Computation of Target-Measurement Association Probabilities Using the Matrix Permanent

摘要

The computation of measurement (or missed detection) to target-association probabilities is an essential part of many tracking algorithms. This paper generalizes previous work to show the relationship between measurement assignment and the matrix permanent given the possibility of missed detections and clutter. A simple expression is given for the target-measurement association probabilities in terms of the matrix permanent and efficient algorithms for computing the matrix permanent are summarized.