Publications
Search
1.
Han, Zicheng; Becerik-Gerber, Burcin; Liu, Ruying; Lucas, Gale; Doolab, Yogin; Zein, Sara; Atkins, Margo; Blass, Deborah; Miles, Lachlan
Emergency dynamics modeling with an agent-based framework: Active shooter scenario case study Journal Article
In: International Journal of Disaster Risk Reduction, vol. 142, pp. 106226, 2026, ISSN: 22124209.
@article{han_emergency_2026,
title = {Emergency dynamics modeling with an agent-based framework: Active shooter scenario case study},
author = {Zicheng Han and Burcin Becerik-Gerber and Ruying Liu and Gale Lucas and Yogin Doolab and Sara Zein and Margo Atkins and Deborah Blass and Lachlan Miles},
url = {https://linkinghub.elsevier.com/retrieve/pii/S2212420926002384},
doi = {10.1016/j.ijdrr.2026.106226},
issn = {22124209},
year = {2026},
date = {2026-05-01},
urldate = {2026-06-16},
journal = {International Journal of Disaster Risk Reduction},
volume = {142},
pages = {106226},
abstract = {Modeling human-induced emergencies requires a structured representation of interactions among occupants, hazards, and built environments. While existing simulation approaches provide important insights, behavioral processes, hazard progression, and architectural conditions are often examined separately. This study develops a parameterized agent-based modeling framework that integrates heterogeneous occupant behaviors, hazard dynamics, and environmental configurations within a commercially available crowd-simulation platform to extend its usability within the academic and professional communities. The framework comprises environmental modeling, agent definition, interaction logic, risk evaluation, and simulation execution, enabling systematic configuration of behavioral rules and scenario parameters through customized scripting. An Active Shooter Incident (ASI) in a school building is simulated as a demonstrative case study to illustrate the framework's capability to configure and test emergency scenarios under different behavioral, hazard and environmental assumptions. Using the framework, the study examines the effects of classroom occupancy, door width, and door-locking mechanisms on occupant safety, evaluated using a token-based high-risk metric and factorial statistical analysis. Results indicate that higher occupancy increases exposure risk, timely door locking substantially improves safety outcomes, and the influence of door width is context-dependent. The consistency of these patterns with prior research supports the behavioral plausibility of the modeling approach. Overall, the study contributes a practical and extensible framework for operationalizing human–hazard–environment interactions in similar agent-based simulations and for generating scenario-specific insights to support design and preparedness strategies in human-induced emergency contexts.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Modeling human-induced emergencies requires a structured representation of interactions among occupants, hazards, and built environments. While existing simulation approaches provide important insights, behavioral processes, hazard progression, and architectural conditions are often examined separately. This study develops a parameterized agent-based modeling framework that integrates heterogeneous occupant behaviors, hazard dynamics, and environmental configurations within a commercially available crowd-simulation platform to extend its usability within the academic and professional communities. The framework comprises environmental modeling, agent definition, interaction logic, risk evaluation, and simulation execution, enabling systematic configuration of behavioral rules and scenario parameters through customized scripting. An Active Shooter Incident (ASI) in a school building is simulated as a demonstrative case study to illustrate the framework's capability to configure and test emergency scenarios under different behavioral, hazard and environmental assumptions. Using the framework, the study examines the effects of classroom occupancy, door width, and door-locking mechanisms on occupant safety, evaluated using a token-based high-risk metric and factorial statistical analysis. Results indicate that higher occupancy increases exposure risk, timely door locking substantially improves safety outcomes, and the influence of door width is context-dependent. The consistency of these patterns with prior research supports the behavioral plausibility of the modeling approach. Overall, the study contributes a practical and extensible framework for operationalizing human–hazard–environment interactions in similar agent-based simulations and for generating scenario-specific insights to support design and preparedness strategies in human-induced emergency contexts.
2.
Kwon, Deuksin; Shrestha, Kaleen; Han, Bin; Lin, Spencer; Hale, James; Gratch, Jonathan; Matarić, Maja; Lucas, Gale M.
Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution Miscellaneous
2026, (arXiv:2602.07414 [cs.AI]).
@misc{kwon_can_2026,
title = {Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution},
author = {Deuksin Kwon and Kaleen Shrestha and Bin Han and Spencer Lin and James Hale and Jonathan Gratch and Maja Matarić and Gale M. Lucas},
url = {http://arxiv.org/abs/2602.07414},
doi = {10.48550/arXiv.2602.07414},
year = {2026},
date = {2026-02-01},
urldate = {2026-06-18},
publisher = {arXiv},
abstract = {Large language models (LLMs) are increasingly used to simulate human behavior in social settings such as legal mediation, negotiation, and dispute resolution. However, it remains unclear whether these simulations reproduce the personality-behavior patterns observed in humans. Human personality, for instance, shapes how individuals navigate social interactions, including strategic choices and behaviors in emotionally charged interactions. This raises the question: Can LLMs, when prompted with personality traits, reproduce personality-driven differences in human conflict behavior? To explore this, we introduce an evaluation framework that enables direct comparison of human-human and LLM-LLM behaviors in dispute resolution dialogues with respect to Big Five Inventory (BFI) personality traits. This framework provides a set of interpretable metrics related to strategic behavior and conflict outcomes. We additionally contribute a novel dataset creation methodology for LLM dispute resolution dialogues with matched scenarios and personality traits with respect to human conversations. Finally, we demonstrate the use of our evaluation framework with three contemporary closed-source LLMs and show significant divergences in how personality manifests in conflict across different LLMs compared to human data, challenging the assumption that personality-prompted agents can serve as reliable behavioral proxies in socially impactful applications. Our work highlights the need for psychological grounding and validation in AI simulations before real-world use.},
note = {arXiv:2602.07414 [cs.AI]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Large language models (LLMs) are increasingly used to simulate human behavior in social settings such as legal mediation, negotiation, and dispute resolution. However, it remains unclear whether these simulations reproduce the personality-behavior patterns observed in humans. Human personality, for instance, shapes how individuals navigate social interactions, including strategic choices and behaviors in emotionally charged interactions. This raises the question: Can LLMs, when prompted with personality traits, reproduce personality-driven differences in human conflict behavior? To explore this, we introduce an evaluation framework that enables direct comparison of human-human and LLM-LLM behaviors in dispute resolution dialogues with respect to Big Five Inventory (BFI) personality traits. This framework provides a set of interpretable metrics related to strategic behavior and conflict outcomes. We additionally contribute a novel dataset creation methodology for LLM dispute resolution dialogues with matched scenarios and personality traits with respect to human conversations. Finally, we demonstrate the use of our evaluation framework with three contemporary closed-source LLMs and show significant divergences in how personality manifests in conflict across different LLMs compared to human data, challenging the assumption that personality-prompted agents can serve as reliable behavioral proxies in socially impactful applications. Our work highlights the need for psychological grounding and validation in AI simulations before real-world use.
3.
Zhang, Tianyi; Zhou, Xiaolin; Wang, Yunzhe; Cambria, Erik; Traum, David; Mao, Rui
Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods Miscellaneous
2025, (arXiv:2510.20252 [cs.AI]).
