We first propose a two-dimensional categorization mechanism to classify existing work depending on the size of crowds and the time-scale of the crowd
phenomena of interest. Four evaluation criteria have also been introduced to evaluate existing crowd simulation systems from the point of view of both a modeler and an end-user.
We have discussed some influential existing work in crowd modeling and simulation regarding their major features, performance as well as the technologies used in this work. We have also discussed some open problems in the area. This article will provide the researchers with useful information and insights on the state of the art of the technologies in crowd modeling and simulation as well as future research directions.
Traditional methods for constructing crowd simulations often have shortcomings in terms of realism, and data-driven methods are an effective approach to enhancing the visual realism of crowd simulation. However, existing work mainly constructs crowd simulations through prediction-based approaches based on deep learning or by fitting the parameters of traditional methods, which limits the expressiveness of the model. In response to these limitations, this paper introduces a method capable of generating realistic pedestrian crowds. This approach uses a Generative Adversarial Network, complemented by a transformer module, to learn behavioral patterns from actual crowd trajectories. We use a transformer module to extract trajectory features of the crowd, then convert the spatial relationships between individuals into sequences using a special data processing mechanism, and use the transformer module to extract social features of the crowd, while guiding the movement of each individual with their target direction. During training, we simultaneously learn from real crowd data and simulation data resolving collisions by traditional methods, to enhance the collision avoidance behavior of virtual crowds while maintaining the movement patterns of real crowds, resulting in more general collision avoidance behavior. The crowds generated by the model are not limited to specific scenarios and show generalization capabilities. Compared to other models, our method shows better performance on publicly available large-scale pedestrian datasets after training. Our code is publicly available at https://github.com/ydp91/NPCGAN.
Data-driven crowd modeling has now become a popular and effective approach for generating realistic crowd simulation and has been applied to a range of applications, such as anomaly detection and game design. In the past decades, a number of data-driven crowd modeling techniques have been proposed, providing many options for people to generate virtual crowd simulation. This article provides a comprehensive survey of these state-of-the-art data-driven modeling techniques. We first describe the commonly used datasets for crowd modeling. Then, we categorize and discuss the state-of-the-art data-driven crowd modeling methods. After that, data-driven crowd model validation techniques are discussed. Finally, six promising future research topics of data-driven crowd modeling are discussed.
Multiagent time-critical dynamic decision making is a challenging task in many real-world applications where a trade-off between solution quality and computational tractability is required. In this paper, we present a formal representation for modelling time-critical multiagent dynamic decision problems based on interactive dynamic influence diagrams (I-DIDs). The new representation called time-critical I-DIDs (TC-IDIDs) represents space-temporal abstraction by providing time-index to nodes and the model is defined in terms of the condensed and deployed forms. The condensed form is a static model of TC-IDIDs and can be expanded into its dynamic version. To facilitate the conversion between the two forms, we exploit the notion of object-orientation design to develop flexible and reusable TC-IDIDs. The difficulty on expanding TC-IDIDs is to select a proper time sequence to index nodes in the condensed form so that the expanded TC-IDIDs can be solved efficiently without compromising the quality of the policy. For this purpose, we propose two methods to build the condensed form of TC-IDIDs. We evaluate the solution quality and time complexity in three well-studied problems and provide results in support.
Situation awareness is the perception of environmental elements within a volume of time and space, the comprehension of their meaning, and the projection of their status in the near future. The quality of situation awareness directly affects the decision making process for human soldiers in Military Operations on Urban Terrain (MOUT). It is therefore important to accurately model situation awareness in order to generate realistic tactical behaviors for the non-player characters (also known as bots) in MOUT simulations. This is a very challenging problem due to the time constraints in decision-making process and the heterogeneous cue types involved in MOUT. Although there are some theoretical models on situation awareness, they generally do not provide computational mechanisms suitable for MOUT simulations. In this paper, we propose a computational model of situation awareness for the bots in MOUT simulations. The computational model aims to form up situation awareness quickly with some key cues of the tactical situation. It is also designed to work together with some novel features that help to produce realistic tactical behaviors. These features include case-based reasoning, qualitative spatial representation and expectations. The effectiveness of the computational model is assessed with Twilight City, a virtual environment that we have built for MOUT simulations.
