VIRTUAL AGENT

• US 20170206095A1
• Filed: 01/14/2016
• Published: 07/20/2017
• Est. Priority Date: 01/14/2016
• Status: Active Grant

• Current Assignee: Samsung Electronics Co. Ltd.
• Original Assignee: Samsung Electronics Co. Ltd.
• Inventors: GIBBS, Simon J., LIOT, Anthony, SONG, Yu
• Granted Patent: US 10,664,741 B2
• Time in Patent Office: Days
• CPC Class Codes: G06F 2203/011 Emotion or mood input determined on the basis of sensed human body parameters such as pulse, heart rate or beat, temperature of skin, facial expressions, iris, voice pitch, brain activity patterns
• CPC Class Codes: G06F 3/011 Arrangements for interaction with the human body, e.g. for user immersion in virtual reality blind teaching G09B21/00
• CPC Class Codes: G06F 3/167 Audio in a user interface, e.g. using voice commands for navigating, audio feedback
• CPC Class Codes: G06F 9/453 Help systems
• CPC Class Codes: G06N 3/006 based on simulated virtual individual or collective life forms, e.g. social simulations or particle swarm optimisation [PSO]
• CPC Class Codes: G06T 13/40 of characters, e.g. humans, animals or virtual beings

Abstract

In one aspect, a method for operating a virtual agent will be described. An interaction context is obtained. An agent state is determined based on the obtained interaction context. The agent state indicates an activity of the virtual agent. The emotion of the virtual agent is updated based on the obtained interaction context, the determined agent state and/or a personality of the virtual agent. One or more behaviors are selected. Each behavior involves or indicates a change in an appearance of the virtual agent or generation of audio. Various embodiments relate to devices, servers, software and systems arranged to implement one or more of the above operations.

Claims

1. A method of operating a virtual agent, comprising:

obtaining an interaction context, the interaction context including at least one of a communication context, target information, situation information and people information;

determining an agent state based on the obtained interaction context, the agent state indicating an activity of the agent;

updating emotion of the virtual agent based on the obtained interaction context, the determined agent state and a personality of the virtual agent; and

selecting one or more behaviors based on the obtained interaction context and the updated emotion of the virtual agent wherein each behavior involves an adjustment of an appearance of the virtual agent or generation of audio.

2. The method of claim 1 further comprising:

determining the interaction context by sensing an environment around a device.

3. The method of claim 1, further comprising:

displaying a virtual agent model that visually represents the virtual agent; and

rendering the one or more behaviors at the virtual agent model, which changes the appearance of the virtual agent model.

4. The method of claim 1 wherein the same interaction context has a different effect on emotion of a second virtual agent because of a different personality of the second virtual agent.

5. The method of claim 1 wherein each behavior is associated with one or more tags, each tag being associated with at least one of an activity, emotion, mood and personality.

6. The method of claim 1 wherein the personality is not adjusted based on the interaction context.

7. The method of claim 1 wherein the personality is adjustable by a user.

8. The method of claim 1 further comprising:

updating mood of the virtual agent based on the interaction context, the agent state and the personality of the virtual agent wherein the mood is arranged to change more slowly in response to the interaction context than the emotion.

9. The method of claim 8 wherein the emotion includes one or more emotion parameters and wherein the mood is based on a moving average of the emotion parameters over a predetermined period of time.

10. The method of claim 1 further comprising:

obtaining a behavior corpus, which includes a plurality of predefined behaviors;

obtaining a rule set including a plurality of predefined rules, each rule being associated with a selector and a trigger, the trigger being a condition that matches one or more of a particular interaction context, emotion and agent state wherein the selector is a search criteria used to search the behavior corpus; and

when at least one of the obtained interaction context, updated emotion and determined agent state match a trigger of a rule in the rule set, selecting the one or more behaviors based on the selector associated with the matching trigger and rule.

11. The method of claim 10, further comprising:

choosing one or more candidate behaviors based on the selector and the updated emotion;

determining that two or more of the candidate behaviors conflict; and

selecting one of the conflicting candidate behaviors for implementation based on a priority indicator.

12. The method of claim 11 wherein candidate behaviors conflict when the conflicting candidate behaviors involve adjusting the same part of a body simulated by the virtual agent.

13. The method of claim 1 further comprising:

rendering the virtual agent at a plurality of frames per second; and

performing the method on a frame by frame basis.

14. The method of claim 1 wherein the interaction context is obtained using a network of two or more sensor devices.

15. The method of claim 1 wherein the virtual agent is represented by a graphical model that is three-dimensional.

16. A computer readable storage medium including executable computer code embodied in a tangible form wherein the computer readable medium comprises:

executable computer code operable to obtain an interaction context, the interaction context including at least one of a communication context, target information, situation information and people information;

executable computer code operable to determine an agent state based on the obtained interaction context, the agent state indicating activity of the agent;

executable computer code operable to update emotion of the virtual agent based on the obtained interaction context, the determined agent state and a personality of the virtual agent; and

executable computer code operable to select one or more behaviors based on the obtained interaction context and the updated emotion wherein each behavior involves an adjustment of an appearance of the virtual agent or generation of audio.

17. A device comprising:

at least one processor; and

memory circuitry including a computer readable storage medium that is arranged to store computer code in a tangible form wherein the computer code, when executed by the at least one processor, causes the device to;

obtain an interaction context, the interaction context including at least one of a communication context, target information, situation information and people information;

determine an agent state based on the obtained interaction context, the agent state indicating activity of the agent;

update emotion of the virtual agent based on the obtained interaction context, the determined agent state and a personality of the virtual agent; and

select one or more behaviors based on the obtained interaction context and the updated emotion wherein each behavior involves an adjustment of an appearance of the virtual agent or generation of audio.

18. The device of claim 17 further comprising:

a sensor unit including one or more sensors wherein the interaction context is based at least in part on sensor data received from the sensor unit.

19. The device of claim 17 further comprising a microphone that is arranged to receive audio information and wherein the interaction context is based at least in part on the audio information received using the microphone.

20. The device of claim 17 is one of a refrigerator, a laptop computer, a computer, a tablet, a phone, a watch, glasses and a virtual reality device.