System for locating users

Patent metadata

Field	Value
Publication number	US-10477355-B1
Application number	US-201715841216-A
Country	US
Kind code	B1
Filing date	Dec 13, 2017
Priority date	Dec 13, 2017
Publication date	Nov 12, 2019
Grant date	Nov 12, 2019

Abstract

A surface within a facility, such as a floor, is equipped with electromagnetic sensors that generate data indicative of the presence of objects, such as the feet of users, at or close to the floor. The data may also be indicative of cross-coupling resulting from a transfer of an electromagnetic signal by the object, such as from one foot to another of a user. The data is processed to determine presence of individual object representations, such as representing a foot. Information about object representations such as their movement, cross-coupling, physical distance between, or features of the object representation may be determined. This data may be used to maintain the identity of particular object representations, track object representations over time, determine that particular object representations are associated with a single user and so forth. Adjacent object representations may avoid merging by using a top-down cross sectioning technique.

First claim

What is claimed is:

1. A system comprising:
a plurality of floor devices, each of the plurality of floor devices comprising:
a flooring material;
a support structure configured to support the flooring material;
a first communication interface;
a transmitter coupled to an antenna, wherein the antenna is underneath the flooring material at a first location;
a receiver coupled to the antenna;
a first memory, storing first computer-executable instructions; and
a first hardware processor to execute the first computer-executable instructions to:
use the transmitter to transmit a transmitted signal that is radiated by the antenna, wherein the transmitted signal is indicative of the first location;
use the receiver to receive one or more received signals with the antenna;
generate first sensor data indicative of a characteristic of the one or more received signals at a first time;
send the first sensor data using the first communication interface;
generate second sensor data indicative of a characteristic of the one or more received signals at a second time; and
send the second sensor data using the first communication interface; and

a server comprising:
a second communication interface;
a second memory, storing second computer-executable instructions; and
a second hardware processor to execute the second computer-executable instructions to:
receive the first sensor data and the second sensor data from the plurality of floor devices;
determine, from the first sensor data, a first object representation that is representative of a first object that is proximate to the flooring material, wherein the first object representation comprises sensor data having values that exceed a threshold value;
determine, from the second sensor data, a second object representation that is representative of a second object that is proximate to the flooring material, wherein the second object representation comprises sensor data having values that exceed the threshold value;
determine a first user identifier associated with the first object representation;
determine that at least a portion of a first perimeter of the first object representation overlaps a portion of a second perimeter of the second object representation; and
associate the first user identifier with the second object representation.

2. The system of claim 1 , wherein the first sensor data is representative of a received signal strength at the first time and the second sensor data is representative of a received signal strength at the second time; and
the threshold value is indicative of a minimum signal strength representative of detection of an object.

3. The system of claim 1 , further comprising:
determine, from the second sensor data, a third object representation that is representative of a third object that is proximate to the flooring material at the second time, wherein the third object representation comprises sensor data values that exceed the threshold value;
determine, based on the second sensor data, that a first electromagnetic signal has been transferred from the antenna at the first location that is within the first perimeter of the first object representation, to a second antenna at a second location that is within a third perimeter of the third object representation; and
determine the first object and the third object are associated with one another.

4. A method comprising:
accessing first sensor data at a first time from one or more sensors at a floor;
accessing second sensor data at a second time from the one or more sensors;
determining, from the first sensor data, a first object representation that is representative of a first object that is above the floor, wherein the first object representation comprises sensor data values that exceed a threshold value;
determining, from the second sensor data, a second object representation that is representative of a second object that is above the floor, wherein the second object representation comprises sensor data values that exceed the threshold value;
determining that the first object and the second object are a same object; and
determining one or more user identifiers associated with the first object representation and the second object representation based on feature data of the first object representation.

5. The method of claim 4 , further comprising:
generating user track data indicative of movement with respect to the floor of a user associated with the one or more user identifiers.

6. The method of claim 4 , wherein the determining that the first object and the second object are the same object further comprises:
determining a first location that is within a first perimeter of the first object representation; and
determining the first location is within a second perimeter of the second object representation.

7. The method of claim 4 , wherein the determining that the first object and the second object are the same object further comprises:
determining a set of prior locations of the first object at times before the second time;
determining a predicted location for the first object at the second time;
determining a region of location uncertainty associated with the predicted location; and
determining that at least a portion of the second object representation overlaps the region of location uncertainty.

8. The method of claim 7 , wherein the determining the region of location uncertainty is based on one or more of:
maximum expected displacement of a foot during successive times;
statistical estimation error;
direction of travel indicated by the set of prior locations of the first object; or
speed of the first object during at least a portion of the successive times.

9. The method of claim 4 , further comprising:
accessing third sensor data at a third time from the one or more sensors;
determining, from the third sensor data, a third object representation that is representative of a third object that is above the floor, wherein the third object representation comprises sensor data values that exceed the threshold value;
determining a first electromagnetic signal has been transferred, from a first sensor of the one or more sensors that is at a first location within a first perimeter of the first object representation, to a second sensor of the one or more sensors that is at a second location within a second perimeter of the third object representation; and
determining the first object and the third object are associated with one another.

10. The method of claim 4 , further comprising:
accessing third sensor data at a third time from the one or more sensors;
determining, from the third sensor data, a third object representation that is representative of a third object that is above the floor, wherein the third object representation comprises sensor data values that exceed the threshold value;
determining the first object is a first foot of a user and the third object is a second foot of the user;
determining, using the first sensor data and the third sensor data, gait data indicative of one or more of:
timing between successive steps of one or more of the first foot and the second foot;
a distance between the first foot and the second foot when both are in contact with the floor; and
a duration of simultaneous contact of the first foot and the second foot with the floor;

accessing historical gait data that associates particular gait data with particular user identifiers;
determining a correspondence between the gait data and the particular gait data of a particular user identifier exceeds a second threshold value; and
associating the particular user identifier with the first object and the second object.

Assignees

• Amazon Tech Inc

Inventors

• Niranjayan Somasundaram
• Mccleary Patrick Ian
• Bowen Matthew Gregory
• Orlov Nikolai
• Stephanick James

Classifications

• A61B5/1128

  using image analysis (A61B5/1127 takes precedence) · CPC title

• A61B5/112

  Gait analysis · CPC title

• A61B5/1113

  Local tracking of patients, e.g. in a hospital or private home · CPC title

• H04W4/38

  for collecting sensor information · CPC title

• H04W4/33

  for indoor environments, e.g. buildings · CPC title