Self-adaptive device intelligence as a service enterprise infrastructure for sensor-rich environments

Patent metadata

Field	Value
Publication number	US-10754328-B2
Application number	US-201514844286-A
Country	US
Kind code	B2
Filing date	Sep 3, 2015
Priority date	Sep 5, 2014
Publication date	Aug 25, 2020
Grant date	Aug 25, 2020

Abstract

A device intelligence architecture configures and controls on-site devices and performs environment monitoring to facilitate effective device functionality. The architecture facilitates efficient use of devices (e.g., robotics) in an unstructured and dynamic environment, which will allow deployments in many more environments than is currently possible. The architecture stores and shares information between segregated devices to avoid the silo effect of vendor specific stacks. The architecture also models capabilities of devices and maps actions to intelligence packages to deploy a specific intelligence package to the device. The architecture also implements distributed processing. For instance, computationally intensive tasks may be offloaded to the back end processing, with action updates resulting from the processing pushed to the devices.

First claim

The invention claimed is:

1. A method comprising:
in a device control system in communication via a network with a plurality of devices in a deployment location:
implementing a data stream interface between the deployment location and the device control system;
receiving, through the data stream interface, a data stream comprising environmental characteristics of the deployment location;
accessing a device intelligence package library, the device intelligence package library comprising device intelligence packages, each of the device intelligence packages configured to govern operational characteristics of specific devices amongst the plurality of devices at the deployment location;
determining a selected device among the specific devices to control;
comparing the environmental characteristics to a device ontology model for the selected device and responsively identifying a selected first device intelligence package, in the device intelligence package library, with which to control the selected device;
offloading, by the device control system, a selected computation of the selected device intelligence package to an analytics system, the selected computation including a motion optimization computation, a machine learning task, or both;
receiving an analytics result for the selected computation from the analytics system;
responsive to the analytics result, controlling, by the device control system, a first aspect of the selected device by executing in the device control system a first portion of the selected first device intelligence package thereby generating control instructions data and transmitting the control instructions data to the selected device via the network;
communicating at least a second portion of the selected first device intelligence package to the selected device for execution by the selected device to control a second aspect of the selected device;
after controlling the selected device with the selected first device intelligence package, analyzing the data stream and determining a change in an operational environment and the environmental characteristics at the deployment location;
responsive to the change in the environmental characteristics, identifying a second device intelligence package in the device intelligence package library for controlling the selected device; and
executing the identified second device intelligence package to control the selected device.

2. The method of claim 1 , where:
the environmental characteristics comprise:
state information for the specific device.

3. The method of claim 1 , where:
the environmental characteristics comprise:
state information for the deployment location.

4. The method of claim 1 , where:
the selected intelligence package comprises a robot motion control package.

5. The method of claim 4 , where:
the robot motion control package comprises a series of actions for the selected device to execute to complete a robot motion task.

6. The method of claim 1 , wherein the first portion of the selected device intelligence package comprises a system level intelligence package comprising an orchestration algorithm, and wherein the second portion of the selected device intelligence package comprises an edge level intelligence package including a neural net model for visual analysis of the environment to be communicated to the selected device for execution by the selected device.

7. The method of claim 1 , further comprising:
receiving, by the control system from the selected device, information regarding an object encountered by the selected device;
building a model of the object based on the information; and
associating the model with the object for use by a second selected device that encounters the object.

8. A system comprising:
a communication interface for access to a network between a deployment location and a device control system;
a memory configured to store a data stream received through the communication interface, the data stream comprising environmental characteristics of the deployment location;
a device intelligence package library storing device intelligence packages configured to govern operational characteristics of a selected device at the deployment location; and
control circuitry configured to:
after controlling a selected device with a first device intelligence package, analyze the data stream comprising the environmental characteristics of the deployment location;
determine a change in operating environment and the environmental characteristics at the deployment location responsive to the data stream;
compare the environmental characteristics to a device ontology model for the selected device and responsively identify a second device intelligence package for controlling the selected device given the change in operating environment and environmental characteristics at the deployment location;
offload, by the device control system, a selected computation of the selected device intelligence package to an analytics system, the selected computation including a motion optimization computation, a machine learning task, or both;
receive an analytics result for the selected computation from the analytics system;
responsive to the analytics result, control a first aspect of the selected device by executing in the device control system a first portion of the second device intelligence package thereby generating control instructions data and transmitting the control instructions data to the selected device through the communication interface via a network; and
deliver at least a second portion of the second device intelligence package to the selected device for execution by the selected device to control a second aspect of the selected device.

9. The system of claim 8 , where:
the environmental characteristics comprise:
state information for the selected device.

10. The system of claim 8 , where:
the environmental characteristics comprise:
state information for the deployment location.

11. The system of claim 8 , where:
the intelligence package comprises a robot motion control package.

12. The system of claim 11 , where:
the robot motion control package comprises a series of actions for the selected device to execute to complete a robot motion task.

13. The system of claim 8 , wherein the first portion of the selected device intelligence package comprises a system level intelligence package comprising an orchestration algorithm, and wherein the second portion of the selected device intelligence package comprises an edge level intelligence package including a neural net model for visual analysis of the environment to be communicated to the selected device for execution by the selected device.

14. The system of claim 8 , where the control circuitry is further configured to:
receive, by the control system from the selected device, information regarding an object encountered by the selected device;
build a model of the object based on the information; and
associate the model with the object for use by a second selected device that encounters the object.

15. A system comprising:
a communication interface configured to establish a data connection between:
a deployment location and a device control system; and
an analytics system and the device control system;

a memory configured to store device characteristics of a selected device at the deployment location received through the communication interface via a network;
a device intelligence package library storing device intelligence packages configured to determine operational characteristics of the selected device at the deployme

Assignees

• Accenture Global Services Ltd
• Accenture Global Solutions Ltd

Inventors

• Mullan Pramila
• Mui Michael
• Chintalapally Anuraag
• Au Cindy

Classifications

• G05B19/4185 Primary

  characterised by the network communication · CPC title

• B25J9/1605

  Simulation of manipulator lay-out, design, modelling of manipulator · CPC title

• G06F8/60

  Software deployment · CPC title

• G05B2219/40518

  Motion and task planning · CPC title

• G05B2219/32129

  Select program for specified machine from library, file server · CPC title