Encoder signal analysis system for high-resolution position measurement

• US 7,865,324 B2
• Filed: 01/07/2010
• Issued: 01/04/2011
• Est. Priority Date: 02/03/2006
• Status: Expired due to Fees

• Current Assignee: Moog Incorporated
• Original Assignee: Moog Incorporated
• Inventors: Schultze, Gary, Lindberg, Paul M., Hollingsworth, Philip
• Primary Examiner(s): Kundu; Sujoy K
• Application Number:
  US12/655,773
• Publication Number:
  US 20100114518A1
• Time in Patent Office: 362 Days
• Field of Search: 702/57, 702/73, 702/85, 702/94, 702/104, 702/106, 702/150, 702/151, 318/127, 318/128, 318/490, 318/560, 318/652
• US Class Current: 702/94
• CPC Class Codes: G01D 18/001 Calibrating encoders
• CPC Class Codes: G01D 5/24476 Signal processing G01D5/2448 - G01D5/24495 take precedence
• CPC Class Codes: G01D 5/2448 Correction of gain, threshold, offset or phase control
• CPC Class Codes: G01D 5/2449 using hard-stored calibration data
• CPC Class Codes: H02P 6/006 Controlling linear motors

Abstract

An apparatus and method for identifying the position of a magnetic shaft are provided. N field sensors are adjacently positioned at fixed locations relative to the shaft'"'"'s periodic field, corresponding to 180/N relative phase shifts. A table provides N>2 predetermined signal models and a pre-identified position associated with each. An interpolator compares a representation of the N measured sensor signals to at least two predetermined models to generate a correction signal that provides another pre-identified position. The correction signal depends on N sensors for every position of the shaft. The correction signal is used to incrementally choose said another pre-identified position from the table as an approximate position of the shaft in an iterative process to find the minimum correction signal and identify the position.

Claims

1. An apparatus for identifying the position of a first member relative to a defined path;

said path being defined by a second member;

one of said first and second member providing a periodic field along at least a portion of said defined path, said apparatus comprising;

a storage means comprising an addressable table comprising a plurality of pre-determined sets of N sequentially ordered values, where N is greater than 2, and a value for a pre-identified position associated with each predetermined set;

an array of spaced field sensors positioned at fixed locations with respect to the other of said first and second member not providing said periodic field and adjacent to at least a portion of said defined path, each of said field sensors sensing said portion of said periodic field at its fixed location and thereby generating a sensor signal in response thereto; and

a statistical interpolator comprising an error signal calculation module and an interpolation module;

said error signal calculation module comprising a sensor signal processing circuit operably connected to said array for receiving said sensor signals, said sensor signal processing circuit generating a set of N sensor signals in a predetermined sequence associated with the position of said first member in response to said sensor signals, said error signal calculation module also operably connected to said storage means for selectively receiving said plurality of predetermined sets of N sequentially ordered values, said error signal calculation module further comprising a comparator for comparing said set of N sensor signals to at least one of said plurality of predetermined sets of N sequentially ordered values associated with one pre-identified position to generate a correction signal based on said set of N sensor signals, wherein said correction signal is determined by at least two of said predetermined sets and by said set of sensor N signals for every position of said first member relative to said defined path; and

said interpolation module being operably connected to said error signal calculation module to receive said correction signal and to return an indexing signal associated with another pre-identified position to said error signal calculation module, said statistical interpolator programmed to use said correction signal to incrementally choose said another pre-identified position from said table as an approximation of the position of said first member in an iterative process to find the minimum correction signal and to thereby identify the position of said first member relative to said defined path.

2. The apparatus of claim 1, wherein said array of spaced field sensors comprises M field sensors spaced by a distance corresponding to a non-zero integer multiple of 180/M degrees of said periodic field.

3. The apparatus of claim 2, wherein M equals N, each of said regularly spaced sensors providing one of said N sensor signals.

4. The apparatus of claim 3, wherein M is equal to 3.

5. The apparatus of claim 2, wherein M equals two times N, said array comprising N pairs of regularly spaced field sensors, each pair comprising one field sensor physically located relative to the other field sensor to sense said periodic field shifted in phase by 180 degrees relative to the other field sensor, said sensor signal processing circuit comprising a differential circuit for each of said N pairs to subtract one of a pair of sensor signals from the other to generate said set of N sensor signals associated with the position of said first member, thereby substantially compensating for slow field offset variations.

6. The apparatus of claim 1, wherein said periodic field is provided by an array of permanent magnets on the shaft of a linear motor, and wherein said field sensors are magnetic sensors.

