US 12,403,402 B1
Multi-agent planning system for controlling non-player agents in a game
Christopher M. Ward, Somerville, MA (US); Trevor J. Santarra, Fremont, CA (US); Jeffrey D. Orkin, Arlington, MA (US); and Sarah M. Rudzki, Methuen, MA (US)
Assigned to Bitpart AI, Inc., Methuen, MA (US)
Filed by Bitpart AI, Inc., Methuen, MA (US)
Filed on Mar. 14, 2025, as Appl. No. 19/080,586.
Int. Cl. A63F 13/60 (2014.01); A63F 13/47 (2014.01); A63F 13/55 (2014.01); A63F 13/56 (2014.01); A63F 13/57 (2014.01); A63F 13/573 (2014.01); A63F 13/67 (2014.01); G06N 3/08 (2023.01); G06N 20/00 (2019.01); A63F 13/497 (2014.01); A63F 13/65 (2014.01); G06N 3/006 (2023.01)
CPC A63F 13/60 (2014.09) [A63F 13/47 (2014.09); A63F 13/55 (2014.09); A63F 13/56 (2014.09); A63F 13/57 (2014.09); A63F 13/573 (2014.09); A63F 13/67 (2014.09); G06N 3/08 (2013.01); G06N 20/00 (2019.01); A63F 13/497 (2014.09); A63F 13/65 (2014.09); G06N 3/006 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A method associated with a game platform, comprising:
responsive to an input, receiving a data set representing one or more transcripts, wherein a transcript represents a portion of a narrative and comprises a set of one or more items, with each item in the transcript comprising one of: an action, a line of dialogue, a reference to a task in the transcript or in some other transcript, and combinations thereof, wherein a task is a subset of the set of one or more items;
transforming the data set into re-composable combinations of tasks structured as a multi-modal hierarchical network of tasks, wherein an individual path in the hierarchical network of tasks represents a plan that is a particular implementation of a task that has one or more ways of being implemented; and
during a session of a game being controlled by a game engine associated with the game platform and involving a player, real-time orchestrating behaviors of one or more non-player agents by receiving a set of observations occurring in the game and, in response, implementing a set of one or more plans in the hierarchical network of tasks, wherein orchestration occurs remotely from the game engine;
wherein orchestrating behavior of a given agent includes receiving an observation message representing a game event, searching the multi-modal hierarchical network of tasks to identify a particular plan, based on the identified particular plan, generating a set of one or more cues, and providing the set of one or more cues to control the game engine in real-time.