Controller and method of controlling an apparatus using predictive filters

• US 7,107,108 B2
• Filed: 06/05/2002
• Issued: 09/12/2006
• Est. Priority Date: 06/05/2001
• Status: Expired due to Term

• Current Assignee: Zarbana Digital Fund LLC (Intellectual Ventures LLC)
• Original Assignee: Elverson Hopewell LLC (Intellectual Ventures LLC)
• Inventors: Pohr, Bernd, Woergoetter, Florentin
• Primary Examiner(s): BARNES-BULLOCK, CRYSTAL JOY
• Application Number:
  US10/479,741
• Publication Number:
  US 20050027373A1
• Time in Patent Office: 1,560 Days
• Field of Search: 700/28, 700/29, 700/32, 700/33, 700 37- 39, 700/44, 700/45, 700/47, 700/51, 700/54, 700/55, 706/14, 706 21- 23, 706/58, 318/561
• US Class Current: 700/39
• CPC Class Codes: G05B 13/0265 the criterion being a learning criterion

Abstract

In the field of electronic or computerised control apparatuses, known methods of compensating for disturbance signals rely on less than ideal analytical or empirical models. There is provided a controller (200) or method of controlling which observes and learns the correlation between various measured signals and automatically learns how to control the apparatus. It has a primary input (x_R) for the measured state of the apparatus, and signal processing means (604, 606) responsive to the input signal for generating a control signal (V) to maintain the apparatus in a desired state, the controller (200) having further input(s) (X₁, X_n) for additional measurements of the apparatus or its environment, the signal processing means using a weighted average (606) of results of different fixed impulse responses (611–615) to each input to modify the control signal, and means (630, 800, 802) for conditioning the response automatically in response to temporal cross-correlation observed between the measurement signal(s) and the control signal. The controller can be pre-taught with an appropriate response, or it can learn it during initial operation.

Claims

1. A controller for controlling an apparatus, comprising:

first input means for receiving a primary input signal representing a measured state of the apparatus;

signal processing means responsive to said primary input signal for generating a control signal to alter the state of the apparatus to maintain a desired state;

second input means for receiving a signal representing at least one additional measurement of the state of the apparatus,

said signal processing means including correcting means for correcting said control signal in response to said signal representing at least one additional measurement signal, and conditioning means for conditioning a response of said correcting means based at least in part on temporal cross-correlation between said signal representing at least one additional measurement signal and said control signal.

2. A controller as claimed in claim 1, wherein the conditioning means comprises means for adjusting a gain and frequency response of a response of said correcting means.

3. A controller as claimed in claim 1, wherein the correcting means comprises:

means for filtering the signal representing at least one additional measurement via a plurality of filters having different fixed impulse responses; and

means for forming a weighted sum of the filtered signal to determine a correction to the control signal.

4. The controller of claim 3, wherein at least a portion of said plurality of filters comprise predictive filters.

5. A controller as claimed in claim 1, wherein the conditioning means is to multiply the signal of each path with a derivative of the control signal and to integrate the product of said signals over time.

6. A controller as claimed in claim 5, wherein said integration is capable of being held at a fixed value such that said conditioning is maintained in the absence of stimulation of said apparatus.

7. A controller as claimed in claim 1, further comprising a plurality of additional inputs, at least a portion of said plurality of additional inputs including associated correcting means and/or conditioning means.

8. A controller as claimed in claim 7, wherein different filter responses are used by at least a portion of said correcting means.

9. A controller as claimed in claim 7, wherein different quantities of filters are used by at least a portion of said correcting means.

10. A controller as claimed in claim 1 arranged to generate plural control signals based on at least some of the same input signals, the signal processing means including correcting means and conditioning means for generation of at least a portion of said plural control signals.

11. A controller as claimed in claim 10, wherein a first control signal comprises an additional measurement input to a correcting means for generating a second control signal.

