Efficiently maintaining moving micro clusters for clustering moving objects

Clustering moving objects is a challenging task, especially when space consumption must be flexibly and efficiently adjusted based on dynamic object movement. In this paper we develop an efficient approach to dynamically maintain a small set of moving micro clusters (MMCs) to represent groups of similar moving objects. Global clusters can then be generated by any clustering algorithm, whenever is needed, from these representative MMCs, not from individual objects. Under our approach, each MMC is represented by a vector that summarizes the position and velocity information of its member objects. Based on this summarized information, a set of simple formulas is developed to efficiently predict when the contents of MMCs must be updated. MMCs are also organized into a moving cluster feature (MCF) tree so they can be efficiently merged for conserving space and accelerating global clustering.