Vehicle determination system and method using a kalman filter and critical milepost data

• US 9,606,240 B2
• Filed: 11/27/2007
• Issued: 03/28/2017
• Est. Priority Date: 11/27/2007
• Status: Active Grant

• Current Assignee: Westinghouse Air Brake Technologies Corporation
• Original Assignee: General Electric Company
• Inventors: Bonanni, Pierino Gianni, Chan, David So Keung, Mathews, Harry Kirk, Vachhani, Pramod
• Primary Examiner(s): Refai, Ramsey
• Application Number:
  US11/998,103
• Publication Number:
  US 20090138199A1
• Time in Patent Office: 3,409 Days
• Field of Search: 701/480, 701/479
• US Class Current: 1/1
• CPC Class Codes: G01C 21/28 with correlation of data from several navigational instruments
• CPC Class Codes: G01S 19/47 the supplementary measurement being an inertial measurement, e.g. tightly coupled inertial
• CPC Class Codes: G01S 19/50 whereby the position solution is constrained to lie upon a particular curve or surface, e.g. for locomotives on railway tracks

Abstract

A vehicle location determination system and method provide an estimate of real time location of the vehicle along a route in response solely to vehicle GPS information and vehicle speed information such that the estimated real time distance is robust to errors and disturbances associated with both the vehicle GPS information and vehicle speed information to ensure the estimated real time location information is accurate.

Claims

1. A vehicle location determination system (LDS) comprising:

a computer or processor;

a speed bias based speed integration module;

a Kalman filter based estimator, wherein the computer or processor is programmed to estimate real time location of the vehicle along a route via the Kalman filter based estimator that estimates the real time location in response solely to both GPS information and vehicle speed information; and

a critical milepost module comprising a database of identifying entries that are distinct and independent from the GPS information and the vehicle speed information, wherein the computer or processor is further programmed to reset the speed bias based integration module based upon the identifying entries in the database such that the estimated real time location is robust to cumulative errors and disturbances associated with both the GPS information and vehicle speed information to ensure the estimated real time location information is accurate.

2. The vehicle LDS of claim 1, wherein the computer or processor is further programmed to reject GPS data or vehicle speed data that exceeds respective threshold limits.

3. The vehicle LDS of claim 1, wherein the computer or processor is further programmed to intermittently reset and provide a corrected reference value of speed bias based on the identifying entries in the critical milepost database.

4. The vehicle LDS of claim 3, wherein the computer or processor is further programmed to process the most recent GPS information via projection of GPS values onto the critical milepost database.

5. The vehicle LDS of claim 4, wherein the database comprises known latitude data, longitude data, and cumulative distance data associated with the route.

6. The vehicle LDS of claim 4, wherein the projection is a great arc projection.

7. The vehicle LDS of claim 6, wherein the computer or processor is further programmed to generate a quality indicator signal associated with the estimated real time location information in response to GPS projection data and associated covariance data.

8. The vehicle LDS of claim 4, wherein the projection is a non-linear least squares projection.

9. The vehicle LDS of claim 1, wherein the Kalman filter based estimator comprises a two-state variable Kalman filter, wherein first state variable comprises vehicle position along the route and the second state variable comprises vehicle speed bias based upon the critical milepost database.

10. The vehicle LDS of claim 1, wherein the Kalman filter based estimator comprises a single-state variable Kalman filter, wherein the single-state variable comprises vehicle position along the route.

11. The vehicle LDS of claim 1, wherein the vehicle comprises a locomotive and the route comprises a locomotive track.

12. A method of determining real time location of a vehicle along a route, the method comprising:

providing a location determination system (LDS) comprising a computer or processor, a GPS processing module, speed bias based speed integration module, and a Kalman filter based estimator;

programming the computer or processor to measure GPS information via the GPS processing module, and measuring the GPS information via the GPS processing module associated with the vehicle;

programming the computer or processor to measure speed information via the speed bias based speed integration module and measuring the speed information via the speed bias based speed integration module associated with the vehicle;

estimating real time distance of the vehicle along the route solely in response to both the GPS information and the vehicle speed information via the Kalman filter based estimator; and

intermittently communicating a corrected value of speed bias to the speed bias based speed integration module and resetting the speed bias based speed integration module in response to both the corrected value of speed bias and predetermined critical milepost data that is distinct and independent of the GPS information and the vehicle speed information, such that the estimated real time distance is robust to cumulative errors and disturbances associated with both the GPS information and vehicle speed information to ensure the estimated real time location information is accurate.