Generating realistic behaviors for the non-player characters (also known as bots) is an important task for Military Operations on Urban Terrain (MOUT) simulations. Most of the current bots reasoning models are based on complete rational analysis of all alternatives. However, such models may not be adequate in time critical and uncertain situations. In such situations, due to time constraints and incomplete information, humans rely mainly on experience rather than some structured analysis of the given situation to make decisions. In our previous work, we have developed SNAP, a time critical decision-making framework which aims to imitate human decision-making processes for MOUT simulations. A major limitation that we found about SNAP is its difficulty in dealing with dynamic changes in MOUT simulations. In this paper, we propose to address this problem by incorporating a novel feature of expectations into SNAP. The effectiveness of this new feature is assessed with Twilight City, a virtual environment for MOUT simulations. Experimental results show that our expectation model works well in dealing with dynamic changes in MOUT. Copyright © 2009 John Wiley & Sons, Ltd. Generating realistic behaviors for the non-player characters (also known as bots) is an important task for Military Operations on Urban Terrain (MOUT) simulations. In such situations, due to time constraints and incomplete information, humans rely mainly on experience rather than some structured analysis of the given situation to make decisions. To address this issue, we had incorporated expectations in SNAP, a time critical decision-making framework which aims to imitate human decision-making processes for MOUT simulations.
The quality of situation awareness directly affects the decision making process for human soldiers in Military Operations
on Urban Terrain (MOUT). It is important to accurately model situation awareness to generate realistic tactical behaviors
for the non-player characters (also known as bots) in MOUT simulations. This is a very challenging problem due to the time
constraints and the heterogeneous cue types in MOUT. Although there are some theoretical models on situation awareness, they
generally do not provide computational mechanisms suitable for MOUT simulations. In this paper, we propose a computational
model of situation awareness for the bots in MOUT simulations. The model forms up situation awareness quickly with key cues.
It is also designed to work with some novel features. They include case-based reasoning, qualitative spatial representation and expectations. The effectiveness of the computational model is assessed with Twilight City, a virtual environment that we have built for MOUT simulations.
Military Operations in Urban Terrain (MOUT) create unique cognitive demands for small unit leaders. Years of experience are typically needed to master these demands. However, most platoon leaders tend to have limited experience in Army operations generally, and in MOUT specifically. Developed from in-depth interviews with veterans having MOUT combat experience, this report provides a comprehensive and detailed overview of the decisions platoon leaders must make during five stages of building clearing: secure perimeter, approach building, enter building, clear building and maintain security, and evacuate building. The report also discusses five general skills that are required throughout building clearing missions: think like the enemy, maintain situational awareness and the big picture, project into the future, apply rules of engagement, and lead subordinates.
We describe our work on Twilight City, a virtual training environment for Military Operations on Urbanized Terrain (MOUT). MOUT are often characterized by building-to-building, room-to-room, and person-to-person close combat, therefore the virtual training environment for MOUT requires high fidelity in 3D graphics and realistic behaviors of various objects. Twilight City is constructed on a commercial multi-player first-person shooter (FPS) game engine, the Unreal Tournament (UT). Various modifications to the UT game engine have been made to incorporate various aspects in real MOUT scenarios. Case studies show that the AI bots are able to demonstrate some human-like tactical behavior. Human factor tests are conducted to evaluate the performance of Twilight City. The results demonstrate the effectiveness of Twilight City as an urban warfare simulation environment. The lessons learned from these experiments are also discussed which provide directions for future improvement.
