CyberAIMs: a tool for teaching adversarial and systems thinking

"CyberAIMs stands for Cyber Agents’ Interactive Modeling and Simulation. We designed this tool in order to use it as an educational tool to teach Master students in a Cyber security course. This paper aims to describe the model and explain the design choices behind CyberAIMs in terms of associating them with the emerging concepts within cyber security curriculum, namely adversarial and systems thinking. The preliminary results indicate that the current distribution of values and entities allows most of the defense agents to avoid losing all their resources to their attack counterparts. We intend to use this tool as part of a lab with students in Information Security and further extend our target users, by including others who need training in adversarial and systems thinking. We conclude by providing rough results from running simulations with the tool and giving further directions of our future research, in order to improve the usability and level of detail for this tool."

Machine learning for approximated sensors for Marksmanship training

"All people in the military must be proficient on the basics – to shoot, move, and communicate. Basic Rifle Marksmanship (BRM) is required for both noncommissioned and enlisted Warfighters in all branches of military service, with training on BRM skills being conducted in a series of “dry fire”, simulation, and live drills. In all phases of training, Warfighters receive instruction on the four fundamentals of shooting: breathing, body position, sight picture, and trigger squeeze. Within simulation, this training is conducted in a 1:4 to 1:8 range; one instructor per 4-8 students. While realtime individualized feedback is a goal of instruction, it is not a reality, as instructors must attend to the needs of many students. In an effort to aid instructors in providing valuable individualized feedback, a tutoring system was developed which automatically diagnosed novice performance when compared to experts. This system was used to diagnose novice performance with extensive physical sensors applied to the weapons. This work investigates whether machine learning can aid in the diagnosis of the novice performance, without the physical sensors, and analyzes the degree to which the sensors are necessary."

"A hybrid machine learning approach to Automated Scenario Generation (ASG) to support adaptive instruction in virtual simulations and games"

"This paper examines machine learning methods to automatically generate a large number of child scenarios from a small number of parent scenarios in support of adaptive instruction conducted in virtual simulations and game-based platforms. Adaptive instructional systems (AISs) include Intelligent Tutoring Systems (ITSs), intelligent mentors, recommender systems, personal assistants, and intelligent instructional media. AISs attempt to tailor instruction for individuals and teams based on their learning needs (e.g., knowledge or skill deficiencies), goals, and preferences. This often requires much more content than current non-adaptive systems which provide one or a very limited set of training scenarios to address a given set of learning objectives. The goal of the research described in this paper is to reduce the authoring burden for developing a large number of unique and relevant training scenarios. The methodology presented also ranks the resulting scenarios with respect to a set of author-specified learning objectives and learner/team competency in the domain of instruction. The unique contributions of this paper are tied to its hybrid machine learning approach, and consideration for both learning objectives and learner/team competency in automatically ranking generated scenarios."

"Stratagems: an innovative approach for increasing cognitive authenticity in gamebased environments"

"Combat Search and Rescue aircrew are tasked with challenging missions, usually under conditions of time pressure, dynamic conditions, and a high degree of uncertainty. Many skills required for successful outcomes (e.g., solving problems quickly; accurately evaluating risks; adapting to rapidly changing environmental conditions) are acquired and maintained through first-hand experiences. This paper describes a research project to develop a pedagogically effective game-based trainer for cognitive skills required for Combat Rescue Helicopter aircrew. Training scenarios are developed using cognitive interviewing techniques in order to promote cognitive authenticity. The trainer will pose critical decisions and provide expert feedback utilizing ShadowBox®, a proven coaching technique that enables trainees to obtain insight into the decisionmaking processes and reasoning of experts."

Modelling and simulation of a fire department’s response to emergency incidents

"After statistical analysis of the database of a fire department covering eight years of consecutive incident records from January 2009 to December 2016, we developed a modelling and simulation (M&S) approach that could be replicated for fire departments across Canada. Our M&S framework involved two different simulation models running on separate platforms: (i) an Incident Generation Engine, which simulates the ‘arrival’ of emergency incidents, and (ii) a Response Simulation Model. The first model is a discrete event simulation model using CPNTools 4.0, generating inputs for the second model, which is an agent-based simulation model developed using AnyLogic. We discuss the principal elements of the two simulation models, and report on findings from our simulation experiments."

