US 12,268,964 B2
World prediction within a game application environment
Dustin Sutherland, Coquitlam (CA); and Toru Tokuhara, New Westminster (CA)
Assigned to Electronic Arts Inc., Redwood City, CA (US)
Filed by Electronic Arts Inc., Redwood City, CA (US)
Filed on Mar. 30, 2023, as Appl. No. 18/193,554.
Prior Publication US 2024/0325916 A1, Oct. 3, 2024
Int. Cl. A63F 13/577 (2014.01); A63F 13/57 (2014.01)
CPC A63F 13/577 (2014.09) [A63F 13/57 (2014.09); A63F 2300/643 (2013.01)] 20 Claims
OG exemplary drawing
 
1. A computer-implemented method for world prediction within a game application, the method comprising:
by a hardware processor configured with computer executable instructions,
executing a game application comprising a game environment, the game environment comprising a plurality of virtual objects, wherein each of the plurality of virtual objects is associated with positional data;
generating a plurality of instances, wherein each instance comprises a future projection of a first set of the plurality of virtual objects by a set number of frames;
determining a position of a bounding box associated with each of the plurality of virtual objects in the first set in each of the plurality of instances based at least in part on the positional data corresponding to individual virtual objects of the first set;
predicting a collision between two or more virtual objects of the plurality of virtual objects in the first set, wherein predicting the collision comprises:
determining if overlap occurs between individual bounding boxes of the two or more virtual objects,
in response to a determination that the individual bounding boxes of the two or more virtual objects overlap,
generating at least one nearfield bounding box for each of the two or more virtual objects, and
determining if overlap occurs between the at least one nearfield bounding boxes of the two or more virtual objects;
in response to a determination that individual nearfield bounding boxes of the two or more virtual objects overlap, generating collision data based on a predicted collision; and
generating an action for at least one virtual object of the plurality of virtual objects based on the collision data.