Capturing and encoding human variability is an increasingly important issue in human behavior models (HBMs) for military simulations. In this paper, we define variability, describe its importance in several military simulation applications, examine the sources of variability in human behavior, and outline an approach to variability that will allow us to explore approaches to low-cost, realistic variability in human behavior models. Capturing and encoding human variability is an increasingly important issue in human behavior models (HBMs) for military simulations. During the development of TacAir-Soar [1], a detailed model of expert human pilots flying tactical fixed-wing air missions, we faced contradictory requirements with respect to variability. Variability seemed at odds with validation requirements (how could the system be validated if it had any variability at all?) but yet necessary for realistic models (how could we possibly claim to model human behavior without variability?). For TacAir-Soar, we minimized the deliberate introduction of variability, mostly because it was sufficiently challenging to generate correct behavior. Significant variability did emerge in TacAir-Soar from the interaction between a complex environment (other computer and human controlled entities) and the complex knowledge encoded in TacAir-Soar; observers did not feel that the behavior was overly rote or predictable. Thus, we avoided modeling variability explicitly, which would have complicated validation, but realized adequate variability through rich interactions with the environment. We are currently exploring how to create HBMs that exhibit realistic variability in M&S systems, derived from our experiences in developing TacAir-Soar and the VIRTE MOUT adversaries. In this paper, we define variability, examine sources of variability in human behavior, and identify constraints and issues that arise when attempting to model these sources of variability in HBMs. We introduce potential sources of variability in computer-based models of human behavior and describe how they map to the sources of variability in human behavior. We also describe a "strawman" proposal for architecturally supporting realistic variability. We are implementing this proposal in order to explore these questions empirically.
From a general cognitive perspective, decision making is the process of selecting a choice or course of action from a set of alternatives. A large number of time critical decision making models have been developed over the course of several decades. This paper reviews both the underlying cognitive processes and several decision making models. In the first section, we briefly describe the primary underlying cognitive processes and issues that are common to most, if not all, decision making models, with a focus on attention, working memory, and reasoning. The second section reviews several of the most prominent high-level models of decision making, especially those developed for military contexts
El original está disponible en www.journals.cambridge.org Case-based reasoning (CBR) is an approach to problem solving that emphasizes the role of prior experience during future problem solving (i.e., new problems are solved by reusing and if necessary adapting the solutions to similar problems that were solved in the past). It has enjoyed considerable success in a wide variety of problem solving tasks and domains. Following a brief overview of the traditional problem-solving cycle in CBR, we examine the cognitive science foundations of CBR and its relationship to analogical reasoning. We then review a representative selection of CBR research in the past few decades on aspects of retrieval, reuse, revision, and retention. Peer reviewed
Anyone who watches the television news has seen images of firefighters rescuing people from burning buildings and paramedics treating bombing victims. How do these individuals make the split-second decisions that save lives? Most studies of decision making, based on artificial tasks assigned in laboratory settings, view people as biased and unskilled. Gary Klein is one of the developers of the naturalistic decision-making approach, which views people as inherently skilled and experienced. Since 1985, Klein has conducted fieldwork to find out how people tackle challenges in difficult, nonroutine situations. Sources of Power is based on observations of humans acting under such real-life constraints as time pressure, high stakes, personal responsibility, and shifting conditions. In addition to providing information that can be used by professionals in management, psychology, engineering, and other fields, the book presents an overview of the research approach of naturalistic decision making and expands our knowledge of the strengths people bring to difficult tasks.
This paper describes requirements for synthetic adversaries for urban combat training and MOUTBots, a prototype application. The MOUTBots use a commercial computer game to define, implement and test basic behavior representation requirements and the Soar architecture as the engine for knowledge representation and execution. We describe how these components aided the development of the prototype and present an initial evaluation against competence, taskability, fidelity, variability, transparency, and efficiency requirements.
The vicious urban battle for the insurgent-controlled city of Fallujah in November 2004 was a turning point in the ongoing counterinsurgency campaign in Iraq. It demonstrated the resolve of the Iraqi government to fight terrorists domestically, using both multinational and Iraqi forces, and its results included a returning population willing to vote in national elections held in January 2005. Ballard tells the story of the Fallujah campaign, beginning with the horrific deaths of the American Blackwater contractors in March 2004, and continuing through the battle, the painstaking reconstruction of the city, and the precedent-setting elections that followed. Based on first-person accounts, interviews, and official documents, this book gives readers rare insight into the significant actions and innovative techniques of the year-long fight for the city.
Opening with a historical overview of the initial crisis in Fallujah and the similar coalition battle in Najaf, the book includes a detailed account of the planning and execution of the operation to retake the city. Finally, it describes the political and military lessons proven in Fallujah, including coalition force integration, information operations, urban combat techniques, interagency coordination and innovative reconstruction procedures. This is the story of real combat in Iraq—told in a way every American should understand.
