A machine learning approach to crater detection from topographic data

Craters are distinctive features on the surfaces of most terrestrial planets. Craters reveal the relative ages of surface units and provide information on surface geology. Extracting craters is one of the fundamental tasks in planetary research. Although many automated crater detection algorithms have been developed to exact craters from image or topographic data, most of them are applicable only in particular regions, and only a few can be widely used, especially in complex surface settings. In this study, we present a machine learning approach to crater detection from topographic data. This approach includes two steps: detecting square regions which contain one crater with the use of a boosting algorithm and delineating the rims of the crater in each square region by local terrain analysis and circular Hough transform. A new variant of Haar-like features (scaled Haar-like features) is proposed and combined with traditional Haar-like features and local binary pattern features to enhance the performance of the classifier. Experimental results with the use of Mars topographic data demonstrate that the developed approach can significantly decrease the false positive detection rate while maintaining a relatively high true positive detection rate even in challenging sites.     

Solar magnetic feature detection and tracking for space weather monitoring

We present an automated system for detecting, tracking, and cataloging emerging active regions throughout their evolution and decay using SOHO Michelson Doppler Interferometer (MDI) magnetograms. The SolarMonitor Active Region Tracking (SMART) algorithm relies on consecutive image differencing to remove both quiet-Sun and transient magnetic features, and region-growing techniques to group flux concentrations into classifiable features. We determine magnetic properties such as region size, total flux, flux imbalance, flux emergence rate, Schrijver’s R-value, R^∗ (a modified version of R), and Falconer’s measurement of non-potentiality. A persistence algorithm is used to associate developed active regions with emerging flux regions in previous measurements, and to track regions beyond the limb through multiple solar rotations. We find that the total number and area of magnetic regions on disk vary with the sunspot cycle. While sunspot numbers are a proxy to the solar magnetic field, SMART offers a direct diagnostic of the surface magnetic field and its variation over timescale of hours to years. SMART will form the basis of the active region extraction and tracking algorithm for the Heliophysics Integrated Observatory (HELIO).     

Lorentz Symmetry breaking studies with photons from astrophysical observations

Lorentz Invariance Violation (LIV) may be a good observational window on Quantum Gravity physics. Within last few years, all major gamma-ray experiments have published results from the search for LIV with variable astrophysical sources: gamma-ray bursts with detectors on-board satellites and Active Galactic Nuclei with ground-based experiments. In this paper, the recent time-of-flight studies with unpolarized photons published from the space and ground based observations are reviewed. Various methods used in the time delay searches are described, and their performance discussed. Since no significant time-lag value was found within experimental precision of the measurements, the present results consist of 95% confidence level limits on the Quantum Gravity scale on the linear and quadratic terms in the standard photon dispersion relations.     

Active collision avoidance maneuver under constant thrust

In this paper, the relative position parameters of the target spacecraft are obtained by using the vision measurement and the target maneuver positions are calculated through the isochronous interpolation method. Furthermore, new switch control laws under constant thrust are designed for active collision avoidance maneuver of the chaser along a specified trajectory. The switch control laws are obtained based on the acceleration sequences and the working times of thrusters in three axes which can be respectively computed by the time series analysis method. The perturbations and fuel consumptions are addressed during the computation of the working times.     

翼型气动性能鲁棒性优化设计

讨论了一种在来流速度不确定的情况下对翼型进行鲁棒性设计的方法.介绍了单点及多点设计方法后,阐述了鲁棒性设计方法在翼型气动性能优化中的应用,并对三者进行了比较.引入了代理模型以减少计算量, 并通过遗传算法对翼型进行鲁棒性设计.借助区间分析方法讨论了翼型设计变量的扰动对气动特性上下界值的影响.应用该方法,在提高了翼型的气动性能的同时,降低了该性能对于来流速度的敏感度.      

