CPC A63F 13/60 (2014.09) [A63F 13/42 (2014.09); A63F 13/55 (2014.09); G06N 20/00 (2019.01)] | 20 Claims |
1. A method for automating placement of props on a digital map, comprising:
based on one or more training maps on which props have been placed in an ideal placement, training a machine learning mechanism to output placements, on input target maps, for sets of input target props;
receiving, as input to a prop placement tool, a target map and a set of target props;
wherein the target map defines a set of map structures;
for each target prop in the set of target props, executing the prop placement tool on one or more computing devices to cause the prop placement tool to automatically determine a final placement for the target prop on the target map by:
from a set of possible placements of the target prop on the target map, the prop placement tool determining a set of valid placements for the target prop based, at least in part, on one or more spatial rules;
for each valid placement in the set of valid placements, the prop placement tool generating a prop distribution score based on output of the machine learning mechanism;
for each valid placement in the set of valid placements, the prop placement tool determining a placement score based, at least in part, on the prop distribution score that is generated for the valid placement;
selecting, by the prop placement tool, as the final placement for the target prop on the target map, a valid placement based on one or more placement scores generated for one or more valid placements of the target prop;
generating output, by the prop placement tool, that indicates a placement outcome that includes the final placement for each target prop in the set of target props; and
based on the final placement for each target prop in the set of target props, generating a version of the target map in which the set of target props have been placed on the target map.
|