@misc{zhang_individualized_2025,
title = {Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods},
author = {Tianyi Zhang and Xiaolin Zhou and Yunzhe Wang and Erik Cambria and David Traum and Rui Mao},
url = {http://arxiv.org/abs/2510.20252},
doi = {10.48550/arXiv.2510.20252},
year = {2025},
date = {2025-10-01},
urldate = {2026-07-06},
publisher = {arXiv},
abstract = {Individualized cognitive simulation (ICS) aims to build computational models that approximate the thought processes of specific individuals. While large language models (LLMs) convincingly mimic surface-level human behavior such as role-play, their ability to simulate deeper individualized cognitive processes remains poorly understood. To address this gap, we introduce a novel task that evaluates different cognitive representation methods in ICS. We construct a dataset from recently published novels (later than the release date of the tested LLMs) and propose an 11-condition cognitive evaluation framework to benchmark seven off-the-shelf LLMs in the context of authorial style emulation. We hypothesize that effective cognitive representations can help LLMs generate storytelling that better mirrors the original author. Thus, we test different cognitive representations, e.g., linguistic features, concept mappings, and profile-based information. Results show that combining conceptual and linguistic features is particularly effective in ICS, outperforming static profile-based cues in overall evaluation. Importantly, LLMs are more effective at mimicking linguistic style than narrative structure, underscoring their limits in deeper cognitive simulation. These findings provide a foundation for developing AI systems that adapt to individual ways of thinking and expression, advancing more personalized and human-aligned creative technologies.},
note = {arXiv:2510.20252 [cs.AI]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Individualized cognitive simulation (ICS) aims to build computational models that approximate the thought processes of specific individuals. While large language models (LLMs) convincingly mimic surface-level human behavior such as role-play, their ability to simulate deeper individualized cognitive processes remains poorly understood. To address this gap, we introduce a novel task that evaluates different cognitive representation methods in ICS. We construct a dataset from recently published novels (later than the release date of the tested LLMs) and propose an 11-condition cognitive evaluation framework to benchmark seven off-the-shelf LLMs in the context of authorial style emulation. We hypothesize that effective cognitive representations can help LLMs generate storytelling that better mirrors the original author. Thus, we test different cognitive representations, e.g., linguistic features, concept mappings, and profile-based information. Results show that combining conceptual and linguistic features is particularly effective in ICS, outperforming static profile-based cues in overall evaluation. Importantly, LLMs are more effective at mimicking linguistic style than narrative structure, underscoring their limits in deeper cognitive simulation. These findings provide a foundation for developing AI systems that adapt to individual ways of thinking and expression, advancing more personalized and human-aligned creative technologies.
4.
Liu, Ruying; Wu, Wanjing; Becerik-Gerber, Burcin; Lucas, Gale M.; Laboy, Michelle; Fannon, David
Reinforcement learning for evaluating school safety designs in active shooter incidents Journal Article
In: Advanced Engineering Informatics, vol. 67, pp. 103575, 2025, ISSN: 14740346.
@article{liu_reinforcement_2025,
title = {Reinforcement learning for evaluating school safety designs in active shooter incidents},
author = {Ruying Liu and Wanjing Wu and Burcin Becerik-Gerber and Gale M. Lucas and Michelle Laboy and David Fannon},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1474034625004689},
doi = {10.1016/j.aei.2025.103575},
issn = {14740346},
year = {2025},
date = {2025-09-01},
urldate = {2025-08-19},
journal = {Advanced Engineering Informatics},
volume = {67},
pages = {103575},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
5.
Wang, Yunzhe; Ustun, Volkan; McGroarty, Chris
A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research Miscellaneous
2025, (arXiv:2509.06355 [cs]).
@misc{wang_data-driven_2025,
title = {A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research},
author = {Yunzhe Wang and Volkan Ustun and Chris McGroarty},
url = {http://arxiv.org/abs/2509.06355},
doi = {10.48550/arXiv.2509.06355},
year = {2025},
date = {2025-09-01},
urldate = {2025-09-18},
publisher = {arXiv},
abstract = {Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.},
note = {arXiv:2509.06355 [cs]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.
6.
Wang, Yunzhe; Ustun, Volkan; McGroarty, Chris
A Data-Driven Discretized CS:GO Simulation Environment to Facilitate Strategic Multi-Agent Planning Research Miscellaneous
2025, (arXiv:2509.06355 [cs.AI]).
@misc{wang_data-driven_2025-1,
title = {A Data-Driven Discretized CS:GO Simulation Environment to Facilitate Strategic Multi-Agent Planning Research},
author = {Yunzhe Wang and Volkan Ustun and Chris McGroarty},
url = {http://arxiv.org/abs/2509.06355},
doi = {10.48550/arXiv.2509.06355},
year = {2025},
date = {2025-09-01},
urldate = {2026-06-22},
publisher = {arXiv},
abstract = {Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.},
note = {arXiv:2509.06355 [cs.AI]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.
7.
Ustun, Volkan; Hans, Soham; Kumar, Rajay; Wang, Yunzhe
Abstracting Geo-specific Terrains to Scale Up Reinforcement Learning Miscellaneous
2025, (arXiv:2503.20078 [cs]).
@misc{ustun_abstracting_2025,
title = {Abstracting Geo-specific Terrains to Scale Up Reinforcement Learning},
author = {Volkan Ustun and Soham Hans and Rajay Kumar and Yunzhe Wang},
url = {http://arxiv.org/abs/2503.20078},
doi = {10.48550/arXiv.2503.20078},
year = {2025},
date = {2025-03-01},
urldate = {2025-04-15},
publisher = {arXiv},
abstract = {Multi-agent reinforcement learning (MARL) is increasingly ubiquitous in training dynamic and adaptive synthetic characters for interactive simulations on geo-specific terrains. Frameworks such as Unity's ML-Agents help to make such reinforcement learning experiments more accessible to the simulation community. Military training simulations also benefit from advances in MARL, but they have immense computational requirements due to their complex, continuous, stochastic, partially observable, non-stationary, and doctrine-based nature. Furthermore, these simulations require geo-specific terrains, further exacerbating the computational resources problem. In our research, we leverage Unity's waypoints to automatically generate multi-layered representation abstractions of the geo-specific terrains to scale up reinforcement learning while still allowing the transfer of learned policies between different representations. Our early exploratory results on a novel MARL scenario, where each side has differing objectives, indicate that waypoint-based navigation enables faster and more efficient learning while producing trajectories similar to those taken by expert human players in CSGO gaming environments. This research points out the potential of waypoint-based navigation for reducing the computational costs of developing and training MARL models for military training simulations, where geo-specific terrains and differing objectives are crucial.},
note = {arXiv:2503.20078 [cs]},
keywords = {},
pubstate = {published},
tppubtype = {misc}
}
Multi-agent reinforcement learning (MARL) is increasingly ubiquitous in training dynamic and adaptive synthetic characters for interactive simulations on geo-specific terrains. Frameworks such as Unity's ML-Agents help to make such reinforcement learning experiments more accessible to the simulation community. Military training simulations also benefit from advances in MARL, but they have immense computational requirements due to their complex, continuous, stochastic, partially observable, non-stationary, and doctrine-based nature. Furthermore, these simulations require geo-specific terrains, further exacerbating the computational resources problem. In our research, we leverage Unity's waypoints to automatically generate multi-layered representation abstractions of the geo-specific terrains to scale up reinforcement learning while still allowing the transfer of learned policies between different representations. Our early exploratory results on a novel MARL scenario, where each side has differing objectives, indicate that waypoint-based navigation enables faster and more efficient learning while producing trajectories similar to those taken by expert human players in CSGO gaming environments. This research points out the potential of waypoint-based navigation for reducing the computational costs of developing and training MARL models for military training simulations, where geo-specific terrains and differing objectives are crucial.
8.
Jones, Brennan; Xu, Yan; Li, Qisheng; Scherer, Stefan
Designing a Proactive Context-Aware AI Chatbot for People's Long-Term Goals Proceedings Article
In: Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, pp. 1–7, ACM, Honolulu HI USA, 2024, ISBN: 979-8-4007-0331-7.
