Automatic Signature Collection for Tracking and Identification
Small Business Information
6 New England Executive Park, Burlington, MA, 01803
AbstractGround target tracking performance can be improved through the use of appropriately collected target signatures. In particular, associating reports and tracks is especially enhanced in dense target environments and move-stop-move scenarios. To support association, the signatures need to be collected at a diverse set of geometries so that a report signature can be associated with a signature collected previously at a comparable geometry. In our phase 1 effort, we developed a sensor management algorithm for collecting HRR MTI position and signature data at aspect angles that are predicted to be of use in the future. The algorithm also tasks the collection of FTI position data on stopped targets so as to track targets as they start and stop. For the phase 2, we are proposing to build on the algorithmic framework developed in phase 1. We will be developing more sophisticated sensor models, refining our baseline sensor management algorithms, and demonstrating and testing our algorithms in closed-loop with sensor data. The goal is to illustrate, in a realistic simulation, how an algorithm could exploit predictions of track state to manage an agile radar system to improve signature-aided tracking performance.
* information listed above is at the time of submission.