The conference on Artificial Intelligence in Interactive Digital Entertainment is underway. The proceedings are packed with useful information for game AI developers who want to stay up-to-date with the cutting edge. Below is a list of papers that can be found online for the posters track. Also see the AIIDE ‘07 Papers collected from the web.
Level Annotation and Test by Autonomous Exploration
This paper proposes the use of an autonomous exploring agent to generate and annotate the waypoint graph as an off-line process during level development. The explorer incrementally generates the waypoint graph as it explores the level via the same motion model used for player movement, and then revisits the waypoints to annotate them using image-based techniques. Points where the explorer becomes stuck or falls off of the level are flagged for later investigation by a level designer.
Level Annotation and Test by Autonomous Exploration (310 Kb) — Christian J. Darken
In video games, pathfinding must be done quickly and accurately. Not much computational time is allowed for pathfinding, but realistic looking paths are required. One approach to pathfinding which attempts to satisfy both of these constraints is to perform pathfinding on abstractions of the map. Botea et al.’s Hierarchical Pathfinding A* (HPA*) does this by dividing the map into square sectors and defining entrances between them. Although HPA* performs quick pathfinding which produces near-optimal paths, some improvements can be introduced. Here we discuss a faster path smoothing method, an alternative way to compute the weights of abstract edges, and lazy edge weight computations.
A Believable Agent for First-Person Shooter Games
This paper seems to be based on the Icarus architecture. From the site:
Icarus operates on a recognize-act cycle but, unlike many architectures, focuses on reactive execution of existing skills rather than on problem-space search. Given a top-level skill to pursue, on each cycle the system first checks the objective field for that skill.
If the objectives are true, nothing further needs to be done, but, if not, the interpreter examines the requirements to determine if the preconditions for action are met. If not, Icarus invokes a subskill associated with the failed requirement in an effort to satisfy it; otherwise, it selects one of the alternate means and calls on the primitive action or subskill associated with it.
A Cognitive Architecture for Physical Agents — Dongkyu Choi, Tolga Konik, Negin Nejati, Chunki Park, Pat Langley
Player Autonomy versus Designer Intent: A Case Study of Interactive Tour Guides
This paper explore the tradeoff between player autonomy and designer intent by simulating a system of autonomous museum tour guides. Visitors may have different art preferences or may wish to visit different exhibits on multiple visits. Often, these desires conflict. For example, visitors may wish to see the museum’s most popular work, but that could cause congestion, ruining the experience.
Thus, our task is to build a set of guides that can satisfy their visitors’ goals while also providing quality experiences for all. We present the results of a case study indicating that there is a space in the design spectrum between fully author-controlled narrative and complete player autonomy that reduces the frequency of bad experiences while allowing visitors to realize their own goals.
Player Autonomy versus Designer Intent: A Case Study of Interactive Tour Guides — David L. Roberts, Andrew S. Cantino, Charles L. Isbell
Motivational Ambient and Latent Behaviors in Computer RPGs
This paper seems to be based on ScriptEase. From the site:
Creating realistic artificially-intelligent characters is seen as one of the major challenges of the commercial games industry. Historically, character behavior has been specified using simple finite state machines and, more recently, by AI scripting languages. These languages are relatively “simple”, in part because the language has to serve three user communities: game designers, game programmers, and consumers - each with different levels of programming experience.
The scripting often becomes unwieldy, given that potentially hundreds (thousands) of characters need to be defined, the characters need non-trivial behaviors, and the characters have to interface with the plot constraints. We are currently developing a model for AI scripting called ScriptEase. The model is pattern template based, allowing designers to quickly build complex behaviors without doing explicit programming.
ScriptEase - A Scripting Language for Computer Role-Playing Game — Maria Cutumisu, Duane Szafron, Jonathan Schaeffer, Kevin Waugh
Emotionally Driven Natural Language Generation for Personality Rich Characters in Interactive Games
This paper addresses NLG for interactive games, presenting a novel template-based system that provides two distinct advantages over existing systems. First, the system not only works for dialogue, but enables a character’s personality and emotional state to influence the feel of the utterance. Second, the templates are resuable across characters, thus decreasing the burden on the game author.
Emotionally Driven Natural Language Generation for Personality Rich Characters in Interactive Games — Christina R. Strong, Manish Mehta, Kinshuk Mishra, Alistair Jones, Ashwin Ram
From Synthetic Characters to Virtual Actors
Discusses the extension of an emotionally-driven agent architecture already applied to the creation of emergent narratives. Synthetic characters are enhanced to perform as actors by carrying out a second cognitive appraisal, based on the OCC model, of the emotional impact of their projected actions before execution.
Also see the AIIDE ‘07 Papers that are available online.