@inproceedings{jones_designing_2024,
title = {Designing a Proactive Context-Aware AI Chatbot for People's Long-Term Goals},
author = {Brennan Jones and Yan Xu and Qisheng Li and Stefan Scherer},
url = {https://dl.acm.org/doi/10.1145/3613905.3650912},
doi = {10.1145/3613905.3650912},
isbn = {979-8-4007-0331-7},
year = {2024},
date = {2024-05-01},
urldate = {2024-06-25},
booktitle = {Extended Abstracts of the CHI Conference on Human Factors in Computing Systems},
pages = {1–7},
publisher = {ACM},
address = {Honolulu HI USA},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
9.
Aris, Timothy; Ustun, Volkan; Kumar, Rajay
Training Reinforcement Learning Agents to React to an Ambush for Military Simulations Journal Article
In: FLAIRS, vol. 37, 2024, ISSN: 2334-0762.
@article{aris_training_2024,
title = {Training Reinforcement Learning Agents to React to an Ambush for Military Simulations},
author = {Timothy Aris and Volkan Ustun and Rajay Kumar},
url = {https://journals.flvc.org/FLAIRS/article/view/135578},
doi = {10.32473/flairs.37.1.135578},
issn = {2334-0762},
year = {2024},
date = {2024-05-01},
urldate = {2024-08-13},
journal = {FLAIRS},
volume = {37},
abstract = {There is a need for realistic Opposing Forces (OPFOR)behavior in military training simulations. Current trainingsimulations generally only have simple, non-adaptivebehaviors, requiring human instructors to play the role ofOPFOR in any complicated scenario. This poster addressesthis need by focusing on a specific scenario: trainingreinforcement learning agents to react to an ambush. Itproposes a novel way to check for occlusion algorithmically.It shows vector fields showing the agent’s actions throughthe course of a training run. It shows that a single agentswitching between multiple goals is possible, at least in asimplified environment. Such an approach could reduce theneed to develop different agents for different scenarios.Finally, it shows a competent agent trained on a simplifiedReact to Ambush scenario, demonstrating the plausibility ofa scaled-up version.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
There is a need for realistic Opposing Forces (OPFOR)behavior in military training simulations. Current trainingsimulations generally only have simple, non-adaptivebehaviors, requiring human instructors to play the role ofOPFOR in any complicated scenario. This poster addressesthis need by focusing on a specific scenario: trainingreinforcement learning agents to react to an ambush. Itproposes a novel way to check for occlusion algorithmically.It shows vector fields showing the agent’s actions throughthe course of a training run. It shows that a single agentswitching between multiple goals is possible, at least in asimplified environment. Such an approach could reduce theneed to develop different agents for different scenarios.Finally, it shows a competent agent trained on a simplifiedReact to Ambush scenario, demonstrating the plausibility ofa scaled-up version.
10.
Rizzo, Albert Skip; Hartholt, Arno; Mozgai, Sharon
Settling the Score: Virtual Reality as a Tool to Enhance Trauma-Focused Therapy for PTSD Book Section
In: Rich, Grant J.; Kumar, V. K.; Farley, Frank H. (Ed.): Handbook of Media Psychology, pp. 187–213, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-56536-6 978-3-031-56537-3.
@incollection{rich_settling_2024,
title = {Settling the Score: Virtual Reality as a Tool to Enhance Trauma-Focused Therapy for PTSD},
author = {Albert Skip Rizzo and Arno Hartholt and Sharon Mozgai},
editor = {Grant J. Rich and V. K. Kumar and Frank H. Farley},
url = {https://link.springer.com/10.1007/978-3-031-56537-3_14},
doi = {10.1007/978-3-031-56537-3_14},
isbn = {978-3-031-56536-6 978-3-031-56537-3},
year = {2024},
date = {2024-04-01},
urldate = {2024-06-18},
booktitle = {Handbook of Media Psychology},
pages = {187–213},
publisher = {Springer Nature Switzerland},
address = {Cham},
keywords = {},
pubstate = {published},
tppubtype = {incollection}
}
11.
Liu, Ruying; Becerik-Gerber, Burcin; Lucas, Gale M.
Effectiveness of VR-based training on improving occupants’ response and preparedness for active shooter incidents Journal Article
In: Safety Science, vol. 164, pp. 106175, 2023, ISSN: 09257535.
@article{liu_effectiveness_2023,
title = {Effectiveness of VR-based training on improving occupants’ response and preparedness for active shooter incidents},
author = {Ruying Liu and Burcin Becerik-Gerber and Gale M. Lucas},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0925753523001170},
doi = {10.1016/j.ssci.2023.106175},
issn = {09257535},
year = {2023},
date = {2023-08-01},
urldate = {2023-08-22},
journal = {Safety Science},
volume = {164},
pages = {106175},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
12.
Liu, Ruying; Zhu, Runhe; Becerik‐Gerber, Burcin; Lucas, Gale M.; Southers, Erroll G.
Be prepared: How training and emergency type affect evacuation behaviour Journal Article
In: Computer Assisted Learning, pp. jcal.12812, 2023, ISSN: 0266-4909, 1365-2729.
@article{liu_be_2023,
title = {Be prepared: How training and emergency type affect evacuation behaviour},
author = {Ruying Liu and Runhe Zhu and Burcin Becerik‐Gerber and Gale M. Lucas and Erroll G. Southers},
url = {https://onlinelibrary.wiley.com/doi/10.1111/jcal.12812},
doi = {10.1111/jcal.12812},
issn = {0266-4909, 1365-2729},
year = {2023},
date = {2023-04-01},
urldate = {2023-08-22},
journal = {Computer Assisted Learning},
pages = {jcal.12812},
abstract = {Abstract
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Abstract
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.
13.
Hartholt, Arno; McCullough, Kyle; Fast, Ed; Leeds, Andrew; Mozgai, Sharon; Aris, Tim; Ustun, Volkan; Gordon, Andrew; McGroarty, Christopher
Rapid Prototyping for Simulation and Training with the Rapid Integration & Development Environment (RIDE) Proceedings Article
In: 2021.
@inproceedings{hartholt_rapid_2021,
title = {Rapid Prototyping for Simulation and Training with the Rapid Integration & Development Environment (RIDE)},
author = {Arno Hartholt and Kyle McCullough and Ed Fast and Andrew Leeds and Sharon Mozgai and Tim Aris and Volkan Ustun and Andrew Gordon and Christopher McGroarty},
year = {2021},
date = {2021-11-01},
keywords = {},
pubstate = {published},
tppubtype = {inproceedings}
}
14.
Chen, Meida; Feng, Andrew; Hou, Yu; McCullough, Kyle; Prasad, Pratusha Bhuvana; Soibelman, Lucio
Ground material classification and for UAV-based photogrammetric 3D data A 2D-3D Hybrid Approach Journal Article
In: 2021.