激光选区熔化增材制造缺陷智能监测与过程控制综述

激光选区熔化（SLM）技术被认为是最有应用前景的增材制造技术之一,已应用于航空航天、医疗器械等领域。然而,如何确保构件质量的可靠性和制造的可重复性是SLM面临的最大挑战,已被认为是限制SLM及其他金属增材制造技术发展和工业应用的最大壁垒。其中,主要原因是SLM过程中会产生难以控制的缺陷。因此,对SLM进行过程监测和实时反馈控制是解决这一挑战的重要研究方向,也已成为学术界和工业界的研究热点之一。通过对近十年该领域的文献调研,综述了金属激光增材制造中常见的冶金缺陷及其产生机理,对金属增材制造过程产生的信号及其监测手段,如声信号、光信号及热信号等进行了详细描述;总结了信号数据的处理方法,包括传统的统计处理方法和新兴的基于机器学习的智能监测方法;随后,综述了金属增材制造过程的质量控制方法,包括非闭环控制和闭环控制,并对全文进行了总结,展望了未来SLM智能监测和控制领域值得深入的研究方向。     

人工智能可解释性评估研究综述CSCD

近年来,可解释人工智能(XAI)发展迅速,成为当前人工智能领域的研究热点,已出现多种人工智能解释方法。如何量化评估XAI的可解释性以及解释方法的效果,对研究XAI具有重要意义。XAI的可解释性评估涉及主、客观因素,是一个复杂且有挑战性的工作。综述了XAI的可解释性评估方法,首先,介绍了XAI的可解释性及其评估的概念和分类;其次,总结和梳理了一些可解释性的特性;在此基础上,从可解释性评估方法和可解释性评估框架两方面,综述和分析了当前可解释性评估工作;最后,总结了当前人工智能可解释性评估研究的不足,并展望了其未来发展方向。     

Aerodynamic characteristics of co-flow jet wing with simple high-lift devices

Numerical investigations are conducted to explore the aerodynamic characteristics of three-dimensional Co-Flow Jet(CFJ) wing with simple high-lift devices during low-speed takeoff and landing. Effects of three crucial parameters of CFJ wing, i.e., angle of attack, jet momentum and swept angle, are comprehensively examined. Additionally, the aerodynamic characteristics of two CFJ configurations, i.e., using open and discrete slots for injection, are compared. The results show that applying CFJ te...     

基于流体力学和电磁学方程数值求解的飞行器气动隐身一体化设计

介绍了基于流体力学和电磁学方程数值求解的飞行器气动隐身一体化设计方法.首先介绍了精度相对较高的飞行器气动和隐身特性数值计算方法,即,对于气动性能计算,求解的是结构网格上的NS方程加BL代数湍流模式;对于隐身特性计算,是用时域有限体积法来求解电磁学微分方程以获取RCS值.由于采用了高精度的数值方法,优化时单一设计点的气动性能计算和隐身性能计算变得较为耗时,因此在进行多目标遗传算法优化时本文采用了一种"少量样本计算+Kriging响应面模型建模"的优化策略.针对某类似X-47飞行器的一体化设计算例计算表明,上述设计方法是可行的,实现了优化设计中引入高精度的性能分析方法,有望提高优化结果的可信度.     

航天器姿控系统的PD型学习观测器故障重构

针对满足Lipschitz条件的航天器姿态控制系统这一非线性系统中存在的执行器加性故障、空间干扰与测量噪声问题,提出了基于PD型迭代学习观测器的故障重构方法。该方法具有期望的鲁棒性能指标,能够在系统存在空间干扰与测量噪声情况下实现对突变故障与时变故障等故障类型的精确重构。基于线性矩阵不等式技术给出系统化PD型迭代学习观测器的设计方法,并根据Lyapunov稳定性理论对上述设计方法的稳定性条件进行了理论证明,同时利用鲁棒技术抑制空间干扰与测量噪声对执行器故障重构的影响,通过线性矩阵不等式工具箱求解观测器参数矩阵。最后,将该方法应用到航天器姿态控制系统中,仿真结果证明了该方法的有效性。