Maneuvering target tracking using multiple maneuver model joint probabilistic data association

摘要

The JDPA (joint probabilistic data association) algorithm has been reported to be suitable for tracking multiple targets in the presence of extraneous observations (i.e., clutter, thermal false alarms, other targets) and no reports (missed target detections). In the present work, a novel method for highly maneuvering target tracking is presented which is an extension of the JPDA. Computer simulation indicated that this method can track a highly maneuvering target (7 g) in the presence of extremely dense clutter (with a probability of a false alarm of 0.1) with a sampling interval of 0.5 s.