@article{chen_ground_2021,
title = {Ground material classification and for UAV-based photogrammetric 3D data A 2D-3D Hybrid Approach},
author = {Meida Chen and Andrew Feng and Yu Hou and Kyle McCullough and Pratusha Bhuvana Prasad and Lucio Soibelman},
url = {https://arxiv.org/abs/2109.12221},
doi = {10.48550/ARXIV.2109.12221},
year = {2021},
date = {2021-01-01},
urldate = {2022-09-27},
abstract = {In recent years, photogrammetry has been widely used in many areas to create photorealistic 3D virtual data representing the physical environment. The innovation of small unmanned aerial vehicles (sUAVs) has provided additional high-resolution imaging capabilities with low cost for mapping a relatively large area of interest. These cutting-edge technologies have caught the US Army and Navy's attention for the purpose of rapid 3D battlefield reconstruction, virtual training, and simulations. Our previous works have demonstrated the importance of information extraction from the derived photogrammetric data to create semantic-rich virtual environments (Chen et al., 2019). For example, an increase of simulation realism and fidelity was achieved by segmenting and replacing photogrammetric trees with game-ready tree models. In this work, we further investigated the semantic information extraction problem and focused on the ground material segmentation and object detection tasks. The main innovation of this work was that we leveraged both the original 2D images and the derived 3D photogrammetric data to overcome the challenges faced when using each individual data source. For ground material segmentation, we utilized an existing convolutional neural network architecture (i.e., 3DMV) which was originally designed for segmenting RGB-D sensed indoor data. We improved its performance for outdoor photogrammetric data by introducing a depth pooling layer in the architecture to take into consideration the distance between the source images and the reconstructed terrain model. To test the performance of our improved 3DMV, a ground truth ground material database was created using data from the One World Terrain (OWT) data repository. Finally, a workflow for importing the segmented ground materials into a virtual simulation scene was introduced, and visual results are reported in this paper.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
In recent years, photogrammetry has been widely used in many areas to create photorealistic 3D virtual data representing the physical environment. The innovation of small unmanned aerial vehicles (sUAVs) has provided additional high-resolution imaging capabilities with low cost for mapping a relatively large area of interest. These cutting-edge technologies have caught the US Army and Navy's attention for the purpose of rapid 3D battlefield reconstruction, virtual training, and simulations. Our previous works have demonstrated the importance of information extraction from the derived photogrammetric data to create semantic-rich virtual environments (Chen et al., 2019). For example, an increase of simulation realism and fidelity was achieved by segmenting and replacing photogrammetric trees with game-ready tree models. In this work, we further investigated the semantic information extraction problem and focused on the ground material segmentation and object detection tasks. The main innovation of this work was that we leveraged both the original 2D images and the derived 3D photogrammetric data to overcome the challenges faced when using each individual data source. For ground material segmentation, we utilized an existing convolutional neural network architecture (i.e., 3DMV) which was originally designed for segmenting RGB-D sensed indoor data. We improved its performance for outdoor photogrammetric data by introducing a depth pooling layer in the architecture to take into consideration the distance between the source images and the reconstructed terrain model. To test the performance of our improved 3DMV, a ground truth ground material database was created using data from the One World Terrain (OWT) data repository. Finally, a workflow for importing the segmented ground materials into a virtual simulation scene was introduced, and visual results are reported in this paper.
15.
Oxendine, Christopher; O'Banion, Matt; Wright, William; Irmischer, Ian; Fleming, Steven
Rapid Terrain Generation for GeoVisualization, Simulation, Mission Rehearsal, & Operations Journal Article
In: 2019 State and Future of GEOINT Report, pp. 5, 2019.
@article{oxendine_rapid_2019,
title = {Rapid Terrain Generation for GeoVisualization, Simulation, Mission Rehearsal, & Operations},
author = {Christopher Oxendine and Matt O'Banion and William Wright and Ian Irmischer and Steven Fleming},
url = {https://digitalcommons.usmalibrary.org/usma_research_papers/151/},
year = {2019},
date = {2019-06-01},
journal = {2019 State and Future of GEOINT Report},
pages = {5},
abstract = {Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.
16.
Fleming, Steven D; O’Banion, Matt S; McAlinden, Ryan; Oxendine, Christopher; Wright, William; Irmischer, Ian
Rapid Terrain Generation for Geovisualization, Simulation, Mission Rehearsal & Operations Journal Article
In: Annual Report (State and Future of GEOINT), pp. 5, 2019.
@article{fleming_rapid_2019,
title = {Rapid Terrain Generation for Geovisualization, Simulation, Mission Rehearsal & Operations},
author = {Steven D Fleming and Matt S O’Banion and Ryan McAlinden and Christopher Oxendine and William Wright and Ian Irmischer},
url = {http://trajectorymagazine.com/rapid-terrain-generation/},
year = {2019},
date = {2019-01-01},
journal = {Annual Report (State and Future of GEOINT)},
pages = {5},
abstract = {Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.},
keywords = {},
pubstate = {published},
tppubtype = {article}
}
Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.
Filter
2026
Han, Zicheng; Becerik-Gerber, Burcin; Liu, Ruying; Lucas, Gale; Doolab, Yogin; Zein, Sara; Atkins, Margo; Blass, Deborah; Miles, Lachlan
Emergency dynamics modeling with an agent-based framework: Active shooter scenario case study Journal Article
In: International Journal of Disaster Risk Reduction, vol. 142, pp. 106226, 2026, ISSN: 22124209.
Abstract | Links | BibTeX | Tags: Simulation, Virtual Humans
@article{han_emergency_2026,
title = {Emergency dynamics modeling with an agent-based framework: Active shooter scenario case study},
author = {Zicheng Han and Burcin Becerik-Gerber and Ruying Liu and Gale Lucas and Yogin Doolab and Sara Zein and Margo Atkins and Deborah Blass and Lachlan Miles},
url = {https://linkinghub.elsevier.com/retrieve/pii/S2212420926002384},
doi = {10.1016/j.ijdrr.2026.106226},
issn = {22124209},
year = {2026},
date = {2026-05-01},
urldate = {2026-06-16},
journal = {International Journal of Disaster Risk Reduction},
volume = {142},
pages = {106226},
abstract = {Modeling human-induced emergencies requires a structured representation of interactions among occupants, hazards, and built environments. While existing simulation approaches provide important insights, behavioral processes, hazard progression, and architectural conditions are often examined separately. This study develops a parameterized agent-based modeling framework that integrates heterogeneous occupant behaviors, hazard dynamics, and environmental configurations within a commercially available crowd-simulation platform to extend its usability within the academic and professional communities. The framework comprises environmental modeling, agent definition, interaction logic, risk evaluation, and simulation execution, enabling systematic configuration of behavioral rules and scenario parameters through customized scripting. An Active Shooter Incident (ASI) in a school building is simulated as a demonstrative case study to illustrate the framework's capability to configure and test emergency scenarios under different behavioral, hazard and environmental assumptions. Using the framework, the study examines the effects of classroom occupancy, door width, and door-locking mechanisms on occupant safety, evaluated using a token-based high-risk metric and factorial statistical analysis. Results indicate that higher occupancy increases exposure risk, timely door locking substantially improves safety outcomes, and the influence of door width is context-dependent. The consistency of these patterns with prior research supports the behavioral plausibility of the modeling approach. Overall, the study contributes a practical and extensible framework for operationalizing human–hazard–environment interactions in similar agent-based simulations and for generating scenario-specific insights to support design and preparedness strategies in human-induced emergency contexts.},
keywords = {Simulation, Virtual Humans},
pubstate = {published},
tppubtype = {article}
}
Modeling human-induced emergencies requires a structured representation of interactions among occupants, hazards, and built environments. While existing simulation approaches provide important insights, behavioral processes, hazard progression, and architectural conditions are often examined separately. This study develops a parameterized agent-based modeling framework that integrates heterogeneous occupant behaviors, hazard dynamics, and environmental configurations within a commercially available crowd-simulation platform to extend its usability within the academic and professional communities. The framework comprises environmental modeling, agent definition, interaction logic, risk evaluation, and simulation execution, enabling systematic configuration of behavioral rules and scenario parameters through customized scripting. An Active Shooter Incident (ASI) in a school building is simulated as a demonstrative case study to illustrate the framework's capability to configure and test emergency scenarios under different behavioral, hazard and environmental assumptions. Using the framework, the study examines the effects of classroom occupancy, door width, and door-locking mechanisms on occupant safety, evaluated using a token-based high-risk metric and factorial statistical analysis. Results indicate that higher occupancy increases exposure risk, timely door locking substantially improves safety outcomes, and the influence of door width is context-dependent. The consistency of these patterns with prior research supports the behavioral plausibility of the modeling approach. Overall, the study contributes a practical and extensible framework for operationalizing human–hazard–environment interactions in similar agent-based simulations and for generating scenario-specific insights to support design and preparedness strategies in human-induced emergency contexts.
