Fusion Methods

fusion_method <fusion-type-name>
   ... commands ...
end_fusion_method

Specifies the fusion algorithms used by the track manager.

A fusion algorithm combines information about a single entity from two or more sources into a coherent information set, or track.

<fusion-method> can be:

Each type has its own unique input keywords.

Available Fusion Methods

replacement

fusion_method replacement
end_fusion_method

Correlated measurements and tracks are fused according to a standard set of algorithms. Local track positions are replaced by nonlocal track positions.

weighted_average

fusion_method weighted_average
end_fusion_method

Correlated measurements and tracks are fused according to a standard set of algorithms. Local track positions are combined with nonlocal track positions using the covariance matrices of the local and nonlocal tracks.

Note

If there is no covariance matrix associated with a nonlocal track, the track manager will attempt to use a measurement covariance matrix, generated from the track’s measurement errors in range, elevation, and bearing.

orbit_determination

fusion_method orbit_determination
   number_of_angle_measurements   <int>
   angles_only_linear_tolerance   <length-value>
   angles_only_maximum_iterations <int>
   lambert_convergence_tolerance  <real>
   process_noise_sigmas_XYZ ...
   debug
   debug_filter
end_correlation_method

Orbit determination fusion incorporates algorithms that fuse multiple sensor measurements to provide an inital estimate of a satellite’s orbit, then continue to fuse subsequent measurements to update and refine the initial estimate. Target measurements either from angles-only sensors, providing bearing-elevation tracks; or measurements that also provide range may be used. Once an orbit is initially determined, an ORBIT_DETERMINATION_INITIATED event is triggered. On subsequent orbit determination updates, a ORBIT_DETERMINATION_UPDATED event is triggered.

Note

It is highly recommended to set retain track history true in the track manager in order to accumulate sufficient measurements to initiate orbit determination. Also, in order to manage the total number of measurements to be retained, set the track history retention interval time in any track processor utilizing this fusion strategy.

process_noise_sigmas_XYZ <X-value> <Y-value> <Z-value>

Defines the noise standard deviation used by the embedded filter. The values correspond with accelerations in the entity coordinate system (ECS) of the tracked platform.

Default: 0 0 0

number_of_angle_measurements <integer>

Specify the number of angles-only measurements to be collected before an angles-only iniitial orbit determination attempt is made.

Note

This value must be at least 5.

Default 5

angles_only_linear_tolerance <length-value>

Specify the linear tolerance requried to reach a solution with the angles-only initial orbit determination algorithm.

..note:: It is usually not necessary to set this value.

Default 1000 meters

angles_only_maximum_iterations <integer>

Specify the maximum number of iterations for the angles-only initial orbit determination algorithm to find a solution.

..note:: It is usually not necessary to set this value.

Default 100

lambert_convergence_tolerance <real>

Specify a unitless tolerance for convergence of the lambert universal-variables algorithm of two locations and time.

Default 1.0e-8

debug

Prints debug information to std::out.

debug_filter

Specifies saving of filtering history information (see filter debug).

mtt

fusion_method mtt
end_fusion_method

Correlated measurements and tracks are fused using the Multiple Target Tracker (MTT). If this method is selected, the correlation_method must also be mtt.