Aiming at the problem that it is difficult to accurately track the spawning cluster targets, this paper derives a multi-spawn GGIW-PHD filtering algorithm considering spawning behavior. Firstly, individuals in cluster are modeled through the cluster behavior model, and the cluster target is modeled as an extended target in this paper. The multi-spawn targets parameter model is established under the framework of the basic Gamma Gaussian inverse Wishart probability hypothesis density (GGIW-PHD) filter. Secondly, adopting the coordinate difference measurement partitioning method that is more suitable for the characteristics of the cluster target, which partly reduces the time complexity of the filtering algorithm. Simulation experiments show that the proposed algorithm can effectively track cluster targets, especially when spawning occurs, the filter can quickly adapt to the generated subgroup targets, and has smaller estimation errors for subgroup targets center state and extended state.