Method and system of error modelling for object localization using ultra wide band (uwb) sensors

Patent metadata

Field	Value
Publication number	US-2023115981-A1
Application number	US-202217956006-A
Country	US
Kind code	A1
Filing date	Sep 29, 2022
Priority date	Oct 12, 2021
Publication date	Apr 13, 2023
Grant date	—

Abstract

Ultra Wide Band (UWB) based Real Time Location Systems (RTLS) that are being used for location tracking suffer from due to environment specific errors that are introduced due to factors such as difference in reflection and propagation. The disclosure herein generally relates to object localization, and, more particularly, to a method and system of error modelling for object localization using Ultra Wide Band (UWB) sensors. The error modelling allows the system to correct a determined location of an object being tracked, to determine a corrected location.

First claim

What is claimed is:

1 . A processor implemented method for location tracking, comprising:
obtaining distance value of a tag node with reference to position of a plurality of anchor nodes, by an edge node implemented via one or more hardware processors, wherein the tag node is associated with an object being tracked; determining location (X, Y) of the object in a 2-Dimensional space, based on the obtained distance value of the tag node with reference to the position of the plurality of anchor nodes, by the edge node; and correcting the determined location (X, Y) of the object to obtain a corrected location (X′, Y′), by the edge node, comprising:
calculating a moving average for the determined location (X, Y) over a one second time span; and
applying a multivariate model over the calculated moving average and a corresponding ground truth location, wherein applying a multivariate model comprises:
computing an error function by calculating a Euclidian distance between the determined location and the ground truth location;
determining an argument of the minimum for the error function by applying an argmin function over the computed error function; and
applying values of a bias, a scale factor, and an interaxial factor on the calculated moving average, wherein the values of the bias, the scale factor, and the interaxial factor are computed based on the determined argument of the minimum of the error function.

2 . The method of claim 1 , wherein the multivariate model comprises computed values of the bias, the scale factor, and the interaxial factor for a plurality of pairs of determined location and corresponding ground truth location on a training data set.

3 . A system for location tracking, comprising:
one or more hardware processors; a communication interface; and a memory storing a plurality of instructions, wherein the plurality instructions when executed, cause the one or more hardware processors to:
obtain distance value of a tag node with reference to position of a plurality of anchor nodes, by an edge node implemented via the one or more hardware processors, wherein the tag node is associated with an object being tracked;
determine location (X, Y) of the object in a 2-Dimensional space, based on the obtained distance value of the tag node with reference to the position of the plurality of anchor nodes, by the edge node; and
correct the determined location (X, Y) of the object to obtain a corrected location (X′, Y′), by the edge node, by:
calculating a moving average for the determined location (X, Y) over a one second time span; and
applying a multivariate model over the calculated moving average and a corresponding ground truth location, wherein applying a multivariate model comprises:
computing an error function by calculating a Euclidian distance between the determined location and the ground truth location;
determining an argument of the minimum for the error function by applying an argmin function over the computed error function; and
applying values of a bias, a scale factor, and an interaxial factor on the calculated moving average, wherein the values of the bias, the scale factor, and the interaxial factor are computed based on the determined argument of the minimum of the error function.

4 . The system as claimed in claim 3 , wherein the multivariate model comprises computed values of the bias, the scale factor, and the interaxial factor for a plurality of pairs of determined location and corresponding ground truth location on a training data set.

5 . One or more non-transitory machine-readable information storage mediums comprising one or more instructions which when executed by one or more hardware processors cause:
obtaining distance value of a tag node with reference to position of a plurality of anchor nodes, by an edge node implemented wherein the tag node is associated with an object being tracked; determining location (X, Y) of the object in a 2-Dimensional space, based on the obtained distance value of the tag node with reference to the position of the plurality of anchor nodes, by the edge node; and correcting the determined location (X, Y) of the object to obtain a corrected location (X′, Y′), by the edge node, comprising:
calculating a moving average for the determined location (X, Y) over a one second time span; and
applying a multivariate model over the calculated moving average and a corresponding ground truth location, wherein applying a multivariate model comprises:
computing an error function by calculating a Euclidian distance between the determined location and the ground truth location;
determining an argument of the minimum for the error function by applying an argmin function over the computed error function; and
applying values of a bias, a scale factor, and an interaxial factor on the calculated moving average, wherein the values of the bias, the scale factor, and the interaxial factor are computed based on the determined argument of the minimum of the error function.

6 . The one or more non-transitory machine-readable information storage mediums of claim 5 , wherein the multivariate model comprises computed values of the bias, the scale factor, and the interaxial factor for a plurality of pairs of determined location and corresponding ground truth location on a training data set.

Assignees

• Tata Consultancy Services Ltd

Inventors

• Ghose Avik
• Bhaumik Chirabrata
• Chakravarty Tapas
• Swain Amit

Classifications

• G01S5/0278 Primary

  involving statistical or probabilistic considerations (G01S5/0252, G01S5/0294 take precedence) · CPC title

• G01S5/0284 Primary

  Relative positioning · CPC title

• H04W64/003

  locating network equipment · CPC title

• G01S5/14

  Determining absolute distances from a plurality of spaced points of known location · CPC title