7. The apparatus of claim 6, wherein said magnetic sensors comprise at least one of Hall sensors and magnetoresistive sensors.

8. The apparatus of claim 1, wherein said statistical interpolator comprises a digital signal processor.

9. The apparatus of claim 1, wherein said comparator is adapted to:

calculate a square of the magnitude of the difference between said set of N sensor signals and at least said one of and a second one of said plurality of predetermined sets of N sequentially ordered values associated with a second pre-identified position for each of at least said one and said second predetermined sets to generate at least two error values that are a function of pre-identified position;

calculate a slope of the function at said one pre-identified position defined by said at least two error values to determine a direction and amount by which said at least two error values change as a function of pre-identified position; and

multiply said slope by a scaling factor to generate a correction value associated with said correction signal.

10. The apparatus of claim 9, wherein said interpolation module is adapted to subtract said correction value from said one pre-identified position to choose said another pre-identified position.

11. The apparatus of claim 1, said storage means comprising memory having addressable memory locations, wherein said statistical interpolator comprises an index pointer associated with said indexing signal, said index pointer being a scaled representation of said pre-identified positions, said statistical interpolator adapted to address said addressable memory locations using said index pointer to choose said another pre-identified position.

12. A method for identifying the position of a first member relative to a defined path;

said path being defined by a second member;

one of said first and second member providing a periodic field along at least a portion of said defined path, said method comprising the steps of;

(a) providing an array of more than two spaced field sensors positioned at fixed locations with respect to the other of said first and second member not providing said periodic field and adjacent to at least a portion of said defined path, each of said field sensors sensing said portion of said periodic field at its fixed location and thereby generating a sensor signal in response thereto to provide a set of N sensor signals in a predetermined sequence associated with the position of said first member relative to said defined path, where N is greater than two;

(b) providing a plurality of mathematically representative predetermined sensor signal models, each associated with a pre-identified position of said first member relative to said defined path;

(c) comparing a mathematical representation of said set of N sensor signals to a first one of said plurality of mathematically representative predetermined sensor signal models associated with a first of said pre-identified positions to generate a correction signal; and

(d) using said correction signal to generate an output position signal and to provide another one of said pre-identified positions associated with another one of said plurality of predetermined sensor signal models to be compared to said mathematical representation of said set of N sensor signals.

13. The method of claim 12, further comprising the additional steps of:

(e) iteratively repeating steps (c), and (d) to find a minimum correction signal, thereby identifying the position of said first member relative to said defined path.

14. The method of claim 13, wherein said mathematical representation of said N set of sensor signals is a vector representation, and wherein said comparing step (c) comprises the steps of:

generating a vector representation for at least a first one and a second one of said plurality of mathematically representative predetermined sensor signal models associated with a first and a second pre-identified position respectively;

calculating an error signal representing a square of the magnitude of a difference between said vector representation of said set of N sensor signals and at least said first one and said second one of said vector representations of said plurality of mathematically representative predetermined sensor signal models for each of at least said first and said second mathematically representative predetermined sensor signal models to generate at least two error values that are a function of pre-identified position;

calculating a slope of said function defined by said at least two error values to determine a direction and an amount by which said at least two error values change as a function of pre-identified position at said first pre-identified position; and

multiplying said slope by a scaling factor to generate a correction value associated with said correction signal.

15. The method of claim 14, wherein said using step (d) comprises the step of subtracting said correction value from said first pre-identified position to provide said another pre-identified position.

16. The method of claim 15, wherein said correction value corresponds to said minimum correction signal, each of said slope and said correction value is zero, said another pre-identified position equals said first pre-identified position, and said output position signal identifies said first pre-identified position as the position of said first member relative to said defined path.

17. The method of claim 14, wherein said scaling factor is a positive or negative constant value.

18. The method of claim 14, wherein said scaling factor is automatically adjusted at every iteration of said comparing step.

19. The method of claim 12, wherein said providing step (b) comprises the step of providing a table of said plurality of mathematically representative predetermined sensor signal models indexed by said pre-identified positions associated therewith, and

wherein said using step (d) comprises the step of using said output position signal to step an index pointer in said table to provide said another pre-identified position.

20. The method of claim 12, wherein said array of spaced field sensors comprises M sensors spaced by a distance corresponding to a non-zero integer multiple of 180/M degrees of said periodic field.

21. The method of claim 20, wherein M equals N, each of said spaced sensors providing one of said N sensor signals in said set of said N sensor signals.