Holovulcano: Augmented Reality simulation of volcanic eruptions

"This paper describes an interactive holographic simulation of volcanic eruption. The aim of the project is to use Augmented Reality (AR) technology to visualize different volcanic eruptions for public education, emergency training, and preparedness planning purposes. To achieve this goal, a 3D model of the entire Vulcano Island in Italy has been created using real elevation data. Unity game engine and Microsoft Visual Studio have been used to develop HoloVulcano augmented/virtual reality simulation application. The current version of HoloVulcano simulates normal and unrest situations, single and long lasting Vulcanian, Plinian, and Strombolian eruptions. HoloVulcano has been developed for Microsoft HoloLens AR device. Wearing the HoloLens, users can interact with the volcano through voice, gazing, and gestures and view different eruptions from different points in the island. HoloVulcano will be used for training emergency exercises and public education."

Simulation of crisis affecting critical infrastructures and industrial plants

"This paper proposes a model for reproducing human behaviors and it is devoted to evaluate the operation performance in case of crisis with special attention to cases involving multiple actors such as fire-fighters, police, ambulances, civil protection and national reserve. The paper present the general scheme and the data structure used to conduct experiments related to cases of flooding and hazardous material spills in industrial and urban areas."

"Intelligent tutoring in the wild: leveraging mobile app technology to guide live training"

"Mobile computing technologies are extending how people can interact with educational and training content in a whole new way. Through high resolution displays, intuitive user interfacing, embedded sensing technologies, and well supported app development communities, there is a plethora of content that can be used to build effective materials that target knowledge and skill development. To truly enhance this new training paradigm, extending Intelligent Tutoring System (ITS) to support mobile interactions can provide a new means to managing training in a rich contextualized environment. In this instance, learning takes place in the natural environment where directed experiences focus on the elements surrounding one’s location. In this paper, we describe the development of a new mobile ITS application using the Generalized Intelligent Framework for Tutoring (GIFT; Sottilare, Goldberg, Brawner & Holden, 2012). The domain of land navigation was applied as a use case, with direct support for the United States Military Academy at West Point. We describe the training concept, how GIFT was extended to support this concept from an architectural and assessment standpoint, along with implementation plans for an initial training effectiveness study."

Crime scene evidence collection as a Virtual Reality use case

"Virtual Reality (VR) is an interface strategy that completely immerses the user into a virtual environment. Users with head mounted displays (HMD) see and hear the immersive space as if they are no longer in the real world. Instead they are fully wrapped in the surroundings of the virtual space. While VR has been around for many years, recent advances in commercial technology have provided a surge of costeffective devices to mainstream end-users. With this surge in hardware availability, software developers have been exploring ways to make use of VR to design engaging training experiences. This paper describes the creation of a VR simulation to train the investigative mindset relative to evidence collection tasks at a crime scene. Spatial orientation within a crime scene is critical to understand blood splatter, munitions trajectory and to conceptually re-create events in the crime scene. This paper will explain the challenges and solutions of developing this VR experience. Finally, the paper will provide lessons learned and recommended future development involving this VR training application."

"Model and simulation of a real water distribution network to support emergency plans"

"This paper presents a water distribution network model to evaluate the impact of adverse events, such as faults and/or cyber-attacks, on a real water distribution system in a wider context which involves the interdependency with the electrical grid, in the frame of the Horizon 2020 project ATENA. The model has been developed by using a commercial simulator, which can address both the electrical and the water domain. Specific features and modules have been added to the simulator, in order to calculate the water level in tanks – an important and missing metric to support emergency plans. The interdependency among the electricity grid and the water network is considered throughout pumps, which are at the same time electrical loads and hydraulic devices. Two use cases, dealing with faults or cyber-attacks against the electrical grid affecting critical pumps or treatment stations, are investigated and the simulation results are reported."