Abstract: The aging paint removal of aircraft skin is an important step in aircraft maintenance and repair. Compared with traditional chemical cleaning methods, laser cleaning has the advantages of high efficiency and low pollution, but the problem of substrate damage caused by excessive cleaning of high-energy laser exists, so it is very important to monitor the cleaning process in real time so as to give feedback to the laser in time. Aiming at this problem, a real-time monitoring method based on acousto-optic composite method is proposed, and the optical and acoustic signals are collected and analyzed respectively in the cleaning process. For the spectral detection method, the laser-induced breakdown spectrum is used to calibrate the different characteristic peaks of different elements in different paint layers to realize the inversion of different paint layers; The acoustic signal detection method is to determine the corresponding paint layer by comparing and analyzing the intensity and frequency of the acoustic signal generated by the laser ablation of different paint layers, and then the cleaning condition and removal effect of the paint layer can be deduced by comparing the two methods. The research shows that the acousto-optic composite method can achieve accurate real-time monitoring of the laser paint removal process.