next up previous contents index
Next: 18.4.2 Tracking Techniques Up: 18.4 Multitarget Tracking Previous: 18.4 Multitarget Tracking

18.4.1 Nature of the Problem


Sim89, described broadly in the previous section, is designed to process a so-called mass raid scenario, in which a few hundred primary threats are launched within a one- to two-minute time window, together with about 40 to 60 secondary, anti-satellite launches. The primary targets boost through two stages of powered flight (total boost time is about 300 seconds), with each booster ultimately deploying a single post boost vehicle (PBV). Over the next few hundred seconds, each PBV in turn deploys 10 re-entry vehicles (RVs). The Sim89 environment does not yet include the factor of 10 to 100 increase in object counts due to decoys, as would be expected in the ``real world.''

The data available for the tracking task consist, essentially, of line-of-sight measurements from various sensing platforms to individual objects in the target ensemble at (fairly) regular time intervals. At present, all sensing platforms are assumed to travel in circular orbits about a spherically symmetric earth (neither of these assumptions/simplifications is essential). The current program simulates two classes of sensors: GEO platforms, in geostationary, equatorial orbits, and MEO platforms in polar orbits at altitude . The scan time for GEO sensors is typically taken to be , with for MEO platforms.

Figure 18.16 shows a small portion of the field of view of a MEO sensor at a time about halfway through the RV deployment phase of a typical Sim89 scenario. The circles are the data seen by the sensor at one scan and the crosses are the data seen by the same sensor at a time later. Given such data, the primary tasks of the tracking program are fairly simple to state:

  1. Determine which data at one scan are associated with data from previous scans.
  2. For associated data points over several scans, determine the trajectory of the underlying targets.
Moreover, these tasks must be done in an extremely timely fashion. For example, the ASATs mentioned at the beginning of this subsection have burn times of order 100 seconds, and can hit a space-based asset in less that 250 seconds after launch. In order to allow self/mutual defense among the space-based assets, the tracker must provide good three-dimensional tracks for ASATs within about 60 seconds of launch.

Figure 18.16: Typical Midcourse Data Sets, Consecutive Scans

In order to (in principle) process data from a wide variety of sensors, the Sim89 tracker adopts a simple unified sensing formalism. For each sensor, the standard reference plane is taken to be the plane passing through the center of the earth, normal to the vector from the center of the earth to the (instantaneous) satellite position. Note that these standard frames are not inertial. The two-dimensional data used by the tracking algorithm are the coordinates of the intersection of the reference plane and the line of sight from the sensor to the target. The intersection coordinates are defined in terms of a standard Cartesian basis in the reference frame, with one axis along the normal to the sensor's orbital plane, and the other parallel to the projection of the platform velocity onto the reference plane.

next up previous contents index
Next: 18.4.2 Tracking Techniques Up: 18.4 Multitarget Tracking Previous: 18.4 Multitarget Tracking

Guy Robinson
Wed Mar 1 10:19:35 EST 1995