Developing training scenarios that induce a trainee to utilize specific skills is one of the facets of simulation-based training that requires significant effort. Simulation-based training systems have become more complex in recent years. Because of this added complexity, the amount of effort required to generate and maintain training scenarios has increased. This thesis describes an investigation into automating the scenario generation process. The Automated Scenario Generation System (ASGS) generates expected action flow as contexts in chronological order from several events and tasks with estimated time for the entire training mission. When the training objectives and conditions are defined, the ASGS will automatically generate a scenario, with some randomization to ensure no two equivalent scenarios are identical. This makes it possible to train different groups of trainees sequentially who may have the same level or training objectives without using a single scenario repeatedly. The thesis describes the prototype ASGS and the evaluation results are described and discussed. SVSTM Desktop is used as the development infrastructure for ASGS as prototype training system.

To create a realistic environment, many simulations require simulated agents with human behavior patterns. Manually creating such agents with realistic behavior is often a tedious and time-consuming task. This dissertation describes a new approach that automatically builds human behavior models for simulated agents by observing human performance. The research described in this dissertation synergistically combines Context-Based Reasoning, a paradigm especially developed to model tactical human performance within simulated agents, with Genetic Programming, a machine learning algorithm to construct the behavior knowledge in accordance to the paradigm. This synergistic combination of well-documented AI methodologies has resulted in a new algorithm that effectively and automatically builds simulated agents with human behavior. This algorithm was tested extensively with five different simulated agents created by observing the performance of five humans driving an automobile simulator. The agents show not only the ability/capability to automatically learn and generalize the behavior of the human observed, but they also capture some of the personal behavior patterns observed among the five humans. Furthermore, the agents exhibited a performance that was at least as good as agents developed manually by a knowledgeable engineer.

This dissertation explores modeling collaborative behavior, based on Joint Intentions Theory (JIT), in Context-Based Reasoning (CxBR). Context-Based Reasoning is one of several contextual reasoning paradigms. And, Joint Intentions Theory is the definitive semantic framework for collaborative behaviors. In order to formalize collaborative behaviors in CxBR based on JIT, CxBR is first described in terms of the more popular Belief, Desire, and Intention (BDI) model. Once this description is established JIT is used as a basis for the formalism for collaborative behavior in CxBR. The hypothesis of this dissertation is that this formalism allows for effective collaborative behaviors in CxBR. Additionally, it is also hypothesized that CxBR agents inferring intention from explicitly communicating Contexts allows for more efficient modeling of collaborative behaviors than inferring intention from situational awareness. Four prototypes are built and evaluated to test the hypothesis and the evaluations are favorable. Effective collaboration is demonstrated through cognitive task analysis and through metrics based on JIT definitions. Efficiency is shown through software metric evaluations for volume and complexity of code.

This research focuses on the ability to contextualize observed human behaviors in efforts to automate the process of tactical human performance modeling through learning from observations. This effort to contextualize human behavior is aimed at minimizing the role and involvement of the knowledge engineers required in building intelligent Context-based Reasoning (CxBR) agents. More specifically, the goal is to automatically discover the context in which a human actor is situated when performing a mission to facilitate the learning of such CxBR models. This research is derived from the contextualization problem left behind in Fernlund's research on using the Genetic Context Learner (GenCL) to model CxBR agents from observed human performance [Fernlund, 2004]. To accomplish the process of context discovery, this research proposes two contextualization algorithms: Contextualized Fuzzy ART (CFA) and Context Partitioning and Clustering (COPAC). The former is a more naive approach utilizing the well known Fuzzy ART strategy while the latter is a robust algorithm developed on the principles of CxBR. Using Fernlund's original five drivers, the CFA and COPAC algorithms were tested and evaluated on their ability to effectively contextualize each driver's individualized set of behaviors into well-formed and meaningful context bases as well as generating high-fidelity agents through the integration with Fernlund's GenCL algorithm. The resultant set of agents was able to capture and generalized each driver's individualized behaviors.