9 December 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Time-series InSAR is an advanced technology for real-time monitoring and precise analysis of object deformation. However, large-span steel truss structure bridges have complex bridge structures and deformation forms, which make it difficult for the traditional persistent scatterer interferometry (PSI) method to realize the precise detection of their deformation, resulting in a series of limitations. In order to solve these problems, an improved PSI method is proposed in this article, which is first to carry out preliminary elevation measurement by a small temporal baseline, use the preliminary elevation results to PS-arcs construction based on 3-D spatial network, and utilize the optimal model coherence coefficient adaptive selection based on constant false alarm rate to carry out the parameter estimation so as to realize the accurate monitoring of deformation of large-span steel truss structure bridges. In this article, 24 ascending CSK stripmaps are used to monitor the deformation of Dongting Lake Bridge of the Menghua Railway, and the monitoring results show that the method not only realizes the accurate detection of deformation of large-span steel truss structure bridges but also improves the accuracy and robustness of parameter estimation. The results show that the proposed method improves the accuracy of deformation parameter estimation by comparing with the deformation parameters solved by the traditional method, and the method reduces the average residual phase standard deviation by 10%, which provides the experience and reference for the deformation monitoring of similar bridges.
28 July 2022 in SCIENTIA SINICA Technologica
在月面采样任务中,受土壤反作用力、能耗等因素影响,机械臂采样位形会影响其采样性能。为了准确评估采样位形对采样性能的影响,提高采样任务执行的可靠性,提出了机械臂采样位形分析方法与优选策略。首先,分析机械臂与采样器、采样器与铲挖执行机构之间的耦合关系,建立机械臂与采样器的联合动力学模型;其次,分析采样过程中机械臂位形与土壤反作用力、机械臂末端抗偏移能力、关节运动行程以及运动能耗之间的关系,并构建量化指标对采样性能进行评估;进而,根据构建的采样性能评价指标,构建考虑采样点区域、关节轨迹等多种约束的多目标最优化问题框架,设计基于NSGA-2多目标遗传算法的采样位形优选策略;最后,以嫦娥五号采样机械臂为对象进行实验分析。实验结果验证了本文提出的采样位形分析方法与优选策略的有效性,相关结论也为嫦娥五号采样机械臂实际采样指令的生成提供了参考依据。
1 January 2025 in Electronics Letters
Accurate ground deformation monitoring is crucial for understanding geophysical activities and mitigating disaster risks. D‐InSAR and offset tracking are widely used but face limitations: D‐InSAR struggles with phase unwrapping errors in large‐scale deformation regions, while offset tracking lacks precision in small‐scale deformation. This paper proposes a fusion algorithm based on variance‐masked mean filtering to combine the strengths of both methods. The algorithm filters noise in gentle regions of the offset map and fuses refined results with D‐InSAR to enhance accuracy. Applied to the 2023 Türkiye earthquake, the method effectively reduces noise, corrects unwrapping errors, and achieves superior accuracy in large‐scale deformation scenarios such as earthquakes and landslides.
22 April 2024 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Lunar tidal deformation (LTD), characterized by the vertical ( h 2 ) and horizontal ( l 2 ) displacement Love numbers, is a key to deciphering the interior structure of the Moon. However, the small deformation amplitude of only up to 10 cm makes their measurements very challenging. In this paper, we propose a novel method named Earth-based repeat-pass SAR interferometry (EBRP-InSAR) to measure the LTD and displacement Love numbers. We analyzed the potential and performance of EBRP-InSAR in detail based on LTD model and existing planetary radar capabilities. The error budget and simulation results show that the relative measurement accuracy of LTD could be better than 2 mm. Furthermore, compared with Lunar Laser Ranging (LLR) and Lunar Orbiter Laser Altimeter (LOLA), the two-pass EBRP-InSAR can not only directly estimate h 2 but also l 2 with accuracies better than 10 -3 and 10 -4 , respectively, which is comparable that of LLR and LOLA methods. In addition, through long-term time-series observations, it will be possible to assess the spatial inhomogeneity of LTD response to the forcing potential in the near side with time-series EBRP-InSAR. After spatial smoothing, a space variation as small as on the order of 10 -4 and 10 -5 in h 2 and l 2 , respectively, can be distinguished on the lunar near side. As another measurement technology independent of LLR and LOLA, the EBRP-InSAR is expected to explain the difference between the observations from LLR and LOLA and the modeled lunar interior structure, and firstly estimate the lunar horizontal displacement Love number.
13 June 2025 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
River level monitoring is crucial for hydrological studies, providing essential information for flood forecasting, water resource management, and environmental protection. In this paper, we present a novel method for monitoring river level under bridges using time-series SAR images. Firstly, we transfer a DeepLab V3+ network model for road segmentation to bridge segmentation, fine-tuning it with bridge scattering signal data, while a new loss supervision function CentroidLoss, has been added to the model to improve the integrity of the bridge signal segmentation. Furthermore, the Energy Accumulation Algorithm (EAA) is proposed to improve the accuracy of river level measurements in areas of low signal-to-noise ratio with noise such as ships and waves. Leveraging deep learning and energy accumulation algorithm, the proposed approach accurately extracts bridge scattering signals and precisely estimates the peak positions of the bridge's multiple scattering signals, enabling precise river level monitoring. Sentinel-1A and COSMO-SkyMed data were applied as inputs to our method, and the comparison between the river levels measured by the proposed method and those of local hydrological stations reveals sub-meter level estimation accuracy.
9 March 2026 in IEEE Geoscience and Remote Sensing Letters
Tidal deformation offers critical insights into the internal structure of Near-Earth Asteroids (NEAs). This study leverages the 2029 Earth flyby of asteroid 99942 Apophis, which will approach within 31,000 km. Unlike typical observations constrained by low Signal-to-Noise Ratio (SNR), this encounter allows ground-based radar to support high-precision measurements. We propose a novel physics-signal fusion framework that integrates a dynamic tidal deformation model with the Hybrid Generalized Radon-Fourier Transform (HGRFT) to estimate key structural parameters. Simulation experiments demonstrate that the proposed method maintains high robustness, and the relative error of dynamic viscosity estimation remains within 1% under a low-SNR condition. The approach holds significant implications for future asteroid missions and provides a viable pathway for the characterization of hazardous asteroids during close encounters.