Kwon, Deuksin; Shrestha, Kaleen; Han, Bin; Lin, Spencer; Hale, James; Gratch, Jonathan; Matarić, Maja; Lucas, Gale M.
Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution Miscellaneous
2026, (arXiv:2602.07414 [cs.AI]).
Abstract | Links | BibTeX | Tags: ARO, DTIC, LLM, Simulation, UARC
@misc{kwon_can_2026,
title = {Can LLMs Truly Embody Human Personality? Analyzing AI and Human Behavior Alignment in Dispute Resolution},
author = {Deuksin Kwon and Kaleen Shrestha and Bin Han and Spencer Lin and James Hale and Jonathan Gratch and Maja Matarić and Gale M. Lucas},
url = {http://arxiv.org/abs/2602.07414},
doi = {10.48550/arXiv.2602.07414},
year = {2026},
date = {2026-02-01},
urldate = {2026-06-18},
publisher = {arXiv},
abstract = {Large language models (LLMs) are increasingly used to simulate human behavior in social settings such as legal mediation, negotiation, and dispute resolution. However, it remains unclear whether these simulations reproduce the personality-behavior patterns observed in humans. Human personality, for instance, shapes how individuals navigate social interactions, including strategic choices and behaviors in emotionally charged interactions. This raises the question: Can LLMs, when prompted with personality traits, reproduce personality-driven differences in human conflict behavior? To explore this, we introduce an evaluation framework that enables direct comparison of human-human and LLM-LLM behaviors in dispute resolution dialogues with respect to Big Five Inventory (BFI) personality traits. This framework provides a set of interpretable metrics related to strategic behavior and conflict outcomes. We additionally contribute a novel dataset creation methodology for LLM dispute resolution dialogues with matched scenarios and personality traits with respect to human conversations. Finally, we demonstrate the use of our evaluation framework with three contemporary closed-source LLMs and show significant divergences in how personality manifests in conflict across different LLMs compared to human data, challenging the assumption that personality-prompted agents can serve as reliable behavioral proxies in socially impactful applications. Our work highlights the need for psychological grounding and validation in AI simulations before real-world use.},
note = {arXiv:2602.07414 [cs.AI]},
keywords = {ARO, DTIC, LLM, Simulation, UARC},
pubstate = {published},
tppubtype = {misc}
}
Large language models (LLMs) are increasingly used to simulate human behavior in social settings such as legal mediation, negotiation, and dispute resolution. However, it remains unclear whether these simulations reproduce the personality-behavior patterns observed in humans. Human personality, for instance, shapes how individuals navigate social interactions, including strategic choices and behaviors in emotionally charged interactions. This raises the question: Can LLMs, when prompted with personality traits, reproduce personality-driven differences in human conflict behavior? To explore this, we introduce an evaluation framework that enables direct comparison of human-human and LLM-LLM behaviors in dispute resolution dialogues with respect to Big Five Inventory (BFI) personality traits. This framework provides a set of interpretable metrics related to strategic behavior and conflict outcomes. We additionally contribute a novel dataset creation methodology for LLM dispute resolution dialogues with matched scenarios and personality traits with respect to human conversations. Finally, we demonstrate the use of our evaluation framework with three contemporary closed-source LLMs and show significant divergences in how personality manifests in conflict across different LLMs compared to human data, challenging the assumption that personality-prompted agents can serve as reliable behavioral proxies in socially impactful applications. Our work highlights the need for psychological grounding and validation in AI simulations before real-world use.
2025
Zhang, Tianyi; Zhou, Xiaolin; Wang, Yunzhe; Cambria, Erik; Traum, David; Mao, Rui
Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods Miscellaneous
2025, (arXiv:2510.20252 [cs.AI]).
Abstract | Links | BibTeX | Tags: AI, LLM, Natural Language, Simulation
@misc{zhang_individualized_2025,
title = {Individualized Cognitive Simulation in Large Language Models: Evaluating Different Cognitive Representation Methods},
author = {Tianyi Zhang and Xiaolin Zhou and Yunzhe Wang and Erik Cambria and David Traum and Rui Mao},
url = {http://arxiv.org/abs/2510.20252},
doi = {10.48550/arXiv.2510.20252},
year = {2025},
date = {2025-10-01},
urldate = {2026-07-06},
publisher = {arXiv},
abstract = {Individualized cognitive simulation (ICS) aims to build computational models that approximate the thought processes of specific individuals. While large language models (LLMs) convincingly mimic surface-level human behavior such as role-play, their ability to simulate deeper individualized cognitive processes remains poorly understood. To address this gap, we introduce a novel task that evaluates different cognitive representation methods in ICS. We construct a dataset from recently published novels (later than the release date of the tested LLMs) and propose an 11-condition cognitive evaluation framework to benchmark seven off-the-shelf LLMs in the context of authorial style emulation. We hypothesize that effective cognitive representations can help LLMs generate storytelling that better mirrors the original author. Thus, we test different cognitive representations, e.g., linguistic features, concept mappings, and profile-based information. Results show that combining conceptual and linguistic features is particularly effective in ICS, outperforming static profile-based cues in overall evaluation. Importantly, LLMs are more effective at mimicking linguistic style than narrative structure, underscoring their limits in deeper cognitive simulation. These findings provide a foundation for developing AI systems that adapt to individual ways of thinking and expression, advancing more personalized and human-aligned creative technologies.},
note = {arXiv:2510.20252 [cs.AI]},
keywords = {AI, LLM, Natural Language, Simulation},
pubstate = {published},
tppubtype = {misc}
}
Individualized cognitive simulation (ICS) aims to build computational models that approximate the thought processes of specific individuals. While large language models (LLMs) convincingly mimic surface-level human behavior such as role-play, their ability to simulate deeper individualized cognitive processes remains poorly understood. To address this gap, we introduce a novel task that evaluates different cognitive representation methods in ICS. We construct a dataset from recently published novels (later than the release date of the tested LLMs) and propose an 11-condition cognitive evaluation framework to benchmark seven off-the-shelf LLMs in the context of authorial style emulation. We hypothesize that effective cognitive representations can help LLMs generate storytelling that better mirrors the original author. Thus, we test different cognitive representations, e.g., linguistic features, concept mappings, and profile-based information. Results show that combining conceptual and linguistic features is particularly effective in ICS, outperforming static profile-based cues in overall evaluation. Importantly, LLMs are more effective at mimicking linguistic style than narrative structure, underscoring their limits in deeper cognitive simulation. These findings provide a foundation for developing AI systems that adapt to individual ways of thinking and expression, advancing more personalized and human-aligned creative technologies.
Liu, Ruying; Wu, Wanjing; Becerik-Gerber, Burcin; Lucas, Gale M.; Laboy, Michelle; Fannon, David
Reinforcement learning for evaluating school safety designs in active shooter incidents Journal Article
In: Advanced Engineering Informatics, vol. 67, pp. 103575, 2025, ISSN: 14740346.
