Data sharing among conditionally independent parallel filters

• US 9,104,752 B2
• Filed: 11/16/2012
• Issued: 08/11/2015
• Est. Priority Date: 11/16/2012
• Status: Expired due to Fees

• Current Assignee: Honeywell International Inc.
• Original Assignee: Honeywell International Inc.
• Inventors: Ingvalson, Ryan
• Primary Examiner(s): Allen, Brittany N
• Application Number:
  US13/679,170
• Publication Number:
  US 20140143262A1
• Time in Patent Office: 998 Days
• Field of Search: 707/754
• US Class Current: 1/1
• CPC Class Codes: G06F 16/335 Filtering based on additional data, e.g. user or group profiles filtering in web context G06F16/9535, G06F16/9536
• CPC Class Codes: G06F 9/54 Interprogram communication

Abstract

A method for data sharing between conditionally independent filters is provided. The method comprises initializing a plurality of conditionally independent filters operating in parallel, processing data measurements in each of the conditionally independent filters, sharing conditioning node estimates among the conditionally independent filters, merging the shared conditioning node estimates in each of the conditionally independent filters, and performing a conditional update in each of the conditionally independent filters.

Claims

1. A method for data sharing between a plurality of filters, the method comprising:

initializing at least a first filter and a second filter that operate in parallel with each other, the first filter having a first state vector (x₁), the second filter having a second state vector (x₂), the first filter estimating a distribution of the first state vector that contains a first subset (x_a) and a third subset (x_c), the second filter estimating a distribution of the second state vector that contains a second subset (x_b) and the third subset (x_c), such that;

x₁=[_xc^xa] and x₂=[_xc^xb];

wherein;

the first subset includes one or more unique states of the first filter, the second subset includes one or more unique states of the second filter, and the third subset includes one or more common states between the first and second filters;

the first subset, the second subset, and the third subset are selected such that the first subset and the second subset are conditionally independent given the third subset; and

the third subset is a conditioning node for the first filter and the second filter;

processing data measurements in the first filter;

sharing conditioning node estimates with the second filter when it is time for conditionally independent data sharing;

merging the shared conditioning node estimates in each of the filters; and

performing a conditional update in each of the filters.

2. The method of claim 1, wherein the filters each comprise a Bayesian tracking filter for stochastic state estimation.

3. The method of claim 2, wherein the Bayesian tracking filter comprises a Gaussian filter.

4. The method of claim 1, wherein the data measurements are based on sensor data processed by one or more sensors.

5. The method of claim 1, wherein the conditioning node estimates are based on states and prior data measurements processed in the filters.

6. The method of claim 1, wherein the shared conditioning node estimates are merged with a track fusion algorithm.

7. The method of claim 1, wherein each of the filters is located in a separate filter processing module of a plurality of filter processing modules whose filters are related to each other in a dependence network.

8. The method of claim 7, wherein the dependence network comprises a singly-connected network.

9. The method of claim 1, wherein each of the filters is located in a separate filter processing module that is part of a collaborative network.

10. A method for data sharing between a plurality of filters, the method comprising:

(a) initializing at least a first filter and a second filter that operate in parallel with each other, the first filter having a first state vector (x₁) the second filter having a second state vector (x₂), the first filter estimating a distribution of the first state vector that contains a first subset (x_a) and a third subset (x_c), the second filter estimating a distribution of the second state vector that contains a second subset (x_b) and the third subset (x_c), such that;

x₁=[_xc^xa] and x₂=[_xc^xb];

wherein;

the first subset includes one or more unique states of the first filter, the second subset includes one or more unique states of the second filter, and the third subset includes one or more common states between the first and second filters;

the first subset, the second subset, and the third subset are selected such that the first subset and the second subset are conditionally independent given the third subset; and

the third subset is a conditioning node for the first filter and the second filter;

(b) propagating the first filter forward in time;

(c) determining whether one or more measurements have been made;

(d) if one or more measurements have been made;

(i) updating the first filter with the measurements; and

(ii) determining whether it is time for conditionally independent data sharing;

(e) if one or more measurements have not been made, determining whether it is time for conditionally independent data sharing;

(f) if it is not time for conditionally independent data sharing, repeating the method starting at (b);

(g) if it is time for conditionally independent data sharing, extracting a current conditioning node estimate from the first filter;

(h) sending the current conditioning node estimate to the second filter;

(i) receiving one or more shared conditioning node estimates from the second filter;

(j) merging the one or more shared conditioning node estimates with the current conditioning node estimate; and

(k) updating the first filter with the merged conditioning node estimates.

11. The method of claim 10, wherein the one or more shared conditioning node estimates and the current conditioning node estimate are merged with a track fusion algorithm.

12. The method of claim 10, wherein each of the first filter and the second filter is located in a separate module of a plurality of modules, where the filters are probabilistically dependent upon one another in a network.

13. The method of claim 12, wherein the network comprises a singly-connected network.

14. The method of claim 10, wherein the method further comprises:

(I) after updating the first filter with the merged conditioning node estimates, repeating the method starting at (b).

15. A computer program product, comprising:

a non-transitory computer readable medium having instructions stored thereon executable by a processor to perform a method for data sharing between a plurality of filters according to claim 10.

16. A system for data sharing between a plurality of filters, the system comprising:

a first filter;

at least a second filter operating in parallel with the first filter;

wherein the first filter has a first state vector (x₁), the second filter has a second state vector (x₂), the first filter estimates a distribution of the first state vector that contains a first subset (x_a) and a third subset (x_c), the second filter estimates a distribution of the second state vector that contains a second subset (x_b) and the third subset (x_c), such that;

x₁=[_xc^xa] and x₂=[_xc^xb];

wherein;

the first subset includes one or more unique states of the first filter, the second subset includes one or more unique states of the second filter, and the third subset includes one or more common states between the first and second filters;

the first subset, the second subset, and the third subset are selected such that the first subset and the second subset are conditionally independent given the third subset; and

the third subset is a conditioning node for the first filter and the second filter;

one or more processors in operative communication with the first and second filters;

a non-transitory computer readable medium having instructions stored thereon executable by the one or more processors to perform a method for data sharing comprising;

initializing the first filter and the second filter with the same data measurement set, wherein the first filter has a first probability density function estimate and the second filter has a second probability density function estimate;

receiving a first independent measurement at the first filter;

updating the first probability density function estimate with the first independent measurement;

receiving a second independent measurement at the second filter;

updating the second probability density function estimate with the second independent measurement;

extracting a conditioning node estimate from the first filter;

extracting a conditioning node estimate from the second filter;

combining the conditioning node estimates from the first and second filters to produce a merged conditioning node estimate;

running a conditional update for the first filter with the merged conditioning node estimate; and

running a conditional update for the second filter with the merged conditioning node estimate.

17. The system of claim 16, wherein the merged conditioning node estimate is produced using a track fusion algorithm.

18. The system of claim 16, wherein each of the first and second filters is located in a separate module of a plurality of modules, where the filters are probabilistically dependent upon one another in a network.

19. The system of claim 18, wherein the network comprises a singly-connected network.

20. The system of claim 16, wherein the first filter is located in a first platform of a collaborative network, and the second filter is located in a second platform of the collaborative network.