IP Library › Granted Patent US 12,330,696
Granted Patent B2
US 12,330,696 · App. 18/582,223 · Granted Jun 17, 2025

On-board thermal track misalignment detection system and method therefor

Inventors: Asim Ghanchi (Southlake, TX); Nicholas Dryer (North Barrington, IL); Coleman Barkley (Azle, TX); Michael Saied Saniei (Fort Worth, TX)
Assignee: BNSF Railway Company
B61K9/08B61L23/042B61L23/047B61L25/025G06T7/0002H04N23/90B61L2205/04H04N23/54
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,330,696
App. No.
18/582,223
Granted
Jun 17, 2025
Kind
B2
Abstract

An on-board thermal track misalignment detection system method therefor is presented. The system can use on-board locomotive sensors attached to an end-of-train device to detect (on the edge), signs and symptoms of thermal misalignments of the track. Once detected an alert can be transmitted to prevent potential derailments. The system can also include a forward-facing and rearward-facing imaging sensors (e.g., camera, LiDAR sensor, etc). The system can wirelessly communicate (e.g., via radio) with a leading locomotive to ensure proper air pressure and location. The system can be powered by an on-board battery and/or air pressure device. Advantageously, the system can calculate whether any rail deviation is significant (e.g., via one or more threshold values). The system can also leverage image processing functionality, executed by one or more processors) to find the centerline and the distance between the tracks.

Claims (35)

1. A method of detecting railroad track misalignment, comprising:

receiving, via a first imaging sensor coupled to a rail vehicle, a first rail view of a first section of a first rail;

generating first imaging data related to the first rail via the first imaging sensor;

receiving, via a second imaging sensor coupled to a rail vehicle, a second rail view of a second section of the first rail;

generating second imaging data related to the first rail via the second imaging sensor; and

comparing, via an edge processor, the first and second imaging data to detect misalignments of a railroad track, wherein an adaptive anomaly threshold varies based upon a measured deviation of the rail.

2. The method of claim 1 , wherein the measured deviation is calculated by the edge processor using a distance between the first rail and a second rail.

3. The method of claim 1 , wherein the measured deviation is calculated by the edge processor using a distance between the first rail and a centerline between rails.

4. The method of claim 1 , further comprising a GPS device operably coupled to the edge processor and configured to receive one or more satellite signals to determine the position of the rail vehicle.

5. The method of claim 4 , wherein the edge processor can determine a speed index for a particular section of track using the rail vehicle position.

6. The method of claim 5 , wherein an adaptive anomaly threshold varies based upon the speed index of the rail vehicle location.

7. The method of claim 1 , wherein the first imaging sensor and the second imaging sensor can be positioned within a first sensor module to selectively determine the first rail view and the second rail view.

8. The method of claim 7 , wherein the first imaging sensor is positioned parallel to the first rail.

9. The method of claim 7 , wherein the second imaging sensor is positioned canted downward from parallel to the first rail.

10. The method of claim 7 , further comprising:

receiving, via a third imaging sensor coupled to the rail vehicle, a third rail view of a first section of a second rail;

generating third imaging data related to the second rail via the third imaging sensor;

receiving, via a fourth imaging sensor coupled to the rail vehicle, a fourth rail view of a second section of the second rail; and

generating fourth imaging data related to the second rail via the fourth imaging sensor.

11. The method of claim 10 , wherein the edge processor is disposed proximate the first, second, third, and fourth imaging sensors and configured to receive and compare the third and fourth imaging data to determine whether any differences between the third and fourth imaging data exceed an adaptive anomaly threshold.

12. The method of claim 10 , wherein the third imaging sensor and the fourth imaging sensor can be positioned within a second sensor module to selectively determine the third rail view and the fourth rail view.

13. The method of claim 10 , wherein the first imaging sensor is positioned parallel to the first rail.

14. The method of claim 10 , wherein the second imaging sensor is positioned canted downward from parallel to the first rail.

15. The method of claim 1 , further comprising generating an alert identifying a section of misaligned railroad track and transmitting the alert to one or more recipients.

16. A method of detecting railroad track misalignments via a railroad track misalignment system, comprising:

receiving, via a first imaging sensor coupled to a rail vehicle, a first rail view of a first section of a first rail;

generating first imaging data related to the first rail via the first imaging sensor;

receiving, via a second imaging sensor coupled to a rail vehicle, a second rail view of a second section of the first rail;

generating second imaging data related to the first rail via the second imaging sensor; and

generating inertial sensor data, via an inertial sensor, related to a railcar's vibration or movement;

removing, via an edge processor, the railcar's vibration or movement from the first, second, third, and fourth imaging data using the inertial sensor data; and

comparing, via the edge processor, the first and second or the third and fourth imaging data to detect railroad track misalignments.

17. The method of claim 16 , further comprising a solar panel or battery operably coupled to the system and configured to provide power to one or more components.

18. The method of claim 16 , wherein the edge processor compares two or more data sets generated by the first and second imaging sensors in real-time.

19. The method of claim 16 , further comprising generating an alert identifying a section of misaligned railroad track and transmitting the alert to one or more recipients.

Assignments (1)
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Feb 21, 2024
From: GHANCHI, ASIM; DRYER, NICHOLAS; BARKLEY, COLEMAN; SANIEI, MICHAEL SAIED
To: BNSF RAILWAY COMPANY
Reel/Frame 066511/0875 →
Continuity (3)
Continuation 18185923 · Mar 17, 2023
Continuation 17806316 · Jun 10, 2022
Related Publication 20240190483A1 · Jun 13, 2024
References Cited (19)
US 9340219B2 · Gamache et al. · 2016 [cited by applicant]
US 9796400B2 · Puttagunta et al. · 2017 [cited by applicant]
US 9981673B2 · Martin et al. · 2018 [cited by applicant]
US 10713503B2 · Thakkar et al. · 2020 [cited by applicant]
US 11022982B2 · Naithani et al. · 2021 [cited by applicant]
US 20150268172A1 · Naithani et al. · 2015 [cited by applicant]
US 20170227470A1 · Cesarano · 2017 [cited by applicant]
US 20180339720A1 · Singh · 2018 [cited by examiner]
US 20190054937A1 · Graetz et al. · 2019 [cited by applicant]
US 20200143173A1 · Kurz et al. · 2020 [cited by applicant]
US 20200207389A1 · Moth et al. · 2020 [cited by applicant]
US 20200158656A1 · Chung et al. · 2020 [cited by applicant]
US 20210078622A1 · Miller et al. · 2021 [cited by applicant]
US 20230112991A1 · Wang et al. · 2023 [cited by applicant]
Lu et al., Rolling-stock Automatic In-situ Line Deterioration and Operating Condition Sensing, Jan. 17, 2013. [cited by applicant]
Hong, Train Track Misalignment Detection System, May 31, 2014. [cited by applicant]
Ravishankar et al.; Detecting Defects of Railway Tracks by Using Computer Vision Methodology; Proceedings of the 2021 IISE Annual Conference; 2021; pp. 572-577 (Year: 2021). [cited by applicant]
Mittal et al.; Vision Based Railway Track Monitoring Using Deep learning; arXiv preprint arXic;1711.06423 (2017). [cited by applicant]
International Patent Application No. PCT/US2023/016843 International Search Report and Written Opinion, Jul. 12, 2023, 15 Pages. [cited by applicant]