As UAV technology advances at a fast pace, its ability to track targets has become increasingly valuable across diverse domains. This study begins by highlighting the significance of UAV target tracking and the obstacles it encounters, including shifts in target appearance, disturbances from complex environments, and variations in lighting. Subsequently, it conducts a detailed analysis of traditional target tracking algorithms, those based on correlation filtering, deep learning-driven ones, and those leveraging Transformer architectures. In addition, we will compare the advantages and disadvantages of their representative algorithms in terms of tracking accuracy, real-time performance, robustness and other related factors through examples. Finally, the commonly used UAV target tracking datasets and relevant evaluation indicators are introduced, the current challenges faced by UAV target tracking technology are summarized, and the relevant prospects for future development directions are put forward.