IMM-JVC and IMM-JPDA for closely maneuvering targets

Efficient and reliable multiple target tracking methods require the use of sophisticated data association as well as positional estimation algorithms. Two different approaches can be employed in the development of an assignment strategy for data association. This paper presents a comparative study of two assignment alternatives, namely the JVC (unique association of a measurement to an existing track) and JPDA (nonunique association of a measurement to an existing track) algorithms. The above assignment strategies are then both combined with an interacting multiple model (IMM) positional estimator, and the respective tracking performance of the IMM-JVC and IMM-JPDA algorithms is evaluated using two critical scenarios involving a pair of closely maneuvering targets.. An analysis of the simulation results is carried out, permitting for a comparison of the JVC and JPDA association algorithms, which ultimately demonstrates the superiority of the IMM-JVC algorithm.