With the increasing seriousness of global atmospheric environmental pollution, spaceborne lidar, as a new type of active remote sensing instrument, has become an important load for global atmospheric measurement, which can achieve high-precision measurement of atmospheric components such as greenhouse gases, particulate matter, and aerosols. Compared with the passive detection of spaceborne cameras, lidar has more stringent requirements for the stability of optical systems. The mechanical impact during the orbiting process, the change of gravity field during the orbiting operation, the change of temperature, the release of internal stress, and the jitter of the satellite platform will cause the structural deformation of the main optical machine of the lidar, thereby destroying the consistency of the optical axis of the radar transceiver and causing a decrease in detection efficiency. In addition, the change of the direction of the lidar optical axis relative to the star sensor will lead to the deviation of the radar optical axis relative to the reference measurement attitude. Therefore, high-precision transceiver optical axis monitoring and matching technology is necessary for active detection of remote sensing functions. The optical axis monitoring unit is used to monitor the real-time variation of the transceiver optical axis, which is an important part of the optical axis matching feedback mechanism.Figure 1 shows the main components of the visual axis monitoring system, including the active 785 nm reference light source, the visual axis camera, the CCD focusing lens group, the eyepiece, the receiving optical axis prism, the star sensor reference mirror, the beam combiner and the beam splitter. The active reference light source in the system uses a laser diode with a wavelength of 785 nm. After the tail fiber output laser is collimated and shaped, it first passes through the beam splitter M1. In this process, about one-tenth of the beam energy is reflected into the surveillance camera as a reference optical axis. Subsequently, the reference light is beam-divided again by 1 : 1 through the M2 mirror. This part of the light is reflected by the reference prism of the star sensor and imaged on the surveillance camera to capture the information of the star-sensitive optical axis. The light beam separated by the M2 mirror is reflected by a hollow mirror and enters the telescope system, and is reflected back to the surveillance camera through a light-taking prism located next to the secondary mirror, so that the direction change information of the received optical axis can be obtained. In addition, the emitted laser is split by a beam splitter before output, and a small amount of emitted laser enters the surveillance camera by turning and combining the beams to obtain the change information of the laser emission optical axis. The simulation results show that the imaging quality of the four optical axis monitoring channels reaches the diffraction limit level, and the design accuracy of the receiving and transmitting optical axis monitoring can reach 0.09 μrad and 2.28 μrad respectively.The experimental verification of spaceborne lidar in vacuum and space thermal environment was carried out. A test system in vacuum environment was built to calibrate the optical axis pointing of lidar. The optical axis change data measured by the optical axis monitoring unit and the tank test system were compared. The test results are shown in Fig.12. The pointing fluctuation of the transmitting optical axis measured by the optical axis monitoring unit is ± 1.14 μrad ( corresponding to a pixel jump ). At this time, the pointing jitter of the transmitting laser measured by the monitoring system in the tank is better than ± 1.5 μrad ; the pointing fluctuation of the receiving optical axis measured by the optical axis monitoring unit is ± 1.2 μrad, and the corresponding jitter measured by the tank monitoring system is ± 3.5 μrad. By comparison, it is found that both of them have the same periodic fluctuation, and the fluctuation cycle is consistent with the track thermal environment cycle. The optical axis monitoring accuracy of the transmitting optical axis is limited by the angular resolution of the channel of 2.28 μrad. There is a deviation of about 3 μrad between the visual axis system for monitoring the receiving optical axis and the in-tank test system. The reason is that the beam has a long transmission path in the in-tank test system, which is easily affected by vibration and temperature deformation during the operation of the vacuum equipment. The optical axis data after the stable operation of the lidar is analyzed. The optical axis change data is shown in Fig.13. The data analysis shows that the stability of the optical axis of the lidar is better than 3 μrad after the stable operation of the lidar in orbit.In this paper, a space-based multi-optical axis monitoring method is proposed for the on-orbit stability requirements of Atmospheric Environment Monitoring Satellite (DQ-1) Aerosol and Carbon dioxide Detection Lidar (ACDL), and a set of lidar optical axis monitoring optical machine system is designed. By using the integrated imaging scheme of active laser light source, the on-orbit synchronous high-precision monitoring of lidar transceiver optical axis under space-based coordinate reference is successfully realized. The high precision and high reliability of the monitoring technology in the space environment are verified by the vacuum thermo-optical calibration test in the ground space environment. After the lidar is launched into orbit, the optical axis monitoring unit works normally in orbit. The optical axis monitoring unit used in this paper has high environmental reliability and stability, and can be applied to other optical axis monitoring modules of active and passive space optical loads, which has important reference value and guiding significance.