Reaction-boned silicon carbide (RB-SiC) is considered a new material for large lightweight ground-based space telescopes due to its high specific stiffness, low thermal deformation, and excellent optical quality. The excellent mechanical properties of RB-SiC result in the low efficiency of traditional polishing and mechanical polishing. In this paper, a polishing method for RB-SiC based on a femtosecond laser is proposed to improve surface quality. A theoretical heat conduction model was established in the process of femtosecond laser irradiation of SiC. We analyzed the ablation type and calculated the single-pulse ablation threshold of SiC, which verified the feasibility of femtosecond laser polishing. Further, the effects of polishing parameters on the polished surface quality were analyzed by a series of experiments, and the optimal parameters were selected. It was observed to improve polishing efficiency and can replace the intermediate steps of traditional mechanical polishing.

With advances in scientific foundations and engineering practice, segmented mirrors—a key architecture for realizing extremely large apertures and high-resolution imaging—have become foundational across space astronomy, ground-based telescopes, and advanced manufacturing. In recent years, interferometry, which leverages optical coherence and phase sensitivity, has become a powerful tool for inter-segment co-phasing. Its capabilities have advanced markedly owing to developments in multi-wavelength techniques, high-speed high-dynamic-range detectors, and instantaneous phase-shifting methods. Relative to non-interferometric sensing, interferometry directly encodes and unwraps phase. This enables a unified framework that combines millimeter-scale dynamic range with nanometer-level resolution throughout coarse acquisition, fine phasing, and in situ maintenance. This paper first outlines the degrees of freedom and error sources in segmented mirrors. It then reviews the configurations and acquisition strategies of shearing, Mach–Zehnder, Michelson, Fizeau, and PISTIL interferometers, and systematizes interferogram processing methods—such as phase-shifting, synthetic-wavelength techniques, and digital holography—for retrieving piston and tip/tilt. Accuracy of piston is λ/50–λ/100, and tip/tilt accuracy can reach the arcsecond level, with resolution at the nanometer scale. Finally, we discuss pathways to extend interferometric metrology from segmented mirrors to other discontinuous surfaces (e.g., segmented detectors, segmented gratings, microlens arrays) and outlines future research directions.

Under ground-based observation geometry, infrared radiant-intensity sequences of space objects are jointly influenced by object micromotion, thermal radiation, time-varying viewing conditions, and atmospheric propagation. Existing simulation studies often prescribe the line of sight or simplify the coupling between viewing geometry and atmospheric attenuation, which limits long-duration ground-based sequence analysis. This study develops a physics-informed framework for generating atmosphere-attenuated infrared radiant-intensity sequences of space objects undergoing precession or tumbling. The framework reconstructs observation geometry from azimuth–elevation–range trajectories, updates facet normals through a unified micromotion attitude model, computes visible projected area and transient facet temperature, and incorporates MODTRAN-derived elevation-dependent atmospheric transmittance. Using this framework, we construct IRPeriodic, an eight-class simulated dataset for long-duration univariate time-series classification. We further propose LPD-Net, which integrates large-kernel residual feature extraction, prototype-guided dynamic temporal alignment, and differential periodic representation to capture long-range waveform morphology, sample-dependent temporal correspondence, and segment-level local variation. On IRPeriodic, LPD-Net achieves an accuracy of 0.8618 ± 0.0057, a macro-F1 of 0.8615 ± 0.0061, and a Matthews correlation coefficient of 0.8426 ± 0.0065, outperforming the evaluated neural-network and ROCKET-type baselines. Ablation and synthetic-noise sensitivity analyses indicate that the performance gain is mainly associated with long-context feature extraction, with additional improvements from dynamic alignment and differential periodic statistics. Auxiliary experiments on selected public UCR datasets suggest that the representation is also competitive for univariate time-series classification. These results demonstrate the effectiveness of LPD-Net on the proposed physics-informed benchmark for long-duration ground-based infrared radiant-intensity sequence classification.