Artificial intelligence server for determining route of robot and method for the same

Patent metadata

Field	Value
Publication number	US-11511422-B2
Application number	US-201916556150-A
Country	US
Kind code	B2
Filing date	Aug 29, 2019
Priority date	Jul 30, 2019
Publication date	Nov 29, 2022
Grant date	Nov 29, 2022

Abstract

An artificial intelligence server for determining a route of a robot includes a communication unit and a processor. The communication unit is configured to receive image data for a control area from the robot or a camera installed inside the control area. The processor is configured to calculate a current density for the control area from the image data, calculate a future density for the control area using the calculated current density, determine a priority for each of group areas included in the control area based on the calculated future density, and determine the route of the robot based on the determined priority.

First claim

What is claimed is:

1. An artificial intelligence server for determining a route of a robot, comprising:
a communication unit configured to receive image data for a control area from the robot or by a camera configured to capture images inside the control area, wherein the control area is divided into group areas that are further divided into unit areas; and
a processor configured to:
determine a current density for the control area from the received image data;
determine a future density for the control area using the determined current density and a crowd inflow for each of the unit areas, wherein the crowd inflow is determined by determining an average amount of change in density of each unit area within a predetermined period;
based on a determination that future densities for the group areas are a same density, determine a sub-density average value for a lower group area of each of the group areas and determine a priority order for each of the group areas based on a comparison of the determined sub-density average values for the lower group areas, wherein the lower group area corresponds to groups of areas having a size smaller by one unit area;
based on a determination that future densities for the group areas are different densities, determine the priority order for each of the group areas included in the control area based on the determined future density; and
determine the route of the robot based on the determined priority order, wherein the robot travels within the control area according to the determined route.

2. The artificial intelligence server according to claim 1 , wherein the control area corresponds to a maximum activity range of the robot.

3. The artificial intelligence server according to claim 2 , wherein the current density for the control area comprises a current density for each of the group areas included in the control area, wherein the future density for the control area comprises a future density for each of the group areas.

4. The artificial intelligence server according to claim 3 , wherein the future density is further determined using least one of a crowd movement direction for each of the unit areas, spatial information for the control area, schedule information for the control area, a crowd moving pattern, or a facility usage pattern.

5. The artificial intelligence server according to claim 4 , wherein the future density is further determined using a future density calculation model trained by using a machine learning algorithm or a deep learning algorithm and is configured as an artificial neural network.

6. The artificial intelligence server according to claim 4 , wherein the crowd movement direction for a first unit area is determined by summing movement direction vectors generated by determining directions of faces of users recognized in the first unit area from the received image data.

7. The artificial intelligence server according to claim 6 , wherein the processor is further configured to determine magnitudes of the movement direction vectors in proportion to movement speeds of the recognized faces of the users.

8. The artificial intelligence server according to claim 3 , wherein the priority order of corresponding group areas are determined to be higher based on the future density increasing.

9. The artificial intelligence server according to claim 8 , wherein the route of the robot is determined such that the robot moves to the group areas in a descending order of the priority order.

10. The artificial intelligence server according to claim 3 , wherein the current density for the control area is determined based on the current density determined for each of the unit areas using the received image data.

11. The artificial intelligence server according to claim 10 , wherein the processor is further configured to:
recognize faces of users included in a second unit area from the received image data using a face recognition model, wherein the current density for the second unit area is determined based on a number of the recognized face.

12. The artificial intelligence server according to claim 11 , wherein the face recognition model is trained using a machine learning algorithm or a deep learning algorithm and is configured as an artificial neural network.

13. A method for determining a route of a robot, comprising:
receiving image data for a control area from the robot or by a camera configured to capture images inside the control area, wherein the control area is divided into group areas that are further divided into unit areas;
determining a current density for the control area from the received image data;
determining a future density for the control area using the determined current density and a crowd inflow, wherein the crowd inflow is determined by determining an average amount of change in density of each unit area within a predetermined period;
based on a determination that future densities for the group areas are a same density, determine a sub-density average value for a lower group area of each of the group areas and determine a priority order for each of the group areas based on a comparison of the determined sub-density average values for the lower group areas, wherein the lower group area corresponds to groups of areas having a size smaller by one unit area;
based on a determination that future densities for the group areas are different densities, determining the priority order for each of the group areas included in the control area based on the determined future density; and
determining the route of the robot based on the determined priority order, wherein the robot travels within the control area according to the determined route.

14. A non-transitory recording medium having recorded thereon a program for performing a method for determining a route of a robot, the method comprising:
receiving image data for a control area from the robot or by a camera configured to capture images inside the control area, wherein the control area is divided into group areas that are further divided into unit areas; and
determining a current density for the control area from the received image data;
determining a future density for the control area using the determined current density and a crowd inflow for each of the unit areas, wherein the crowd inflow is determined by determining an average amount of change in density of each unit area within a predetermined period;
based on a determination that future densities for the group areas are a same density, determine a sub-density average value for a lower group area of each of the group areas and determine a priority order for each of the group areas based on a comparison of the determined sub-density average values for the lower group areas, wherein the lower group area corresponds to groups of areas having a size smaller by one unit area;
based on a determination that future densities for the group areas are different densities, determining the priority order for each of the group areas included in the control area based on the determined future density; and
determining the route of the robot based on the determined priority order, wherein the robot travels within the control area according to the determined route.

Assignees

• Lg Electronics Inc

Inventors

• Chae Jonghoon
• Maeng Jichan

Classifications

• B25J19/023

  including video camera means · CPC title

• B25J9/163

  learning, adaptive, model based, rule based expert control · CPC title

• G06V20/53

  Recognition of crowd images, e.g. recognition of crowd congestion · CPC title

• B25J9/1664 Primary

  characterised by motion, path, trajectory planning · CPC title

• G06Q50/10

  Services · CPC title