A GGIW-PHD Filter for Multiple Non-Ellipsoidal Extended Targets Tracking With Varying Number of Sub-Objects

When the extension state of non-ellipsoidal extended target (NET) changes, performance traditional multiple tracking algorithms based on constant number sub-objects will decrease. To solve this problem, paper proposes a gamma Gaussian inverse Wishart probability hypothesis density filter for targets with varying sub-objects, called VN-NET-GGIW-PHD filter. In proposed filter, each NET is considered as combination spatially close and label management introduced to realize association between corresponding sub-objects. Then, by spawning combination, approximating can be adjusted automatically. Furthermore, obtain partition measurement set, an approach clustering fast search find peaks (CFSFDP) algorithm expectation maximization (EM) proposed. Simulation results show that adaptively adjust has better when changes.