Malcolm Gladwell; род. 3 сентября 1963, Хэмпшир) — канадский журналист, поп-социолог. В 2005 году «Time» назвало Малкольма Гладуэлла одним из 100 самых влиятельных людей. Книги и статьи Малкольма часто касаются неожиданных последствий исследований в социальных науках и находят широкое применение в научной работе, в частности в областях социологии, психологии и социальной психологии. Некоторые из его книг занимали первые строки в списке бестселлеров «The New York Times». В 2007 году Малкольм получил первую премию Американской Социологической ассоциации за выдающиеся достижения по отчетам в социальных вопросах. В 2007 году он также получил почетную степень доктора филологии Университета Ватерлоо. Малькольм Гладуелл описывает эксперименты, которые показывают, что человеку с поврежденными эмоциональными центрами крайне трудно принимать решения. Он рассказывает про одного такого пациента, которому было предложено прийти на прием либо во вторник, либо в пятницу. И пациент два часа решал во вторник ему прийти или в пятницу — в столбик выписывал плюсы и минусы, их сравнивал, группировал по разному, всяко переставлял. И в жизни своих домашних он просто убивал вот этим. Если его спрашивали, ты что хочешь: омлет или салат? — это задача минут на сорок. Обычный человек очень просто поступает. Он видит омлет, что-то чувствует и говорит: Хочу! Все. Выбор сделан легко и быстро.
MOUT is a complex operation. Effective combat units are able to identify the high payoff tasks required to accomplish their mission, establish solid SOP's, and train to standard. Movement between buildings is where the majority of casualties occur in the village of Shughart-Gordon. This problem is a direct result of a lack of understanding of the nature of MOUT and lack of training emphasis on the specific collective tasks required. Units that emphasize movement between buildings and achieve a level of proficiency at this task will be attacking the source of the single greatest cause of casualties in MOUT. To do so requires us to understand how the enemy fights, focus on the relationship between fire and movement in cities, and the requirement to maintain continuous security as well as a sense of tempo and tactical patience. If we are successful here we will be able to minimize casualties and set up our squads for success as they close with the enemy. We will then be well on our way to accomplishing the mission.
The Battlefield Information Questionnaire was developed to measure the relative importance of 88 different types of battlefield information to squad leaders in four combat situations: planning, assaulting an objective, consolidating and reorganizing on the objective, and defending the objective from counterattack. This questionnaire was then administered to a group of 106 non-commissioned officers, each having combat experience as a squad leader. The types of information they most wanted to know, and least wanted to know, were highly consistent across combat situations. Averaged across situations, the ten types of information most important to squad leaders in battle were: location of threat personnel, vehicles, and weaponry; casualty collection point location; ammunition remaining; location of personnel in their squad; location of units in contact with the enemy; personnel location in adjacent friendly units; their own location relative to other personnel; location of mines, obstacles, booby traps, and improvised explosive devices; availability of supporting fires; and direction of movement for enemy personnel.
A study of memory for places was carried out to examine five hypotheses about the use of schemata in memory performance: (a) that schemata determine what objects are encoded into memory; (b) that schemata act as frameworks for episodic information; (c) that schema-based information is integrated with episodic information; (d) that schemata facilitate retrieval; and (e) that schemata influence what is communicated at recall. Subjects were taken into what they thought was a graduate student's office and later were tested for memory of the room with either drawing recall, written recall, or verbal recognition. Memory scores for objects were correlated with schema expectancy and saliency ratings. Schema expectancy was positively correlated with recall and recognition. Expected objects were inferred in recall, supporting the integration hypothesis. Comparison of recall and recognition data supported the retrieval hypothesis. Analysis of the written descriptions supported the communication hypothesis. Saliency was positively correlated with recall and recognition for present objects, but was unrelated to retrieval. Saliency was negatively correlated with recognition for nonpresent objects, suggesting a metacognitive strategy in recognition of high-salient objects.
