<p>Mining potentially meaningful movement patterns of moving objects is of great significance for analyzing human movement law, understanding traffic mechanisms, and analyzing wild animal migration activities. In recent years, the growing availability of movement data from sensor networks and location-aware devices has provided great opportunities for investigating the spatiotemporal patterns of moving objects. A systematic review of this rapidly evolving field is necessary to synthesize existing knowledge and direct future research. In this paper, we systematically review the classification and mining methods of moving objects’ movement patterns, identify eight major categories of movement patterns, and systematically compare existing pattern discovery methods, including classical spatial data mining techniques and emerging deep learning models. Based on this, we summarize the key challenges in current research, e.g., scale-adaptive pattern mining, real-time detection, co-mining of multiple patterns, and reproducibility. Furthermore, we outline promising future research directions, such as developing general pattern mining algorithms, integrating deep learning with traditional methods, and incorporating semantic context into pattern interpretation. This study can provide a reference for subsequent research and advance the further development of this field, especially promoting the expansion of current spatial data mining methods for static data to spatiotemporal patterns mining for dynamic streaming data oriented to moving objects.</p>