Links | BibTeX | Tags: DTIC, Simulation
@article{liu_reinforcement_2025,
title = {Reinforcement learning for evaluating school safety designs in active shooter incidents},
author = {Ruying Liu and Wanjing Wu and Burcin Becerik-Gerber and Gale M. Lucas and Michelle Laboy and David Fannon},
url = {https://linkinghub.elsevier.com/retrieve/pii/S1474034625004689},
doi = {10.1016/j.aei.2025.103575},
issn = {14740346},
year = {2025},
date = {2025-09-01},
urldate = {2025-08-19},
journal = {Advanced Engineering Informatics},
volume = {67},
pages = {103575},
keywords = {DTIC, Simulation},
pubstate = {published},
tppubtype = {article}
}
Wang, Yunzhe; Ustun, Volkan; McGroarty, Chris
A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research Miscellaneous
2025, (arXiv:2509.06355 [cs]).
Abstract | Links | BibTeX | Tags: DTIC, Simulation
@misc{wang_data-driven_2025,
title = {A data-driven discretized CS:GO simulation environment to facilitate strategic multi-agent planning research},
author = {Yunzhe Wang and Volkan Ustun and Chris McGroarty},
url = {http://arxiv.org/abs/2509.06355},
doi = {10.48550/arXiv.2509.06355},
year = {2025},
date = {2025-09-01},
urldate = {2025-09-18},
publisher = {arXiv},
abstract = {Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.},
note = {arXiv:2509.06355 [cs]},
keywords = {DTIC, Simulation},
pubstate = {published},
tppubtype = {misc}
}
Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.
Wang, Yunzhe; Ustun, Volkan; McGroarty, Chris
A Data-Driven Discretized CS:GO Simulation Environment to Facilitate Strategic Multi-Agent Planning Research Miscellaneous
2025, (arXiv:2509.06355 [cs.AI]).
Abstract | Links | BibTeX | Tags: DTIC, Simulation, UARC
@misc{wang_data-driven_2025-1,
title = {A Data-Driven Discretized CS:GO Simulation Environment to Facilitate Strategic Multi-Agent Planning Research},
author = {Yunzhe Wang and Volkan Ustun and Chris McGroarty},
url = {http://arxiv.org/abs/2509.06355},
doi = {10.48550/arXiv.2509.06355},
year = {2025},
date = {2025-09-01},
urldate = {2026-06-22},
publisher = {arXiv},
abstract = {Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.},
note = {arXiv:2509.06355 [cs.AI]},
keywords = {DTIC, Simulation, UARC},
pubstate = {published},
tppubtype = {misc}
}
Modern simulation environments for complex multi-agent interactions must balance high-fidelity detail with computational efficiency. We present DECOY, a novel multi-agent simulator that abstracts strategic, long-horizon planning in 3D terrains into high-level discretized simulation while preserving low-level environmental fidelity. Using Counter-Strike: Global Offensive (CS:GO) as a testbed, our framework accurately simulates gameplay using only movement decisions as tactical positioning – without explicitly modeling low-level mechanics such as aiming and shooting. Central to our approach is a waypoint system that simplifies and discretizes continuous states and actions, paired with neural predictive and generative models trained on real CS:GO tournament data to reconstruct event outcomes. Extensive evaluations show that replays generated from human data in DECOY closely match those observed in the original game. Our publicly available simulation environment provides a valuable tool for advancing research in strategic multi-agent planning and behavior generation.
Ustun, Volkan; Hans, Soham; Kumar, Rajay; Wang, Yunzhe
Abstracting Geo-specific Terrains to Scale Up Reinforcement Learning Miscellaneous
2025, (arXiv:2503.20078 [cs]).
Abstract | Links | BibTeX | Tags: DTIC, Simulation
@misc{ustun_abstracting_2025,
title = {Abstracting Geo-specific Terrains to Scale Up Reinforcement Learning},
author = {Volkan Ustun and Soham Hans and Rajay Kumar and Yunzhe Wang},
url = {http://arxiv.org/abs/2503.20078},
doi = {10.48550/arXiv.2503.20078},
year = {2025},
date = {2025-03-01},
urldate = {2025-04-15},
publisher = {arXiv},
abstract = {Multi-agent reinforcement learning (MARL) is increasingly ubiquitous in training dynamic and adaptive synthetic characters for interactive simulations on geo-specific terrains. Frameworks such as Unity's ML-Agents help to make such reinforcement learning experiments more accessible to the simulation community. Military training simulations also benefit from advances in MARL, but they have immense computational requirements due to their complex, continuous, stochastic, partially observable, non-stationary, and doctrine-based nature. Furthermore, these simulations require geo-specific terrains, further exacerbating the computational resources problem. In our research, we leverage Unity's waypoints to automatically generate multi-layered representation abstractions of the geo-specific terrains to scale up reinforcement learning while still allowing the transfer of learned policies between different representations. Our early exploratory results on a novel MARL scenario, where each side has differing objectives, indicate that waypoint-based navigation enables faster and more efficient learning while producing trajectories similar to those taken by expert human players in CSGO gaming environments. This research points out the potential of waypoint-based navigation for reducing the computational costs of developing and training MARL models for military training simulations, where geo-specific terrains and differing objectives are crucial.},
note = {arXiv:2503.20078 [cs]},
keywords = {DTIC, Simulation},
pubstate = {published},
tppubtype = {misc}
}
Multi-agent reinforcement learning (MARL) is increasingly ubiquitous in training dynamic and adaptive synthetic characters for interactive simulations on geo-specific terrains. Frameworks such as Unity's ML-Agents help to make such reinforcement learning experiments more accessible to the simulation community. Military training simulations also benefit from advances in MARL, but they have immense computational requirements due to their complex, continuous, stochastic, partially observable, non-stationary, and doctrine-based nature. Furthermore, these simulations require geo-specific terrains, further exacerbating the computational resources problem. In our research, we leverage Unity's waypoints to automatically generate multi-layered representation abstractions of the geo-specific terrains to scale up reinforcement learning while still allowing the transfer of learned policies between different representations. Our early exploratory results on a novel MARL scenario, where each side has differing objectives, indicate that waypoint-based navigation enables faster and more efficient learning while producing trajectories similar to those taken by expert human players in CSGO gaming environments. This research points out the potential of waypoint-based navigation for reducing the computational costs of developing and training MARL models for military training simulations, where geo-specific terrains and differing objectives are crucial.
2024
Jones, Brennan; Xu, Yan; Li, Qisheng; Scherer, Stefan
Designing a Proactive Context-Aware AI Chatbot for People's Long-Term Goals Proceedings Article
In: Extended Abstracts of the CHI Conference on Human Factors in Computing Systems, pp. 1–7, ACM, Honolulu HI USA, 2024, ISBN: 979-8-4007-0331-7.
Links | BibTeX | Tags: AI, Simulation
@inproceedings{jones_designing_2024,
title = {Designing a Proactive Context-Aware AI Chatbot for People's Long-Term Goals},
author = {Brennan Jones and Yan Xu and Qisheng Li and Stefan Scherer},
url = {https://dl.acm.org/doi/10.1145/3613905.3650912},
doi = {10.1145/3613905.3650912},
isbn = {979-8-4007-0331-7},
year = {2024},
date = {2024-05-01},
urldate = {2024-06-25},
booktitle = {Extended Abstracts of the CHI Conference on Human Factors in Computing Systems},
pages = {1–7},
publisher = {ACM},
address = {Honolulu HI USA},
keywords = {AI, Simulation},
pubstate = {published},
tppubtype = {inproceedings}
}
Aris, Timothy; Ustun, Volkan; Kumar, Rajay
Training Reinforcement Learning Agents to React to an Ambush for Military Simulations Journal Article
In: FLAIRS, vol. 37, 2024, ISSN: 2334-0762.
