Feature-Based Angle-Only Measurement to Radar Track Association
Year: 2025
Type: article
Abstract: Angle-only tracking of airborne targets is challenging due to the ambiguity caused by the absence of a measure for range. If a colocated radar and angle-only sensor work collaboratively, it is possible to associate measurements and tracks using not only kinematics, but also radio frequency (RF) derived features. Combining kinematic distances and RF features in a probabilistic framework enables the stitching together of disparate measures in a coherent manner. In this article, we quantitatively formulate a probabilistic RF feature-based kinematic association technique which improves performance over kinematic association alone. We demonstrate a significant increase in accuracy using this RF feature-based kinematic association technique in challenging radar tracking scenarios with closely spaced targets.
Authors: Joel Goodman, Kevin S. Lorenz, Kevin Wagner, Codie Lewis
Institution: United States Naval Research Laboratory
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,645
OpenAlex ID: W4407449076