SYSTEMS AND METHODS FOR AUTONOMOUS ADAPTATION OF AN AUTOMATION CONTROL SERVICE

• US 20150309492A1
• Filed: 03/25/2015
• Published: 10/29/2015
• Est. Priority Date: 04/29/2014
• Status: Active Grant

• Current Assignee: Cox Communications Incorporated (Cox Enterprises Incorporated)
• Original Assignee: Cox Communications Incorporated (Cox Enterprises Incorporated)
• Inventors: Patel, Dipan, Finnerty, Nadine, Shrum, Edgar V. Jr., Goldberg, Andrew, Taylor, Haynes
• Granted Patent: US 10,983,487 B2
• Time in Patent Office: Days
• US Class Current: 1/1
• CPC Class Codes: G05B 15/02 electric
• CPC Class Codes: G06N 20/00 Machine learning
• CPC Class Codes: G06N 5/04 Inference or reasoning models
• CPC Class Codes: G06N 7/01 Probabilistic graphical models, e.g. probabilistic networks
• CPC Class Codes: H04L 12/2803 Home automation networks

Abstract

Aspects of the disclosure relate to intelligent automation control of an environment based at least on intelligence associated with behavior of an operator and/or equipment.

Claims

1. A method for automation control, comprising:

(a) accessing, by a computing system comprising at least one memory device and at least one processor, one or more information streams from a group of agents associated with an operational environment;

(b) determining, by the computing system, an environment condition of the operational environment based on at least a portion of the one or more information streams;

(c) determining, by the computing system, an agent state of an agent of the group of agents based on at least a portion of the one or more information streams; and

(d) adjusting, by the computing system, an automation control service based at least on one or more of the environment condition or the agent state, the automation control service configured to regulate at least one agent of group of agents.

2. The method of claim 1, further comprising repeating (a), (b), (c), and (d) at least one time.

3. The method of claim 1, further comprising determining, by the computing system, intelligence representative of a behavior of the agent based on at least a portion of the one or more information streams, and

determining historical conditions of the operational environment based on at least a portion of the one or more information streams.

4. The method of claim 3, further comprising predicting a second environment condition of the operational environment based at least on the determined historical conditions, and

predicting a second agent state of the agent based at least on the determined intelligence.

5. The method of claim 4, further comprising further adjusting, by the computing system, the automation control service based at least on one or more of the predicted second environment condition or the predicted second agent state.

6. The method of claim 3, wherein determining, the intelligence comprises generating, by the computing system, a model of the behavior of the agent based at least on unsupervised learning, wherein the model comprises one or more of statistical Markov model, a neural network model, or a Gaussian mixture model.

7. The method of claim 3, wherein determining, by the computing system, the intelligence comprises generating, by the computing system, a model of the behavior of the agent based at least on supervised learning, wherein the model comprises one or more of statistical Bayes model, a neural network model, or regression analysis.

8. The method of claim 3, wherein determining, by the computing system, the intelligence comprises determining a correlation between events associated with one or more of the agent or at least one second agent of the group of agents.

9. The method of claim 3, wherein determining, by the computing system, the historical conditions comprises:

determining a pattern of environment conditions, the pattern comprising one or more of (i) a pattern of events within the operational environment, (ii) a time-domain pattern in the at least one of the one or more information streams, or (iii) a space-domain pattern in the at least one of the one or more information streams;

or

determining correlations between patterns in the at least one of the one or more information streams.

10. A system, comprising:

at least one memory device having instructions encoded thereon; and

at least one processor coupled to the at least memory device and configured to execute the instructions and, in response, further configured at least to;

(a) access one or more information streams from a group of agents associated with an operational environment;

(b) determine an environment condition of the operational environment based on at least a portion of the one or more information streams;

(c) determine an agent state of an agent of the group of agents based on at least a portion of the one or more information streams; and

(d) adjust an automation control service based at least on one or more of the environment condition or the agent state, the automation control service configured to regulate at least one agent of group of agents.

11. The system of claim 10, wherein the at least one processor is further configured, in response to execution of the instructions, to repeat (a), (b), (c), and (d) at least one time.

12. The system of claim 10, wherein the at least one processor is further configured, in response to execution of the instructions, to:

determine intelligence representative of a behavior of the agent based on at least a portion of the one or more information streams, and

determine historical conditions of the operational environment based on at least a portion of the one or more information streams.

13. The system of claim 12, wherein the at least one processor is further configured, in response to execution of the instructions, to:

predict a second environment condition of the operational environment based at least on the determined historical conditions, and

predict a second agent state of the agent based at least on the determined intelligence.

14. The system of claim 13, wherein the at least one processor is further configured, in response to execution of the instructions, to further adjust the automation control service based at least on one or more of the predicted second environment condition or the predicted second agent state.

15. The system of claim 12, wherein to determine the intelligence further comprises generate a model of the behavior of the agent based at least on unsupervised learning, wherein the model comprises one or more of statistical Markov model, a neural network model, or a Gaussian mixture model.

16. The system of claim 12, wherein to determine the intelligence further comprises to generate a model of the behavior of the agent based at least on supervised learning, wherein the model comprises one or more of statistical Bayes model, a neural network model, or regression analysis.

17. The system of claim 12, wherein to determine the intelligence further comprises to determine a correlation between events associated with one or more of the agent or at least one second agent of the group of agents.

18. The system of claim 12, wherein to determine the historical conditions further comprises:

determine a pattern of environment conditions, the pattern comprising one or more of (i) a pattern of events within the operational environment, (ii) a time-domain pattern in the at least one of the one or more information streams, or (iii) a space-domain pattern in the at least one of the one or more information streams;

or

determine correlations between patterns in the at least one of the one or more information streams.