ODTK 6.3.1 is a maintenance release containing various bug fixes and some small feature improvements.
ODTK 6.3.1 Features
User interface
Based on customer feedback, we have updated the Object Properties page to make it easier to use by eliminating the double-click action that was required to get into “edit mode” on a property value, and replaced it with a single-click behavior. We think this will make it much easier for users to quickly make updates.
We have also added a “Copy” button to several of the list boxes that pop up when you edit the CustomDataEditing, ResidualEditing, and CustomSigmas properties on the Scenario, Filter, and Least Squares objects. It was also added to the Least Squares StageList and the Simulator’s CustomTrackingInterval.Schedule list. This will make it easier to quickly create rules that are similar to existing ones you’ve already defined.
Least Squares
The Epoch Location control in least squares estimation has been changed to specify the least squares solution epoch in addition to specifying the epoch of the state to be reported. In prior versions of ODTK, the solution was always anchored to the beginning of each stage and then transitioned to the specified epoch location after convergence was achieved.
A new option was added to allow the user to specify how many iterations can diverge before stopping the least squares run. The default value is two, corresponding to the previously hard-coded limit.
Tracking systems
A preconfigured tracking system for the various ESA ESTRACK tracking stations was added to the ODTK install. While we believe the locations are correct, users are encouraged to verify the measurement statistics settings by processing real measurements.
Reporting and Graphing
Normal and tangential uncertainty data elements are now available for reporting and graphing.
6.3.1 Capabilities added and issues resolved from the prior release
| New Capabilities and Resolved Issues in ODTK 6.3.1 | |
|---|---|
| Tracking No. | Description |
| 38689 | When attempting to integrate an orbit in low altitudes, such as < 90 km, the integrator no longer performs poorly and hangs. |
| 74847 | ODTK would sometimes abnormally terminate while exiting when running on a Windows 8 box. This is corrected. |
| 76249 | The Epoch Location control in least squares estimation has been changed to specify the least squares solution epoch in addition to specifying the epoch of the state to be reported. In prior versions of ODTK, the solution was always anchored to the beginning of each stage and then transitioned to the specified epoch location after convergence was achieved. |
| 76322 | The RunSTKX HTML utility now correctly starts STK 10. |
| 76371 | ODTK no longer terminates when importing an STK facility with a very long name. |
| 76412 | Ephemeris file names generated by the LeastSquares object now contain words "Final" or "Initial" instead of iteration number if you set EphemerisArchiving option to anything other than "AllIterations" (i.e., "Initial Only" or "Initial and Final only"). This will ensure predictable file names regardless of how many iterations it took LS to converge. For example: Sat_Satellite1_LS_01_07_20110701_120000.e would become: Sat_Satellite1_LS_01_Final_20110701_120000.e |
| 76465 | This version fixes the bug when Intel MLK library was disabled (AGI_DISABLE_INTEL_MKL=1) and Variable Lag Smoother (VLS) was running through the variable state space boundary. The bug would cause memory beyond a non-square matrix to be corrupted and application crashing most of the times. Normal VLS operations (with Intel MKL enabled by default) had no issues. |
| 76552 | The ability to specify the number of consecutive iterations with increasing values for the weighted RMS required to declare LS divergence has been added. |
| 76559 |
A tracking system object was added to the install to model ESA's ESTRACK satellite tracking network. |
| 76564 | The least squares object now allows users to now solve for the position, velocity, or position and velocity. |
| 76627 | An error in the deweighting of GPS measurements (pseudorange, carrier phase and related differenced measurements) for the case where either the GPS satellite orbits OR the GPS satellite clocks are estimated (but not when both are estimated) has been corrected. The error did not affect scenarios where GPS orbits and clocks were both taken from a source file such as a SP3 formatted file. |
| 76657 | An error on the Solar Pressure help page is corrected. At the bottom where it discusses the GPS solar pressure models it documents the K2Model parameter and states that its "...multiplied by 1e-12 m/sec^2". The units on the number should be 1e-12 km/sec^2 (the change is from meters to kilometers). |
| 76752 | The ability to report trajectory differences and uncertainties in Normal and Tangential components has been added. |
| 77071 | A capability to specify a transfer epoch different than the solution epoch for LS when transferring the LS solution state to the satellite object has been added. |
| 77256 | The state correction selection made by the user in the least squares stage is now respected when using consider states. |
| 77318 | The ODTK Object properties page has been updated to use single-click to edit the properties rather than double-clicks. A copy button has been added to the CustomDataEditing, ResidualEditing, and the CustomSigmas properties on the Scenario, Filter, and Least Squares objects. It was also added to the Least Squares Stage list and Simulator CustomTrackingInterval Schedule list. |
| 77399 | The behavior of Least Squares processing when Absolute Convergence criteria are enabled has been altered to prevent any further state updates after convergence based on the Absolute Convergence criteria have been met. This was done to correct an issue where a state update performed on the iteration where Absolute Convergence was detected could lead to a new solution which did not satisfy the Absolute Convergence criteria. |
| 77564 | When user would manually change Filter.ResidualEditing.Schedule entry MeasType it didn't take effect until scenario reload. This is fixed now. |
| 77614 | An errant pre-run check in the Filter which prevented the Filter from running when a satellite existed in the scenario with only Eph Pos and/or Eph Vel in the MeasurementStatistics list of the satellite and that same satellite was not included in the Filter satellite list (due to other satellites being explicitly included in the Filter satellite list) has been corrected. |
| 77641 | Least squares solutions can now be transferred to the satellite in cases where the least squares process stopped due to exceeding the maximum number of iterations. |
| 77722 | If you are using CCSDS TDM format tracking data files and the PARTICIPANT names in the files are non-numeric, then you must now define aliases on the corresponding objects in your ODTK scenario and specify the alias rule set to use for each TDM file. This will allow ODTK to properly resolve the names of the PARTICIPANTS in the file to the proper ODTK object. If the PARTICIPANTS have strictly numeric names that correspond to an ODTK tracker ID then these will resolve correctly without using an alias.
Customers currently using CCSDS TDM files and who are relying on the previous behavior of naming their ODTK objects such that they match exactly with the names in the TDM files should update their scenarios to properly define aliases for the objects and specify the alias rule set when loading the TDM files into the scenario. |
| 77858 | The GPScatalog.txt file was updated to reflect the recent launch of SVN64. |
| 78044 | The bug was affecting ODTK 6.1, 6.2, and 6.3. Running the Filter would check out both ODMSS AND ODSNGL licenses from the network license server. The correct behavior in 6.3.1 is to first try to checkout ODMSS license and only if that fails will try to check out an instance of ODSNGL. |