Research on mirror shape testing technology based on deep learning

Research on reinforcement learning multi degree of freedom control system for shape testing of space camera mirror. Based on the optimization design of the perception control execution closed-loop intelligent testing system, reinforcement learning pose joint control training and testing are carried out based on the position information of the laser tracker and the misalignment information data of the laser interferometer. Complete automated testing experiments and validation. Improve the accuracy and efficiency of the testing process, reduce labor costs, and enhance the automation level of space camera installation and inspection.