Detecting divergence or convergence of related objects in motion and applying asymmetric rules

Patent metadata

Field	Value
Publication number	US-2016292885-A1
Application number	US-201514673476-A
Country	US
Kind code	A1
Filing date	Mar 30, 2015
Priority date	Mar 30, 2015
Publication date	Oct 6, 2016
Grant date	—

Abstract

Systems and methods may include receiving first and second motion data about first and second objects, respectively. The systems and methods may include applying an asymmetric correlation rule (ACR) to the first and second motion data. The ACR may define a baseline state of motion for the first object and a correlation relationship between the first and second objects. The systems and methods may include determining whether the first object is in the baseline state and, in response to determining that the first object is not in the baseline state, determining whether a first motion vector of the first object is sufficiently correlated with a second motion vector of the second object. The systems and methods may include, in response to determining both that the first object is not in the baseline state and that the first and second motion vectors are not sufficiently correlated, generating an anomalous behavior notification.

First claim

What is claimed is:

1 . A method comprising:
receiving first motion data corresponding to a first object; receiving second motion data corresponding to a second object; and applying an asymmetric correlation rule to the first motion data and the second motion data, the asymmetric correlation rule defining a baseline state of motion for the first object and a correlation relationship between the first object and the second object, and applying the asymmetric motion rule comprising:
determining, based on the first motion data, whether the first object is in the baseline state of motion;
in response to determining that the first object is not in the baseline state of motion, determining, based on the first motion data and the second motion data, whether a first motion vector of the first object is correlated at a threshold level of correlation with a second motion vector of the second object, the correlation relationship between the first object and the second object identifying the threshold level of correlation between the first motion vector and the second motion vector; and
in response to determining both that the first object is not in the baseline state of motion and that the first motion vector of the first object is not correlated with the second motion vector of the second object at the threshold level of correlation, generating a notification that the first object is behaving anomalously.

2 . The method of claim 1 ,
wherein the asymmetric correlation rule further defines a path for the second object, and wherein applying the asymmetric motion rule further comprises:
determining, based on the second motion data, whether the second object is located on the path;
in response to determining that the second object is not located on the path, generating a notification that the second object is behaving anomalously.

3 . The method of claim 2 ,
wherein the asymmetric correlation rule further defines a time-bound position along the path for the second object, the time-bound position being an expected position of the second object along the path at a particular point in time, and wherein applying the asymmetric motion rule further comprises:
determining, based on the first motion data, whether a position of the second object at the particular point in time is convergent with the time-bound position for the second object at the particular point in time; and
in response to determining that the position of the second object at the particular point in time is not convergent with the time-bound position for the second object at the particular point in time, generating the notification that the second object is behaving anomalously.

4 . The method of claim 1 , further comprising:
after determining that the first motion vector and the second motion vector are correlated at the threshold level of correlation:
receiving third motion data about the first object;
receiving fourth motion data about the second object;
determining, based on the third motion data and the fourth motion data, whether the first motion vector remains correlated at the threshold level of correlation with the second motion vector; and
in response to determining that the first motion vector is no longer correlated at the threshold level of correlation with the second motion vector, generating a notification indicating that the first object and the second object have diverged.

5 . The method of claim 1 , further comprising:
determining, based on the first motion data and the second motion data, that the first object and the second object are currently at different locations but are likely to converge at a particular location and a particular time; and in response to determining that the first object and the second object are currently at different locations but are likely to converge at the particular location and the particular time, performing an action at the particular location to prepare for the first object and the second object to arrive.

6 . The method of claim 1 , wherein the correlation relationship between the first object and the second object specifies that the first motion vector and the second motion vector are correlated at the threshold level of correlation when the first motion vector and the second motion vector indicate that the first object and the second object are likely to converge at a particular location.

7 . The method of claim 1 , wherein the asymmetric correlation rule defines the baseline state of motion for the first object to include a particular motion vector and one or more deviations from the particular motion vector that occur when the first object and the second object are within a particular distance from one another.

8 . The method of claim 1 , further comprising
defining one of the first object and the second object as an anchor object; providing on a map a representation of an anchor motion vector, the first motion vector being the anchor motion vector if the first object is defined as the anchor object, and the second motion vector being the anchor motion vector if the second object is defined as the anchor object; and in response to determining that the first motion vector of the first object is not correlated with the second motion vector of the second object at the threshold level of correlation, providing on the map with the representation of the anchor motion vector a representation of the one of the first motion vector and the second motion vector that is not the anchor motion vector, such that the representation of the one of the first motion vector and the second motion vector that is not the anchor motion vector is only provided on the map with the representation of the anchor motion vector when the first motion vector of the first object is not correlated with the second motion vector of the second object at the threshold level of correlation.

9 . The method of claim 1 , further comprising:
defining one of the first object and the second object as an anchor object; providing on a map a representation of an anchor motion vector, the first motion vector being the anchor motion vector if the first object is defined as the anchor object, and the second motion vector being the anchor motion vector if the second object is defined as the anchor object; and determining whether the first object and the second object are positioned within a threshold distance of one another; and in response to determining that the first object and the second object are not positioned within the threshold distance of one another, providing on the map with the representation of the anchor motion vector a representation of the one of the first object and the second object that is not defined as the anchor object, such that the representation of the one of the first object and the second object that is not the anchor motion vector is only provided on the map with the representation of the anchor motion vector when the first object and the second object are not positioned within the threshold distance of one another.

10 . The method of claim 1 , further comprising:
defining one of the first object and the second object as an anchor object; providing on a map a representation of an anchor motion vector, the first motion vector being the anchor motion vector if the first object is defined as the anchor object, and the second motion vector being the anchor motion vector if the second object is defined as the anchor object; and in response to determining that the first motion vector of the first object is correlated with the second motion vector of the second object at the threshold level of correlation, providing on a map a representation of an anchor motion vector as a representation

Assignees

• Greenspan Steven L
• Lipton Paul C
• Harrison Jr John J
• Parikh Prashant
• Mankovskii Serguei
• Velez-Rojas Maria C
• Ca Inc

Inventors

• Greenspan Steven L
• Lipton Paul C
• Harrison Jr John J
• Parikh Prashant
• Mankovskii Serguei
• Velez-Rojas Maria C

Classifications

• G06Q10/08 Primary

  Logistics, e.g. warehousing, loading or distribution; Inventory or stock management · CPC title

• G06F2218/00

  Aspects of pattern recognition specially adapted for signal processing · CPC title

• G06T7/68

  of symmetry · CPC title

• G06T7/2093 Primary

  Physics · mapped topic

• G06T7/608

  Physics · mapped topic