Abstract | Links | BibTeX | Tags: Simulation, VR
@article{aris_training_2024,
title = {Training Reinforcement Learning Agents to React to an Ambush for Military Simulations},
author = {Timothy Aris and Volkan Ustun and Rajay Kumar},
url = {https://journals.flvc.org/FLAIRS/article/view/135578},
doi = {10.32473/flairs.37.1.135578},
issn = {2334-0762},
year = {2024},
date = {2024-05-01},
urldate = {2024-08-13},
journal = {FLAIRS},
volume = {37},
abstract = {There is a need for realistic Opposing Forces (OPFOR)behavior in military training simulations. Current trainingsimulations generally only have simple, non-adaptivebehaviors, requiring human instructors to play the role ofOPFOR in any complicated scenario. This poster addressesthis need by focusing on a specific scenario: trainingreinforcement learning agents to react to an ambush. Itproposes a novel way to check for occlusion algorithmically.It shows vector fields showing the agent’s actions throughthe course of a training run. It shows that a single agentswitching between multiple goals is possible, at least in asimplified environment. Such an approach could reduce theneed to develop different agents for different scenarios.Finally, it shows a competent agent trained on a simplifiedReact to Ambush scenario, demonstrating the plausibility ofa scaled-up version.},
keywords = {Simulation, VR},
pubstate = {published},
tppubtype = {article}
}
There is a need for realistic Opposing Forces (OPFOR)behavior in military training simulations. Current trainingsimulations generally only have simple, non-adaptivebehaviors, requiring human instructors to play the role ofOPFOR in any complicated scenario. This poster addressesthis need by focusing on a specific scenario: trainingreinforcement learning agents to react to an ambush. Itproposes a novel way to check for occlusion algorithmically.It shows vector fields showing the agent’s actions throughthe course of a training run. It shows that a single agentswitching between multiple goals is possible, at least in asimplified environment. Such an approach could reduce theneed to develop different agents for different scenarios.Finally, it shows a competent agent trained on a simplifiedReact to Ambush scenario, demonstrating the plausibility ofa scaled-up version.
Rizzo, Albert Skip; Hartholt, Arno; Mozgai, Sharon
Settling the Score: Virtual Reality as a Tool to Enhance Trauma-Focused Therapy for PTSD Book Section
In: Rich, Grant J.; Kumar, V. K.; Farley, Frank H. (Ed.): Handbook of Media Psychology, pp. 187–213, Springer Nature Switzerland, Cham, 2024, ISBN: 978-3-031-56536-6 978-3-031-56537-3.
Links | BibTeX | Tags: DTIC, MedVR, Simulation, VR
@incollection{rich_settling_2024,
title = {Settling the Score: Virtual Reality as a Tool to Enhance Trauma-Focused Therapy for PTSD},
author = {Albert Skip Rizzo and Arno Hartholt and Sharon Mozgai},
editor = {Grant J. Rich and V. K. Kumar and Frank H. Farley},
url = {https://link.springer.com/10.1007/978-3-031-56537-3_14},
doi = {10.1007/978-3-031-56537-3_14},
isbn = {978-3-031-56536-6 978-3-031-56537-3},
year = {2024},
date = {2024-04-01},
urldate = {2024-06-18},
booktitle = {Handbook of Media Psychology},
pages = {187–213},
publisher = {Springer Nature Switzerland},
address = {Cham},
keywords = {DTIC, MedVR, Simulation, VR},
pubstate = {published},
tppubtype = {incollection}
}
2023
Liu, Ruying; Becerik-Gerber, Burcin; Lucas, Gale M.
Effectiveness of VR-based training on improving occupants’ response and preparedness for active shooter incidents Journal Article
In: Safety Science, vol. 164, pp. 106175, 2023, ISSN: 09257535.
Links | BibTeX | Tags: DTIC, Simulation, UARC
@article{liu_effectiveness_2023,
title = {Effectiveness of VR-based training on improving occupants’ response and preparedness for active shooter incidents},
author = {Ruying Liu and Burcin Becerik-Gerber and Gale M. Lucas},
url = {https://linkinghub.elsevier.com/retrieve/pii/S0925753523001170},
doi = {10.1016/j.ssci.2023.106175},
issn = {09257535},
year = {2023},
date = {2023-08-01},
urldate = {2023-08-22},
journal = {Safety Science},
volume = {164},
pages = {106175},
keywords = {DTIC, Simulation, UARC},
pubstate = {published},
tppubtype = {article}
}
Liu, Ruying; Zhu, Runhe; Becerik‐Gerber, Burcin; Lucas, Gale M.; Southers, Erroll G.
Be prepared: How training and emergency type affect evacuation behaviour Journal Article
In: Computer Assisted Learning, pp. jcal.12812, 2023, ISSN: 0266-4909, 1365-2729.
Abstract | Links | BibTeX | Tags: DTIC, Simulation, UARC
@article{liu_be_2023,
title = {Be prepared: How training and emergency type affect evacuation behaviour},
author = {Ruying Liu and Runhe Zhu and Burcin Becerik‐Gerber and Gale M. Lucas and Erroll G. Southers},
url = {https://onlinelibrary.wiley.com/doi/10.1111/jcal.12812},
doi = {10.1111/jcal.12812},
issn = {0266-4909, 1365-2729},
year = {2023},
date = {2023-04-01},
urldate = {2023-08-22},
journal = {Computer Assisted Learning},
pages = {jcal.12812},
abstract = {Abstract
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.},
keywords = {DTIC, Simulation, UARC},
pubstate = {published},
tppubtype = {article}
}
Abstract
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.
Background
Video‐based training has been widely adopted by private organizations and public authorities to educate occupants on various types of building emergencies. However, the effectiveness of video‐based training for preparing occupants for building emergencies has not been rigorously studied nor has the impact of emergency type been investigated on training effectiveness.
Objectives
This study examines whether video‐based training is an effective method to prepare occupants for building emergencies and how the effectiveness differs in the context of different building emergencies.
Methods
We simulated fire and active shooter emergencies in a virtual office building and conducted evacuation experiments to examine participants' emergency responses using both objective and subjective metrics. A total of 108 participants were recruited and responded to the fire or active shooter incident with or without video‐based training.
Results and Conclusions
The results revealed that participants with video‐based training more often chose to follow other recommendations when responding to building emergencies instead of simply following others. Results from ANOVA showed that training increased participants' self‐efficacy significantly, especially for those in the active shooter group. Moreover, participants in the active shooter simulation had a higher level of response efficacy than those in the fire emergency simulation. Our results also demonstrated the influence of emergency type on participants' final decisions and considerations of the recommendations.
Implications
Our results suggested that video‐based training is effective in improving participants' emergency preparedness and changing their behaviour patterns to a certain extent such as reducing following behaviour and encouraging safe evacuations. Additionally, statistically significant interactions between video‐based training and emergency types suggested that training effectiveness should be considered in accordance with the emergency type.
,
Lay Description
What is already known about this topic
People can behave differently in different types of building emergencies. Understanding human behaviours in building emergencies is essential for developing emergency preparedness strategies.
Emergency training is important for building occupants and video is a widely used media for emergency training. However, its training effectiveness needs to be evaluated.
What this paper adds
We used virtual environments to investigate evacuation behaviour.
The effectiveness of video‐based training and human responses in building emergencies were studied on both subjective responses and objective measurements.
Video‐based training significantly reduced the occurrence of following behaviours.
