Unmanned Aerial Vehicle (UAV) target tracking is one of the key technologies in aerial intelligent perception systems, playing a vital role in applications such as traffic monitoring, border patrol, disaster response, search and rescue, environmental monitoring, and military reconnaissance. Compared with generic object tracking tasks, UAV platforms exhibit significant differences in imaging perspectives, target scales, motion patterns, and onboard computing capabilities, which pose unique challenges for UAV target tracking, including small targets and drastic scale variations, platform motion and motion blur, complex backgrounds and frequent occlusions, low-light conditions at night, as well as real-time and energy constraints. To address these issues, a large number of UAV-oriented tracking methods have been proposed in recent years, covering traditional correlation filters, deep Siamese networks, and emerging Transformer-based models, achieving continuous performance improvements across multiple UAV benchmark datasets. Despite substantial research efforts, existing survey works primarily focus on generic object tracking or a single technical approach, lacking a systematic summary and comparative analysis from the perspective of UAV application requirements. Unlike previous surveys that mainly classify methods based on model architecture, the innovation of this study lies in establishing a unified UAV target tracking framework centered on five major challenges. First, we analyze typical UAV tracking applications and core challenges. Then, from a challenge-driven perspective, existing methods are categorized and summarized based on small targets and scale variations, rapid motion and motion blur, complex backgrounds and occlusions, low-light night conditions, and lightweight and real-time considerations. Furthermore, we conduct quantitative and qualitative comparisons of representative methods in terms of accuracy, success rate, and computational efficiency on mainstream benchmarks, including UAV123, UAV123@10fps, UAVDT, UAV20L, and DTB70. Finally, we looked ahead to future development directions, such as lightweight deployment, multi-modal fusion, and large model-driven approaches. This work aims to provide a clear technical roadmap and a systematic reference for UAV target tracking research.

3D single object tracking (SOT) based on point clouds is a fundamental task for environmental perception in autonomous driving and dynamic scene understanding in robotics. Recent technological advancements in this field have significantly bolstered the environmental interaction capabilities of intelligent systems. This field faces persistent challenges, including feature degradation induced by point cloud sparsity, representation drift caused by non-rigid deformation, and occlusion in complex scenarios. Traditional appearance matching methods, particularly those relying on Siamese networks, are severely constrained by point cloud characteristics, often failing under rapid motions or structural ambiguities among similar objects. In response, the research paradigm has progressively evolved toward motion-centric modeling approaches. These emerging frameworks utilize spatio-temporal joint modeling and geometric shape completion to attain notable performance gains. Furthermore, the incorporation of attention mechanisms and State Space Model (SSM) has enabled more effective multi-scale spatio-temporal feature association, which is particularly beneficial for long-term tracking scenarios. To the best of our knowledge, this is the first comprehensive survey dedicated to 3D single object tracking in point clouds. We provide a detailed analysis of current tracking methods, scrutinizing their limitations regarding multi-object interference and analyzing the trade-off between accuracy and computational efficiency. Finally, we discuss potential future directions, including the development of lightweight models for edge deployment and the integration of cross-modal fusion strategies.