System and method of space object tracking and surveillance network control

1. A computer implemented method for improved tracking of objects in earth orbit by a Space Surveillance Network (SSN), the method comprising:

receiving a dataset comprising initial state estimations of either a group of uncorrelated objects or each tracked object in earth orbit;

partitioning the dataset into a plurality of clusters of observations;

for each cluster of observations, generating an orbital element set description of the cluster of observations;

for each cluster of observations, determining a true anomaly location on the respective orbit associated with providing a largest increase in accuracy to the respective orbital description;

for each cluster of observations, generating a tasking order configured to cause SSN observation of a projected position-time pair; and

transmitting said generated tasking orders toward the SSN.

2. The computer implemented method of claim 1 , wherein partitioning the dataset is performed using an algorithmic similarity measure.

3. The computer implemented method of claim 2 , wherein the algorithmic similarity measure incorporates global information and spectral clustering.

4. The computer implemented method of claim 1 , wherein determining a true anomaly is made using either Bayesian inference or identification of a true anomaly furthest from all previous observations on orbit.

5. The computer implemented method of claim 1 , wherein the projected position-time pair is associated with the true anomaly location on the respective orbit.

6. The computer implemented method of claim 3 , wherein the spectral clustering is performed in accordance with at least one of a diffusion map technique and an eigenspace technique.

7. The computer implemented method of claim 1 , wherein partitioning the dataset is performed in accordance with expert heuristics.

8. The computer implemented method of claim 1 , wherein generating the orbital element set description is performed by finding a least squares fit of an elliptical vector function to the cluster of observations.

9. The computer implemented method of claim 8 , wherein the least squares fit uses a stochastic gradient descent mechanism.

10. The computer implemented method of claim 1 , wherein generating the orbital element set description is performed using batched least squares.

11. The computer implemented method of claim 1 , wherein the orbital element set description includes differential corrections in orbit determination.

12. The computer implemented method of claim 1 , further comprising receiving dataset updates comprising time-stamped position information associated with each of the plurality of clusters of observations.

13. The computer implemented method of claim 1 , wherein the dataset is partitioned into clusters associated by likelihood of being the same object through the use of algorithmic similarity measure and spectral clustering techniques.

14. The computer implemented method of claim 12 , wherein historical discounting is applied to said time-stamped position information prior to either fitting an elliptical vector value function to each cluster of observations or implementing a standard batched least squares algorithm to incorporate the differential corrections to the initial orbit estimation.

15. The computer implemented method of claim 8 , wherein fitting an elliptical vector value function to a cluster of observations is performed using an adaptive stochastic gradient descent algorithm developed for this purpose.

16. The computer implemented method of claim 1 , wherein determining said true anomaly locations is performed using a Bayesian Inference approach.

17. The computer implemented method of claim 1 , wherein determining said true anomaly locations is performed using identification of a true anomaly furthest from all previous observations on orbit.

18. The computer implemented method of claim 1 , wherein identifying within the SSN Field of View the closest true anomalies to the output of the Bayesian estimation algorithm.

19. The computer implemented method of claim 1 , further comprising transmitting one or more SSN tasking orders toward appropriate SSN elements within the SSN.

20. The computer implemented method of claim 1 , wherein for each cluster of observations said generated tasking order identifies an optimal location in orbit for improving state estimations of the cluster of observations.

21. The computer implemented method of claim 20 , wherein for each cluster of observations said generated tasking order identifies SSN elements suitable for use in viewing the optimal location of the cluster of observations.

22. A computer implemented method for improved tracking of objects in earth orbit by a Space Surveillance Network (SSN), the method comprising:

receiving a set of observations comprising time-stamped position-based observation data for uncorrelated objects along with relevant error estimates for each observation;

determining a number of unique objects represented by the set of observations;

partitioning the set of observations into clusters associated with a unique object;

implementing initial orbital determination methods along with batched least squares differential corrections to produce a state estimation for each object;

implementing a Bayesian inference method to identify an optimal orbital location for improving the state estimations; and

transmitting said generated tasking orders toward the SSN.

23. The computer implemented method of claim 22 , wherein the time-stamped position-based observation data is received in the form of a covariance matrix.

24. The computer implemented method of claim 22 , wherein the Bayesian inference method also identifies a closest possible tasking order using known SSN fields of view.

25. An apparatus for improved tracking of objects in earth orbit by a Space Surveillance Network (SSN), the apparatus comprising:

a tracking/tasking module, operatively coupled to a SSN via a network;

the tracking/tasking module comprising a dataset retrieval module configured for receiving from the SSN a set of observations in the form of a covariance matrix;

the tracking/tasking module comprising a dataset processing module configured for determining a number of unique objects represented by the set of observations;

the tracking/tasking module comprising a SSN tasking module configured for identifying a closest possible tasking order using known SSN fields of view, and transmitting said generated tasking orders toward the SSN.

26. The apparatus of claim 25 , wherein the set of observations comprise comprising time-stamped position-based observation data for uncorrelated objects.

27. The apparatus of claim 25 , wherein the set of observations comprise relevant error estimates for each observation.

28. The apparatus of claim 25 , wherein the dataset processing module is configured for partitioning the set of observations into clusters associated with a unique object.

29. The apparatus of claim 25 , wherein the dataset processing module is configured for implementing initial orbital determination methods along with batched least squares differential corrections to produce a state estimation for each object.

30. The apparatus of claim 25 , wherein the dataset processing module is configured for implementing a Bayesian inference method to identify an optimal orbital location for improving the state estimations.

31. A tangible and non-transient computer readable storage medium storing computer-executable instructions which, when executed by a system of computer processors, adapt the operation of the system to perform a method for improved tracking of objects in earth orbit by a Space Surveillance Network (SSN), comprising:

receiving a set of observations comprising time-stamped position-based observation data for uncorrelated objects along with relevant error estimates for each observation;

determining a number of unique objects represented by the set of observations;

partitioning the set of observations into clusters associated with a unique object;

implementing initial orbital determination methods along with batched least squares differential corrections to produce a state estimation for each object;

identifying an optimal orbital location for improving the state estimations and identifying a closest possible tasking order; and

transmitting said generated tasking orders toward the SSN.

32. The tangible and non-transient computer readable storage medium of claim 31 , wherein the time-stamped position-based observation data is received in the form of a covariance matrix.

33. The tangible and non-transient computer readable storage medium of claim 31 , wherein the identification of an optimal orbital location for improving the state estimations is performed by Bayesian inference.

34. The tangible and non-transient computer readable storage medium of claim 31 , wherein the identification of a closest possible tasking order is performed using known SSN fields of view.