Accurate detection of track occupancy within train tracking systems

Patent metadata

Field	Value
Publication number	US-12330701-B2
Application number	US-202318475663-A
Country	US
Kind code	B2
Filing date	Sep 27, 2023
Priority date	Sep 27, 2023
Publication date	Jun 17, 2025
Grant date	Jun 17, 2025

Abstract

A train tracking system includes a train tracking module, a plurality of input sources providing train related information, and one or more interface(s) associated with the train tracking module, wherein the train tracking module is configured via computer executable instructions and through operation of a processor to receive the train related information from the plurality of input sources via the one or more interface(s), determine a position of a train within a train track map, the position of the train corresponding to a track location of the train in a track network, and wherein the train tracking module further comprises a learning algorithm, the train tracking module being configured to, utilizing the learning algorithm, determine an unexpected occupancy of the track location.

First claim

The invention claimed is:

1. A train tracking system comprising:
train tracking module,
a plurality of input sources providing train related information, and
one or more interface(s) associated with the train tracking module,
wherein the train tracking module is configured via computer executable instructions and through operation of at least one processor to
receive the train related information from the plurality of input sources via the one or more interface(s),
determine a position of a train within a train track map, the position of the train corresponding to a track location of the train in a track network, and

wherein the train tracking module further comprises a learning algorithm, the train tracking module being configured to, utilizing the learning algorithm,
determine an unexpected occupancy of the track location,

wherein the train tracking module is configured to determine a location attribute for the position of the train and for the unexpected occupancy of the track location, and an accuracy estimate of the position of the train in the track location,
wherein the location attribute for the position of the train includes a trusted position and a suspect position,
wherein the known positions of trains within the track network include trusted positions of track locations and
wherein the configuration table comprises multiple inputs, the multiple inputs relating to the plurality of input sources, wherein each input comprises an assigned level of trust, and
wherein the trusted position and the suspect position are based on different combinations of the levels of trust of the multiple inputs.

2. The train tracking system of claim 1 ,
wherein the train tracking module is further configured to, utilizing the learning algorithm, generate a warning for the unexpected occupancy of the track location.

3. The train tracking system of claim 2 ,
wherein the unexpected occupancy of the track location is determined based on a disagreement of the position of the train and a planned route of the train, or a disagreement between the train's identifier and an expected identifier for the track location.

4. The train tracking system of claim 1 ,
wherein training/input data for the learning algorithm comprise known positions of trains within the track network, global positioning system (GPS) history for the track location, track layout information of the track location, history of reports including confirmations or rejections of trains at the track location, and/or dispatcher feedback.

5. The train tracking system of claim 4 , further comprising:
a dispatch system,
wherein the train tracking module is operably coupled to the dispatch system, or wherein the train tracking module is implemented by the dispatch system, and
wherein the confirmations or rejections of trains at the track location and dispatcher feedback are provided by a dispatcher via the dispatch system.

6. The train tracking system of claim 1 , further comprising:
a configuration table configured to be accessed by the train tracking module,
wherein the location attribute including the trusted position and the suspect position is determined utilizing the configuration table.

7. The train tracking system of claim 1 ,
wherein the plurality of input sources comprises global positioning system (GPS) messages, positive train control (PTC) messages, track circuit indications, form-based authorities, train route plans, end of train (EOT) messages, head of train (HOT) messages, verbal crew input, work authorities, parallel tracks, and/or locations of other trains.

8. A method for train tracking, the method comprising through operation of at least one processor:
collecting train related information from a plurality of input sources via one or more interface(s),
determining a position of a train within a train track map based on the train related information, the position of the train corresponding to a track location of the train in a track network,
determining an unexpected occupancy of the track location utilizing a learning algorithm,
determining, utilizing the learning algorithm, a location attribute for the position of the train and the unexpected occupancy of the track location, and an accuracy estimate of the position of the train in the track location, and
accessing a configuration table,
wherein the location attribute including a trusted position and a suspect position is determined utilizing the configuration table,
wherein the configuration table comprises multiple inputs, the multiple inputs relating to the plurality of input sources, wherein each input comprises an assigned level of trust, and
wherein the trusted position and the suspect position are based on different combinations of the levels of trust of the multiple inputs.

9. The method of claim 8 , further comprising:
generating a warning for the unexpected occupancy of the track location utilizing the learning algorithm.

10. The method of claim 8 , wherein determining the unexpected occupancy of the track location is based on a disagreement of the position of the train and a planned route of the train, or a disagreement between the train's identifier and an expected identifier for the track location.

11. The method of claim 8 , further comprising:
receiving, by the learning algorithm, training and input data, the training and input data comprising known positions of trains within the track network, global positioning system (GPS) history for the track location, track layout information of the track location, history of reports of trains at the track location, and/or dispatcher feedback.

12. The method of claim 11 ,
wherein the history of reports of trains at the track location includes confirmations or rejections of trains at the track location, and
wherein the confirmations or rejections of trains at the track location are provided by a dispatcher via the dispatch system.

13. The method of claim 8 ,
wherein the plurality of input sources comprises global positioning system (GPS) messages, positive train control (PTC) messages, track circuit indications, form-based authorities, train route plans, end of train (EOT) messages, head of train (HOT) messages, verbal crew input, work authorities, parallel tracks, and/or locations of other trains.

14. A non-transitory computer readable medium encoded with processor executable instructions that when executed by at least one processor, cause the at least one processor to carry out a method for train tracking as claimed in claim 8 .

Assignees

• Siemens Mobility Inc

Inventors

• Mckenna Monica

Classifications

• B61L27/10

  Operations, e.g. scheduling or time tables · CPC title

• B61L21/06

  Vehicle-on-line indication; Monitoring locking and release of the route · CPC title

• B61L27/30

  Trackside multiple control systems, e.g. switch-over between different systems · CPC title

• B61L27/40

  Handling position reports or trackside vehicle data · CPC title

• B60L2240/625

  by GSM · CPC title