Multiple-Hypothesis Tracking with Unframed Sensor Measurements

As with most multi-target tracking paradigms, multiple-hypothesis tracking assumes framed sensor data. This assumption is appropriate for active radar, sonar, and imaging sensors, but is not well matched to the detection data from many passive sensors. In this paper, we derive the transformation to express unframed data in an equivalent framed-data setting, under a Poisson detection assumption. This leads to the classical MHT formulation, but with the generality of (possibly) multiple measurements per target. We explore performance of a simplified MHT solution. Finally, we explore the question of determining the optimal frame rate in the unframed-to-framed data transformation, balancing the desire for non-myopic reasoning with mitigating violations of the point-target assumption.<sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">1</sup>IEEE Aerospace Conference, Big Sky MT, March 2025. <sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup><sup xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">2</sup>IEEEAC paper #2712, Final Version, Updated 2024-11-27.