Spectrometers based on acousto-optic tunable filters (AOTFs) have several advantages, such as stable temperature adaptability, no moving parts, and wavelength selection through electrical modulation, compared with the traditional grating and Fourier transform spectrometers. Therefore, AOTF spectrometers can realize stable in situ measurement on the lunar surface under wide temperature ranges and low light environments. AOTF imaging spectrometers were first employed for in situ measurement of the lunar surface in the Chinese Chang’e project. The visible and near-infrared imaging spectrometer and the lunar mineralogical spectrometer have been successfully deployed on board the Chang’e-3/4 and Chang’e-5 missions. In this review, we investigate the performance indicators, structural design, selected AOTF performance parameters, data acquisition of the three lunar in situ spectral instruments used in the Chang’e missions. In addition, we also show the scientific achievement of lunar technology based on in situ spectral data.

Near-infrared spectroscopy has been widely applied in various fields such as food analysis and agricultural testing. However, the conventional method of scanning the full spectrum of the sample and then invoking the model to analyze and predict results has a large amount of collected data, redundant information, slow acquisition speed, and high model complexity. This paper proposes a feature wavelength selection approach based on acousto-optical tunable filter (AOTF) spectroscopy and automatic machine learning (AutoML). Based on the programmable selection of sub nm center wavelengths achieved by the AOTF, it is capable of rapid acquisition of combinations of feature wavelengths of samples selected using AutoML algorithms, enabling the rapid output of target substance detection results in the field. The experimental setup was designed and application validation experiments were carried out to verify that the method could significantly reduce the number of NIR sampling points, increase the sampling speed, and improve the accuracy and predictability of NIR data models while simplifying the modelling process and broadening the application scenarios.