Research
Bitpart’s AI NPC Engine is based in SCIENCE. We employ an approach called Episodic Planning, which is an evolution of two threads of research that originated in the game industry and at MIT. Dr. Jeff Orkin pioneered the use of automated planning systems in games with his work on the GOAP system for FEAR. At MIT, Jeff explored learning generative behavior and dialogue from human data with The Restaurant Game research project. Bitpart’s AI Director is powered by an HTN planner that we learn from a combination of LLM-generated narrative, human demonstrations, and other existing content (e.g. design docs, hand-written stories, movie scripts, etc).
Media
Peer-Reviewed Publications
Orkin (2006), Three States and a Plan: The AI of F.E.A.R. Proceedings of the Game Developers Conference (GDC).
Orkin (2005), Agent Architecture Considerations for Real-Time Planning in Games. Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE).
Orkin (2004), Symbolic Representation of Game World State: Toward Real-Time Planning in Games. Proceedings of the AAAI Workshop on Challenges in Game AI.
Orkin (2003), Applying Goal-Oriented Planning for Games. AI Game Programming Wisdom 2, Charles River Media.
Orkin (2013), Collective Artificial Intelligence: Simulated Role-Playing from Crowdsourced Data. PhD Thesis, MIT Media Lab.
Orkin & Roy (2012), Understanding Speech in Interactive Narratives with Crowdsourced Data. Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE).
Orkin, Smith, Reckman & Roy (2010), Semi-Automatic Task Recognition for Interactive Narratives with EAT & RUN. Proceedings of the 3rd Intelligent Narrative Technologies Workshop (INT3).
Orkin & Roy (2009), Automatic Learning and Generation of Social Behavior from Collective Human Gameplay. Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems (AAMAS).
Orkin & Roy (2007), The Restaurant Game: Learning Social Behavior and Language from Thousands of Players Online. Journal of Game Development, 3(1), 39-60.
Related Work
The AI of Horizon Zero Dawn. The Guerrilla Games blog.
Humphreys (2013). Exploring HTN Planners through Example. Game AI Pro: Collected Wisdom of Game AI Professionals.
HTN Planning in Transformers: Fall of Cybertron. AI and Games #14.
Gorniak & Davis. (2007). SquadSmart: Hierarchical Planning and Coordinated Plan Execution for Squads of Characters. Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE).
Hoang, Lee-Urban, and Munoz-Avila. (2005). Hierarchical Plan Representations for Encoding Strategic Game AI. Proceedings of the Artificial Intelligence and Interactive Digital Entertainment Conference (AIIDE).
Bartheye & Jacopin. (2008). Connecting PDDL-based off-the-shelf planners to an arcade game. Proceedings of the European Conference on Artificial Intelligence, Workshop on Artificial Intelligence in . (ECAI AIG).
Burke et al. (2001). Creature Smarts: The Art and Architecture of a Virtual Brain. Proceedings of the 2001 Computer Game Developers Conference (CGDC).
Ghallab et al. (1998). The Planning Domain Definition Language V1.2 (PDDL). The 1998 AI Planning Systems Competition. The Fourth International Conference on Artificial Intelligence (AIPS).
Nau et al. (1999). Simple Hierarchical Ordered Planner (SHOP). Proceedings of the Sixteenth International Joint Conference on Artificial Intelligence (IJCAI).
Rich & Sidner. (1999). COLLAGEN: A Collaboration Manager for Software Interface Agents. User Modeling and User-Adapted Interaction 8(3).
Grosz & Sidner. (1986). Attention, Intentions, and the Structure of Discourse. Computational Linguistics 12(3).
Traverso, Ghallab, and Nau. (2004). Automated Planning: Theory & Practice. Morgan Kaufmann Publishers.
Nilsson. (1998). Artificial Intelligence: A New Synthesis. Morgan Kaufmann Publishers.
Russell & Norvig. (1995). Artificial Intelligence: A Modern Approach. Prentice Hall.