Active Projects
Plot Expressivity in planning-based representations of story
Current planning-based models of story show promise in their ability to create a wide range of storylines with little or no human intervention. At their core, the knowledge representations they use match well with atomic elements of story structure. However, planning-based methods for combining action into sequences fall short of producing many critical structural properties of plot lines. This project seeks to add more expressive range to the generative capacity of efficient story planning systems. In particular, our current work looks at the addition of characters’ dynamic intentions, the performance of actions that fail, and the modeling of characters’ mistaken beliefs and how those false beliefs lead to more compelling stories.
More about this project.
intelligent game architectures
The Intelligent Game Architectures effort in the LN group is an ongoing effort to build architectures for game experiences that facilitate the integration of compute-intensive AI systems (primarily for narrative generation) with existing commercial game engines. The work has produced a cloud-based architecture that integrates AI planning and other intelligent systems with off-the-shelf game engines (e.g., Unity and Unreal). .
More about this project.
automated cinematography
The process of controlling a camera to tell a story, whether in a virtual world or on a real-world sound stage, relies on judgement about the conventions of films and television, the technical aspects of a camera and its operation within a specific environment and an underlying theory of camera shots as planned, intentional communicative action. Our work looks to provide both the foundational models of communicative action that underly cinematic conventions and low level tools and techniques for automatically creating effective cinematic communication within a 3D environment such as a game or virtual world.
More about this project.
Past Projects
ALICE: a model for sustaining technology-rich adaptive learning spaces and interactive content environments
The process of controlling a camera to tell a story, whether in a virtual world or on a real-world sound stage, relies on judgement about the conventions of films and television, the technical aspects of a camera and its operation within a specific environment and an underlying theory of camera shots as planned, intentional communicative action. Our work looks to provide both the foundational models of communicative action that underly cinematic conventions and low level tools and techniques for automatically creating effective cinematic communication within a 3D environment such as a game or virtual world.
More about this project.
circus
The Conflict and Intention Revision in Coherent User-focused Stories (CIRCUS) project team worked to develop new cognitively informed plan-based models of narrative action and to demonstrate that these models can be used both to control a virtual environment and to make effective predictions about the results of users’ mental models of the stories that they characterize. Motivated by psychological models of plans and plan reasoning, the team built on prior work in plan generation and plan-related communication to develop an architecture for creating understandable interaction in narrative-oriented virtual environments
More about CIRCUS
Creating Effective Task Descriptions from Action Plans
More about creating effective task descriptions.
declarative representation and reasoning in game-based learning environments
This project is developing task-based models of interaction within virtual worlds and leveraging cognitive and computational models of story to provide intelligent help and tutoring. The work seeks toprovide explicit and well-founded models of user misconceptions about a task domain and comparably well-defined methods for remediating those misconceptions in an automated fashion.
More about this project.
HI-FIVES
HI-FIVES was a project that sought to develop tools for novice game makers and integrate those tools in to middle school STEM courses throughout North Carolina public schools. The project, led by NC State’s Friday Insttute, and in collaboration with the NCSU Kenan Fellows and the NC Department of Public Instruction.
IC-CRIME
IC-CRIME is a revolutionary system that allows collaborative investigations of three dimensional representations of crime scenes. Crime scene investigators and other authorized personnel can collaborate with each other in viewing, exploring, and annotating these virtual crime scene models at any time from any web browser, anywhere in the world.
Within IC-CRIME, a group of investigators collaborate in a virtual meeting space, which appears in the form of a virtual investigative laboratory. From a given laboratory, multiple different crime scenes can be simultaneously accessed by an unlimited number of other investigators or forensic experts who no longer need to be physically co-located in order to collaborate on tasks related to the physical space of a crime scene.
More about IC-CRIME
Interactive Exploration of Complex Datasets Via the Effective Generation of Text and Graphics
The project on interactive exploration of complex datasets via the effective generation of text and graphics looked to integrate research on the automatic generation of visualizations and textual descriptions of datasets as a primary means for exploration of large amounts of data.
narrative for sensemaking
Stories provide one central mode for making sense of the world around us. We tell stories to others about our daily experiences, and tell stories to ourselves to better understand what has happened, what is happening and what might happen in the future. The Narrative for Sense-making project is a large, multi-year, multi-disciplinary project working to generate tools to support sense-making from real-world event data. The project is targeted at providing tools for intelligence analysts, who work to make sense of activities characterized within large amounts of desperate, low-level event data, but the research has applicability to a broad range of contexts. As an example, the initial domain being used in the project is the popular MOBA game Defense of the Ancients 2.
More about Narrative for Sense-Making
Planning-Based Visualization of Static Network Attack Analysis
The Planning-Based Visualization of Static Network Attack Analysis project, part of a larger project focused on Computer-aided Human Centric Cyber Situation Awareness, developed automated game-based tools for cyber-situation awareness. The Heavy Iron system read network activity logs and automatically produced 3D movie-like cinematics that conveyed the activities of human attackers on a network, highlighting relevant actions, goals and objectives and other aspects of the attackers’ plans, using Hollywood-style cinematic conventions.
More about Planning-Based Visualization of Static Network Attach Analysis