22. The method of claim 20, wherein M equals 2N, wherein said providing step (a) comprises the step of providing said array as N pairs of regularly spaced field sensors, each pair comprising one field sensor physically located relative to the other field sensor to sense said periodic field shifted in phase by 180 degrees relative to the other field sensor, said array generating N combined sensor signals from said sensor signals to provide said set of N sensor signals in said pre-determined sequence associated with the position, and

wherein said method further comprises the step of generating said N combined sensor signals by subtracting one sensor signal generated by said one field sensor generated by a second sensor signal from said other field sensor for each of said N pairs, thereby substantially compensating for slow field offset variations.

23. An apparatus for identifying the position of a first member relative to a defined path;

said path being defined by a second member;

one of said first and second member providing a periodic field along at least a portion of said defined path, said apparatus comprising;

an array of spaced field sensors of number greater than two positioned at fixed locations with respect to the one of said first and second member not providing said periodic field and adjacent to at least a portion of said defined path, each of said field sensors sensing said portion of said periodic field at its fixed location and thereby generating a sensor signal in response thereto;

a storage means comprising an addressable table comprising a plurality of mathematically representative predetermined sensor signal models, and a value for a pre-identified position associated with each predetermined sensor signal model; and

a statistical interpolator comprising an error signal calculation module and an interpolation module;

said error signal calculation module comprising a sensor signal processing circuit operably connected to said array for receiving said sensor signals, said sensor signal processing circuit generating a set of N sensor signals in a predetermined sequence associated with the position of said first member in response to said sensor signals, said error signal calculation module also operably connected to said storage means for selectively receiving said plurality of mathematically representative predetermined sensor signal models, said error signal calculation module further comprising a comparator for comparing a mathematical representation of said set of N sensor signals to at least one of said plurality of mathematically representative predetermined sensor signal models associated with one pre-identified position to generate a correction signal based on said set of N sensor signals, wherein said correction signal is determined by comparing said at least one of said plurality or mathematically representative predetermined signal models to said mathematical representation set of said set of N signals for every position of said first member relative to said defined path; and

said interpolation module being operably connected to said error signal calculation module to receive said correction signal and to return an indexing signal associated with another pre-identified position to said error signal calculation module, said statistical interpolator programmed to use said correction signal to incrementally choose said another pre-identified position from said table as an approximation of the position of said first member in an iterative process to find the minimum correction signal and to thereby identify the position of said first member relative to said defined path.

24. The apparatus of claim 23, wherein said array of spaced field sensors comprises M field sensors spaced by a distance corresponding to a non-zero integer multiple of 180/M degrees of said periodic field.

25. The apparatus of claim 24, wherein M equals N, each of said spaced sensors providing one of said N sensor signals.

26. The apparatus of claim 25, wherein M is equal to 3.

27. The apparatus of claim 24, wherein M equals two times N, said array comprising N pairs of regularly spaced field sensors, each pair comprising one field sensor physically located relative to the other field sensor to sense said periodic field shifted in phase by 180 degrees relative to the other field sensor, said sensor signal processing circuit comprising a differential circuit for each of said N pairs to subtract one of a pair of sensor signals from the other to generate said set of N sensor signals associated with the position of said first member, thereby substantially compensating for slow field offset variations.

28. The apparatus of claim 23, wherein said periodic field is provided by an array of permanent magnets on the shaft of a linear motor, and wherein said field sensors are magnetic sensors.

29. The apparatus of claim 28, wherein said magnetic sensors comprise at least one of Hall sensors and magnetoresistive sensors.

30. The apparatus of claim 23, wherein said statistical interpolator comprises a digital signal processor.

31. The apparatus of claim 23, wherein said comparator is adapted to:

calculate a square of the magnitude of a vector difference between said mathematical representation of said set of N sensor signals and a vector representation of each of at least said one of and a second one of said plurality of mathematically representative predetermined sets of N ordered values associated with a second pre-identified position for each of at least said one and said second mathematically representative predetermined sets to generate at least two error values that are a function of pre-identified position;

calculate a slope of the function defined by said at least two error values to determine a direction and amount by which said at least two error values change as a function of pre-identified position; and

multiply said slope by a scaling factor to generate a correction value associated with said correction signal.

32. The apparatus of claim 31, wherein said interpolation module is adapted to subtract said correction value from said one pre-identified position to generate said indexing signal for providing said another pre-identified position.

33. The apparatus of claim 31, said storage means comprising memory having addressable memory locations, wherein said statistical interpolator comprises an index pointer associated with said indexing signal, said index pointer being a scaled representation of said pre-identified positions, said statistical interpolator adapted to address said addressable memory locations using said index pointer to choose said another pre-identified position.