The different natures of the fire emergency and active shooter incidents shape the effectiveness of video‐based training.
Implications of study findings for practitioners
Video‐based training can improve building occupants' emergency preparedness to a certain extent.
Emergency training media should be designed considering the influence of emergency type.
2021
Hartholt, Arno; McCullough, Kyle; Fast, Ed; Leeds, Andrew; Mozgai, Sharon; Aris, Tim; Ustun, Volkan; Gordon, Andrew; McGroarty, Christopher
Rapid Prototyping for Simulation and Training with the Rapid Integration & Development Environment (RIDE) Proceedings Article
In: 2021.
BibTeX | Tags: AI, DTIC, Machine Learning, Simulation, UARC
@inproceedings{hartholt_rapid_2021,
title = {Rapid Prototyping for Simulation and Training with the Rapid Integration & Development Environment (RIDE)},
author = {Arno Hartholt and Kyle McCullough and Ed Fast and Andrew Leeds and Sharon Mozgai and Tim Aris and Volkan Ustun and Andrew Gordon and Christopher McGroarty},
year = {2021},
date = {2021-11-01},
keywords = {AI, DTIC, Machine Learning, Simulation, UARC},
pubstate = {published},
tppubtype = {inproceedings}
}
Chen, Meida; Feng, Andrew; Hou, Yu; McCullough, Kyle; Prasad, Pratusha Bhuvana; Soibelman, Lucio
Ground material classification and for UAV-based photogrammetric 3D data A 2D-3D Hybrid Approach Journal Article
In: 2021.
Abstract | Links | BibTeX | Tags: DTIC, Simulation, UARC
@article{chen_ground_2021,
title = {Ground material classification and for UAV-based photogrammetric 3D data A 2D-3D Hybrid Approach},
author = {Meida Chen and Andrew Feng and Yu Hou and Kyle McCullough and Pratusha Bhuvana Prasad and Lucio Soibelman},
url = {https://arxiv.org/abs/2109.12221},
doi = {10.48550/ARXIV.2109.12221},
year = {2021},
date = {2021-01-01},
urldate = {2022-09-27},
abstract = {In recent years, photogrammetry has been widely used in many areas to create photorealistic 3D virtual data representing the physical environment. The innovation of small unmanned aerial vehicles (sUAVs) has provided additional high-resolution imaging capabilities with low cost for mapping a relatively large area of interest. These cutting-edge technologies have caught the US Army and Navy's attention for the purpose of rapid 3D battlefield reconstruction, virtual training, and simulations. Our previous works have demonstrated the importance of information extraction from the derived photogrammetric data to create semantic-rich virtual environments (Chen et al., 2019). For example, an increase of simulation realism and fidelity was achieved by segmenting and replacing photogrammetric trees with game-ready tree models. In this work, we further investigated the semantic information extraction problem and focused on the ground material segmentation and object detection tasks. The main innovation of this work was that we leveraged both the original 2D images and the derived 3D photogrammetric data to overcome the challenges faced when using each individual data source. For ground material segmentation, we utilized an existing convolutional neural network architecture (i.e., 3DMV) which was originally designed for segmenting RGB-D sensed indoor data. We improved its performance for outdoor photogrammetric data by introducing a depth pooling layer in the architecture to take into consideration the distance between the source images and the reconstructed terrain model. To test the performance of our improved 3DMV, a ground truth ground material database was created using data from the One World Terrain (OWT) data repository. Finally, a workflow for importing the segmented ground materials into a virtual simulation scene was introduced, and visual results are reported in this paper.},
keywords = {DTIC, Simulation, UARC},
pubstate = {published},
tppubtype = {article}
}
In recent years, photogrammetry has been widely used in many areas to create photorealistic 3D virtual data representing the physical environment. The innovation of small unmanned aerial vehicles (sUAVs) has provided additional high-resolution imaging capabilities with low cost for mapping a relatively large area of interest. These cutting-edge technologies have caught the US Army and Navy's attention for the purpose of rapid 3D battlefield reconstruction, virtual training, and simulations. Our previous works have demonstrated the importance of information extraction from the derived photogrammetric data to create semantic-rich virtual environments (Chen et al., 2019). For example, an increase of simulation realism and fidelity was achieved by segmenting and replacing photogrammetric trees with game-ready tree models. In this work, we further investigated the semantic information extraction problem and focused on the ground material segmentation and object detection tasks. The main innovation of this work was that we leveraged both the original 2D images and the derived 3D photogrammetric data to overcome the challenges faced when using each individual data source. For ground material segmentation, we utilized an existing convolutional neural network architecture (i.e., 3DMV) which was originally designed for segmenting RGB-D sensed indoor data. We improved its performance for outdoor photogrammetric data by introducing a depth pooling layer in the architecture to take into consideration the distance between the source images and the reconstructed terrain model. To test the performance of our improved 3DMV, a ground truth ground material database was created using data from the One World Terrain (OWT) data repository. Finally, a workflow for importing the segmented ground materials into a virtual simulation scene was introduced, and visual results are reported in this paper.
2019
Oxendine, Christopher; O'Banion, Matt; Wright, William; Irmischer, Ian; Fleming, Steven
Rapid Terrain Generation for GeoVisualization, Simulation, Mission Rehearsal, & Operations Journal Article
In: 2019 State and Future of GEOINT Report, pp. 5, 2019.
Abstract | Links | BibTeX | Tags: Simulation
@article{oxendine_rapid_2019,
title = {Rapid Terrain Generation for GeoVisualization, Simulation, Mission Rehearsal, & Operations},
author = {Christopher Oxendine and Matt O'Banion and William Wright and Ian Irmischer and Steven Fleming},
url = {https://digitalcommons.usmalibrary.org/usma_research_papers/151/},
year = {2019},
date = {2019-06-01},
journal = {2019 State and Future of GEOINT Report},
pages = {5},
abstract = {Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.},
keywords = {Simulation},
pubstate = {published},
tppubtype = {article}
}
Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.
Fleming, Steven D; O’Banion, Matt S; McAlinden, Ryan; Oxendine, Christopher; Wright, William; Irmischer, Ian
Rapid Terrain Generation for Geovisualization, Simulation, Mission Rehearsal & Operations Journal Article
In: Annual Report (State and Future of GEOINT), pp. 5, 2019.
Abstract | Links | BibTeX | Tags: DoD, DTIC, Simulation, STG
@article{fleming_rapid_2019,
title = {Rapid Terrain Generation for Geovisualization, Simulation, Mission Rehearsal & Operations},
author = {Steven D Fleming and Matt S O’Banion and Ryan McAlinden and Christopher Oxendine and William Wright and Ian Irmischer},
url = {http://trajectorymagazine.com/rapid-terrain-generation/},
year = {2019},
date = {2019-01-01},
journal = {Annual Report (State and Future of GEOINT)},
pages = {5},
abstract = {Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.},
keywords = {DoD, DTIC, Simulation, STG},
pubstate = {published},
tppubtype = {article}
}
Geospecific 3D terrain representation (aka reality modeling) is revolutionizing geovisualization, simulation, and engineering practices around the world. In tandem with the rapid growth in unmanned aerial systems (UAS) and small satellites, reality modeling advancements now allow geospatial intelligence (GEOINT) practitioners to generate three-dimensional models from a decentralized collection of digital images to meet mission needs in both urban and rural environments. Scalable mesh models deliver enhanced, real-world visualization for engineers, geospatial teams, combatant, and combat support organizations. In this, reality modeling provides a detailed understanding of the physical environment, and models allow installation engineers and GEOINT practitioners to quickly generate updated, high-precision 3D reality meshes to provide real-world digital context for the decision-making process.