Abstract: Tool wear condition monitoring is one of the key technologies in the machining industry to ensure machining quality, improve production efficiency and ensure the safe operation of equipment. Tool wear condition monitoring is of great significance to improve the machining quality, service life and productivity of tools. In this paper, based on sensor selection, feature extraction, machine learning and deep learning, the research progress of tool wear monitoring in the cutting process is systematically introduced. The selection of wear signals, the common methods of feature extraction in the time-domain, frequency domain and time-frequency domain, and the application of machine learning and deep learning in tool wear condition monitoring are focused on in this paper. On this basis, the research difficulties of the monitoring of the wear state of cutting tools are analyzed, and the existing problems are summarized. Finally, the future development trend of tool wear monitoring technology in cutting is prospected.