John Boyd is often known exclusively for the so-called 'OODA' loop model he developed. This model refers to a decision-making process and to the idea that military victory goes to the side that can complete the cycle from observation to action the fastest. This book aims to redress this state of affairs and re-examines John Boyd's original contribution to strategic theory. By highlighting diverse sources that shaped Boyd's thinking, and by offering a comprehensive overview of Boyd's work, this volume demonstrates that the common interpretation of the meaning of Boyd's OODA loop concept is incomplete. It also shows that Boyd's work is much more comprehensive, richer and deeper than is generally thought. With his ideas featuring in the literature on Network Centric Warfare, a key element of the US and NATO's so-called 'military transformation' programmes, as well as in the debate on Fourth Generation Warfare, Boyd continues to exert a strong influence on Western military thinking. Dr Osinga demonstrates how Boyd's work can helps us to understand the new strategic threats in the post- 9/11 world, and establishes why John Boyd should be regarded as one of the most important (post)modern strategic theorists.
This paper describes Gamebots, a multi-agent system infrastructure derived from an Internet-based multi-player video game called Unreal Tournament. The game allows characters to be controlled over client-server network connections by feeding sensory information to client players (humans and agents). Unlike other standard test-beds, the Gamebots domain allows both human players and agents, or bots, to play simultaneously; thus providing the opportunity to study human team behavior and to construct agents that play collaboratively with humans. The Gamebots system provides a built-in scripting language giving interested researchers the ability to create their own multi-agent tasks and environments for the simulation. Keywords Multiple Agent Systems, Agent Infrastructure, Multi-Agent Teamwork, Human-Agent Interaction, Virtual World Simulation, Video Game Bot. 1.
Panic in The Military
- Schultzdp
Schultz DP. Panic in The Military. Technical Report, Uni-versity of North Carolina, Storming Media, January, 1971.
Qualitative Physics for MOUT, 39th Annual Simulation Symposium
- S P Zhou
- S P Ting
Zhou SP, Ting SP. Qualitative Physics for MOUT, 39th Annual Simulation Symposium, Huntsville, AL, USA, April 2–6, 2006.
Fighting For Fallujah. A New Dawn for Iraq
- Jr Ballard
Ballard JR. Fighting For Fallujah. A New Dawn for Iraq. Greenwood Publishing Group: Connecticut, USA, 2006.
Official History of the Canadian Army in the Second World War Volume III. Queen's Printer: Ottawa
- Cp Stacey
Stacey CP. Official History of the Canadian Army in the Second World War Volume III. Queen's Printer: Ottawa, 1960.
Also pub-lished in AI Magazine
- Wray Re
- Je Laird
- A Nuxoll
- D Stokes
- A Kerfoot
Wray RE, Laird JE, Nuxoll A, Stokes D, Kerfoot A. Synthetic Adversaries for Urban Combat Training In Proceedings of the 2004 Innovative Applications of Artificial Intelligence Con-ference, San Jose, CA, July 2004. AAAI Press. Also pub-lished in AI Magazine 2005, 26(3): 82–92.
Schema theory: An introduction
- Widmayersa
Widmayer SA. Schema theory: An introduction. http:// chd.gse.gmu.edu/immersion/knowledgebase/strategies/ cognitivism/SchemaTheory.htm
JavaBot for Unreal Tournament
- A N Marshall
Marshall AN. JavaBot for Unreal Tournament, 2003; http://
utbot.sourceforge.net
Science of Emotion: What People Believe, What Evidence Shows and Where to Go From Here. Human Behavior in Military Contexts
- Lf Barrett
Barrett LF. Science of Emotion: What People Believe, What
Evidence Shows and Where to Go From Here. Human Behavior
in Military Contexts. The National Academies Press:
Washington, USA, 2007.
The Power of Think without Thinking. Little, Brown and Co
- M Gladwell
- Blink
Gladwell M. Blink: The Power of Think without Thinking.
Little, Brown and Co.: New York, USA, 2005.
- McDermott
A 3D virtual world test bed for multiagent research
- R Adobbati
- A N Marshall
- A Scholer
- S Tejada
- Gamebots
- Evans K
Variability in Human Behavior Modeling for Military Simulations
- Wrayre Lairdje
GameBots: A 3D virtual world test bed for multiagent research
- Adobbatir Marshallan
- Scholera Tejadas
A Review of Time Critical Decision Making Models and Human Cognitive Processes
- Furmanskic Azumar Dailym