13. The method of claim 12, further comprising rejecting GPS data or vehicle speed data via the Kalman filter based estimator that exceeds respective threshold limits.

14. The method of claim 12, wherein estimating real time distance of the vehicle along the route further comprises processing the most recent GPS information via projection of GPS values onto the database comprising the critical milepost entries and generating the corrected value of speed bias therefrom via the Kalman filter based estimator.

15. The method of claim 14, wherein projection of GPS values onto a database comprising the critical milepost entries comprises projection of GPS values onto a database comprising known latitude data, longitude data, and cumulative distance data associated with the route.

16. The method of claim 14, wherein projection of GPS values onto the critical milepost database comprises a great arc projection of GPS values onto a database comprising known latitude data, longitude data, and cumulative distance data associated with the route.

17. The method of claim 14, wherein projection of GPS values onto the critical milepost database comprises a non-linear least squares projection of GPS values onto a database comprising known latitude data, longitude data, and cumulative distance data associated with the route.

18. The method of claim 12, wherein estimating real time distance of the vehicle along the route comprises processing the GPS information and the vehicle speed information via the Kalman filter based estimator to generate a quality indicator signal associated with the estimated real time location information in response to GPS projection data and associated GPS covariance data.

19. The method of claim 12, wherein estimating real time distance of the vehicle along the route comprises processing the GPS information and the vehicle speed information via a two-state variable Kalman filter based estimator, wherein first state variable comprises vehicle position along the route and the second state variable comprises vehicle speed bias based upon the critical milepost database entries.

20. The method of claim 12, wherein estimating real time distance of the vehicle along the route comprises processing the GPS information and the vehicle speed information via a single-state variable Kalman filter based estimator, wherein the single-state variable comprises vehicle position along the route.

21. The method of claim 12, wherein estimating real time distance of the vehicle along the route comprises estimating real time distance of a locomotive along a locomotive track.

22. A vehicle location determination system (LDS) comprising:

a computer or processor;

a GPS processing module;

a speed integrator module;

a Kalman filter based estimator module;

wherein the computer or processor is programmed to estimate real time distance of the vehicle along a route via the Kalman filter based estimator that estimates the real time location in response solely to both GPS information from the GPS processing module and vehicle speed information from the speed integrator module; and

a critical milepost module comprising a database of identifying entries that are distinct and independent from the GPS information and the vehicle speed information, wherein the computer or processor is further programmed to reset the speed integrator module based upon the identifying entries in the database such that the estimated real time distance is robust to cumulative errors and disturbances associated with both the GPS information and vehicle speed information to ensure an estimated real time location is accurate.

23. The LDS of claim 22, wherein the identifying entries comprise:

longitude data;

latitude data; and

cumulative distance data, wherein the longitude data, latitude data, and cumulative distance data together define regular milepost and critical milepost estimation locations.

24. The LDS of claim 23, wherein the LDS further comprises a reset module and is further programmed to reset the speed integrator module via the reset module in response to an updated speed bias that is determined at each critical milepost identifying entry.

25. The LDS of claim 23, wherein the GPS processing module is programmed process the most recent GPS information via a great arc projection of GPS values onto the milepost database.

26. The LDS of claim 25, wherein the GPS processing module is further programmed to perform an unscented transformation (UT) for covariance calculations during GPS processing that is carried out only when a GPS value updates and the value is in range of the milepost database.

27. The LDS of claim 23, wherein the GPS processing module is programmed to process the most recent GPS information via a nonlinear least-squares projection of GPS values onto the milepost database.