Granted Patent B2

US 12,254,312 · App. 17/806,794 · Granted Mar 18, 2025

Inventors: Saraswathi Sailaja Perumalla (Visakhapatnam, IN); Sarbajit K. Rakshit (Kolkata, IN); Sowjanya Rao (Hyderabad, IN)

Assignee: International Business Machines Corporation

G06F8/71G06F8/30G06F11/3013G06F11/302G06F11/3409G06F8/60G07C5/0808

Abstract

Claims (49)

1. A computer-implemented method (CIM) comprising:

monitoring, using a set of sensor device(s), an artificial intelligence (AI) enabled edge device from a set of AI enabled edge devices storing edge device control code in a code repository including a plurality of code branches corresponding to a plurality of contextual scenarios;

during the monitoring, operating the AI enabled edge device in a first contextual scenario;

analyzing, by the AI enabled edge device and based on the monitoring, performance of the AI enabled edge device in the first contextual scenario to identify a code branch name for the first contextual scenario;

detecting that the AI enabled edge device is self-adapting code for the first contextual scenario; and

in response to the detecting and based on the identified code branch name, determining whether the plurality of code branches in the code repository includes a branch corresponding to the first contextual scenario.

2. The CIM of claim 1 further comprising:

using an ontology hierarchy of the plurality of contextual scenarios to segment the contextual scenarios;

using the ontology hierarchy of the plurality of contextual scenarios to correlate how the contextual scenarios are related; and

creating a first hierarchical code branch in a version control tool system corresponding to the first contextual scenario.

3. The CIM of claim 1 further comprising:

predicting a second contextual scenario where the AI enabled edge device can perform a predetermined activity; and

proactively deploying self-adapted code on the AI enabled edge device.

4. The CIM of claim 1 further comprising:

performing dynamic code branching of self-adapted code upon successful execution of the first contextual scenario by the AI enabled edge device.

5. The CIM of claim 1 wherein the set of sensor device(s) includes a video camera that outputs a visual feed.

6. The CIM of claim 1 wherein the set of sensor device(s) includes an Internet of Things (IoT) sensor device.

7. A computer-implemented method (CIM) comprising:

monitoring, using a set of sensor device(s), an artificial intelligence (AI) enabled autonomous vehicle (AV) from a set of AI enabled AVs storing edge device control code in a code repository including a plurality of code branches corresponding to a plurality of contextual scenarios;

during the monitoring, operating the AV in a first contextual scenario;

analyzing, by the AV and based on the monitoring by the set of sensor device(s), performance of the AV in the first contextual scenario to identify a code branch name for the first contextual scenario;

detecting that the AV is self-adapting code for the first contextual scenario; and

in response to the detecting and based on the identified code branch name, determining whether the plurality of code branches in the code repository includes a branch corresponding to the first contextual scenario.

8. The CIM of claim 7 further comprising:

using an ontology hierarchy of the plurality of contextual scenarios to segment the contextual scenarios;

using the ontology hierarchy of the plurality of contextual scenarios to correlate how the contextual scenarios are related; and

creating a first hierarchical code branch in a version control tool system corresponding to the first contextual scenario.

9. The CIM of claim 7 further comprising:

performing dynamic code branching of self-adapted code upon successful execution of the first contextual scenario by the AV.

10. The CIM of claim 7 wherein the set of sensor device(s) includes a video camera that outputs a visual feed.

11. A computer-implemented method (CIM) comprising:

monitoring, using a set of sensor device(s), an artificial intelligence (AI) enabled industrial robotic device (IRD) from a set of AI enabled IRDs storing edge device control code in a code repository including a plurality of code branches corresponding to a plurality of contextual scenarios;

during the monitoring, operating the IRD in a first contextual scenario;

analyzing, by the IRD and based on the monitoring, performance of the IRD in the first contextual scenario to identify a code branch name for the first contextual scenario;

detecting that the IRD is self-adapting code for the first contextual scenario; and

in response to the detecting and based on the identified code branch name, determining whether the plurality of code branches in the code repository includes a branch corresponding to the first contextual scenario.

12. The CIM of claim 11 further comprising:

using an ontology hierarchy of the plurality of contextual scenarios to segment the contextual scenarios;

using the ontology hierarchy of the plurality of contextual scenarios to correlate how the contextual scenarios are related; and

creating a first hierarchical code branch in a version control tool system corresponding to the first contextual scenario.

13. The CIM of claim 11 further comprising:

performing dynamic code branching of self-adapted code upon successful execution of the first contextual scenario by the IRD.

14. The CIM of claim 11 wherein the set of sensor device(s) includes an Internet of Things (IoT) sensor device.

15. The CIM of claim 1 , further comprising updating the code repository based on the self-adapted code.

16. The CIM of claim 15 , wherein the updating comprises:

in response to determining that the plurality of code branches does not include the branch corresponding to the first contextual scenario, creating a hierarchical code branch corresponding to the first contextual scenario in the code repository; and

storing the self-adapted code in the hierarchical code branch.

17. The CIM of claim 15 , wherein the updating comprises, in response to determining that the plurality of code branches includes the branch corresponding to the first contextual scenario, applying changes to code stored in the branch based on the self-adapting.

18. The CIM of claim 15 , further comprising deploying the self-adapted code on a second device from the set of AI enabled edge devices in response to determining that the second device is operating in the first contextual scenario.

Assignments (1)

Continuity (1)

References Cited (15)