12. A controller as claimed in claim 11, wherein said measurement input comprises a derivative of said first control signal.

13. A controller as claimed in claim 1, wherein the correcting means comprises a fixed response.

14. A controller as claimed in claim 1, wherein said correcting means comprises a non-zero response, wherein said controller is adapted to adjust the response of the correcting means.

15. A controller as claimed in claim 1, wherein said correcting means comprises one or more filter components.

16. The controller of claim 1, wherein said signal representing at least one additional measurement of the state of the apparatus comprises a disturbance signal.

17. A method of controlling a physical apparatus, the method comprising:

receiving a primary input signal representing a measured state of the apparatus;

generating a control signal to alter the state of the apparatus in response to said primary input signal;

receiving at least one further input signal representing one or more additional measurements of the apparatus;

including corrections in said control signal in response to said additional measurement signal; and

conditioning the response of said corrections based at least in part on temporal cross-correlation between said additional measurement signal and said control signal.

18. A method as claimed in claim 17, further comprising filtering the additional signal via a plurality of filter functions, at least a portion of said filter functions employing different fixed impulse responses, forming a weighted sum of the differently filtered signals to derive the correction to be applied, and adjusting the weightings of the different signal paths in response to their respective observed correlation with the control signal.

19. The method of claim 18, wherein at least a portion of said filter functions comprise predictive filter based functions.

20. The method of claim 19, wherein at least a portion of said filter functions comprise cross-modal predictive filter based functions.

21. A method as claimed in claim 17, wherein the conditioning begins from a zero condition.

22. The method of claim 17, wherein said at least one further input signal comprises a disturbance signal.

23. The controller of claim 22, wherein at least a portion of said plurality of filters comprise cross-modal predictive electronic filters.

24. A method as claimed in claim 17, wherein the conditioning begins with a non-zero set of conditions.

25. A method as claimed in claim 24, wherein the conditioning begins with a non-zero set of conditions learned in another apparatus.

26. A method of controlling an apparatus, comprising:

receiving an input signal;

generating a control signal based at least in part on said input signal;

receiving a disturbance signal; and

conditioning the control signal based at least in part on temporal cross-correlation between said input signal and said disturbance signal.

27. The method of claim 26, wherein conditioning further comprises filtering at least a portion of said input signal and said disturbance signal via a plurality of filter functions.

28. The method of claim 27, wherein at least a portion of said filter functions comprise predictive filter based functions.

29. The method of claim 28, wherein at least a portion of said filter functions comprise cross-modal predictive filter based functions.

30. The method of claim 26, wherein the conditioning is based at least in part on a zero condition. condition.

31. The method of claim 26, wherein the conditioning is based at least in part on a non-zero set of conditions.

32. The method of claim 26, further comprising:

receiving a plurality of disturbance signals;

filtering at least a portion of the disturbance signals; and

providing a summed disturbance signal based at least in part on said filtering.

33. An apparatus, comprising:

a controller adapted to receive at least an input signal and a disturbance signal, said controller further adapted to;

generate a control signal based at least in part on said input signal; and

condition the control signal based at least in part on temporal cross-correlation between said input signal and said disturbance signal.

34. The apparatus of claim 33, wherein said controller is further adapted to filter at least a portion of said input signal and said disturbance signal via a plurality of filter functions.

35. The apparatus of claim 34, wherein at least a portion of said filter functions comprise predictive filter based functions.

36. The apparatus of claim 35, wherein at least a portion of said filter functions comprise cross-modal predictive filter based functions.

37. The apparatus of claim 33, wherein the conditioning is based at least in part on a zero condition.

38. The apparatus of claim 33, wherein the conditioning is based at least in part on a non-zero set of conditions.

39. The apparatus of claim 33, wherein said controller is further adapted to:

receive a plurality of disturbance signals;

filter at least a portion of the disturbance signals; and

provide a summed disturbance signal based at least in part on said filtering.