Artificial intelligence for emulating human playstyles

1. A non-transitory, computer-readable storage medium storing computer readable instructions that, when executed by one or more computing devices, causes the one or more computing devices to perform operations comprising:

generating an image indicating positional information about objects in a virtual environment within a video game, the objects including at least one computer-controlled character;

generating data including non-positional information about the objects in the virtual environment;

selecting a first artificial intelligence (AI) model from among a plurality of AI models based at least in part on the image and the data;

providing the image and the data as inputs to the first AI model, wherein the first AI model is configured to:

receive, as model inputs, the image and the data;

process the model inputs; and

generate, as outputs, at least one command to control operation of the computer-controlled character;

executing the at least one command to control operation of the computer-controlled character within the virtual environment; and

generating instructions to output the execution of the at least one command by the computer-controlled character within the virtual environment during runtime execution of the video game.

2. The non-transitory, computer-readable storage medium of claim 1 , wherein generating the image includes generating a retina image focused on the computer-controlled character, the retina image divided into a plurality of cells each corresponding to a location in the virtual environment, wherein the cell includes an indicator of which objects from the virtual environment are positioned in the corresponding locations.

3. The non-transitory, computer-readable storage medium of claim 2 , wherein boundaries of the plurality of cells are defined by concentric shapes and by slices of the concentric shapes.

4. The non-transitory, computer-readable storage medium of claim 1 , wherein the image is a matrix image, each cell of the matrix image corresponding to a location in the virtual environment.

5. The non-transitory, computer-readable storage medium of claim 1 , wherein:

the image is a matrix image comprising a plurality of matrix elements;

each matrix element corresponds to a location in the virtual environment;

a value, color, or visual indicator of the matrix elements indicate relational positions of objects in the corresponding locations; and

the matrix image is separate from a video output frame of the video game.

6. The non-transitory, computer-readable storage medium of claim 5 , wherein the matrix elements are mapped to the corresponding locations on a retina image that is focused on the computer-controlled character.

7. The non-transitory, computer-readable storage medium of claim 1 , wherein processing the model inputs includes processing the image using convolutional neural network analysis.

8. The non-transitory, computer-readable storage medium of claim 1 , wherein the operations further include:

checking, against hard coded software rules, consequences of causing computer-controlled character to execute the control operation; and

causing the computer-controlled character to behave in the video game according to the control operation based at least in part on the consequences.

9. The non-transitory, computer-readable storage medium of claim 1 , wherein the computer-controlled character has a role similar to a role of a human-controllable character in the video game.

10. A computer-implemented method for controlling a computer-controlled character, the method comprising:

generating an image indicating positional information about objects in a virtual environment of a video game, the objects including at least the computer-controlled character and a player controlled character;

generating data including non-positional information about the objects in the virtual environment;

selecting a first artificial intelligence (AI) model from among a plurality of AI models based at least in part on the image and the data;

providing the image and the data as inputs to the first AI model, wherein the first AI model is configured to:

receive, as model inputs, the image and the data;

process the model inputs; and

generate, as outputs, at least one command for controlling operation of the computer-controlled character; and

executing the at least one command to control operation of the computer-controlled character within the virtual environment.

11. The computer-implemented method of claim 10 , wherein generating the image includes generating a retina image focused on the computer-controlled character, the retina image divided into a plurality of cells each corresponding to a location in the virtual environment, wherein the cell includes an indicator of which objects in the virtual environment are positioned in the corresponding locations.

12. The computer-implemented method of claim 10 , wherein generating the image further includes generating a matrix image, each cell of the matrix image corresponding to a location in the virtual environment and indicating objects of the virtual environment that are positioned in the corresponding locations.

13. The computer-implemented method of claim 12 , wherein locations are defined by concentric shapes and by slices of the concentric shapes.

14. The computer-implemented method of claim 10 , wherein the image is a matrix image comprising a plurality of matrix elements,

wherein each matrix element corresponds to a location in the virtual environment;

wherein a value, color, or visual indicator of the matrix elements indicate relational positions of objects in the corresponding locations; and

wherein the matrix image is separate from a video output frame of the video game.

15. The computer-implemented method of claim 10 , wherein processing the model inputs includes processing the image using convolutional neural network analysis.

16. The computer-implemented method of claim 10 , wherein the computer-controlled character has a role similar to a player-controllable character in the video game.

17. A system for controlling a non-player character (NPC), the system comprising:

a player input device;

a non-transitory computer-readable storage medium configured to store code for a video game and a plurality of artificial intelligence (AI) models;

one or more processors configured to execute the code to perform operations comprising:

generating an image indicating positional information about objects in a virtual environment of the video game, the objects including at least a computer-controlled character and a player controlled character;

generating data including non-positional information about the objects in the virtual environment;

selecting a first AI model from among the plurality of AI models based at least in part on the image and the data;

providing the image and the data as inputs to the first AI model, wherein the first AI model is configured to:

receive, as model inputs, the image and the data;

process the model inputs; and

generate, as outputs, at least one command for controlling the computer-controlled character;

executing the at least one command to control operation of the computer-controlled character within the virtual environment; and

executing a command received through the player input device to control a player-controlled character within the virtual environment.

18. The system of claim 17 , wherein the operations further include:

checking, against hard coded software rules, consequences of causing computer-controlled character to execute the control operation; and

causing the computer-controlled character to behave in the video game according to the control operation based at least in part on the consequences.

19. The system of claim 17 , wherein:

the operations further include:

receiving, through the player input device, a selection of an AI model preference; and

wherein selecting the first AI model from among a plurality of AI models is further based at least in part on the AI model preference.