数字图像散斑相关技术的蚁群优化方法

基于蚁群优化方法提出新的数字图像散斑相关算法。该方法模仿了真实蚂蚁从其巢到食物找到最短路径的方式,通过对蚁群优化方法改进,减少迭代次数并改善解的质量。将新的数字图像散斑相关算法应用到计算机模拟的散斑图像和实验获得的散斑图像中,并与广泛使用的Newton-Raphson算法进行了比较。实验结果展示了新算法的精度、可行性和有效性。当数量级为0.01像素,误差离散均方根小于0.002像素。     

COMPOSITE-STEP LIKE FILTER METHODS FOR EQUALITY CONSTRAINT PROBLEMS

In a composite-step approach, a step sk is computed as the sum of two components vk and hk. The normal component vk, which is called the vertical step, aims to improve the linearized feasibility, while the tangential component hk, which is also called horizontal step, concentrates on reducing a model of the merit functions. As a filter method, it reduces both the infeasibility and the objective function. This is the same property of these two methods. In this paper, one concerns the composite-step like filter approach. That is, a step is tangential component hk if the infeasibility is reduced. Or else, sk is a compositestep composed of normal component Vk and tangential component hk.     

An explicit formulation for multiscale modeling of bcc metals

Many materials for specialized applications exhibit a body-centered cubic structure; e.g., tantalum, vanadium, barium and chromium. In addition, the successful modeling of body-centered cubic (bcc) metals is a necessary step toward modeling of common structural materials such as iron. Implicit formulations for this class of materials exist [e.g., Stainier, L., Cuitiño, A., Ortiz, M., 2002. A micromechanical model of hardening, rate sensitivity, and thermal softening in bcc crystals. Journal of the Mechanics and Physics of Solids 50 (7), 1511–1545; Kuchnicki, S., Radovitzky, R., Cuitiño, A., Strachan, A., Ortiz, M., 2007. A pressure-dependent multiscale model for bcc metals], but are impractical to resolve large-scale dynamic deformation processes. In this article, we describe a procedure analogous to Kuchnicki et al. [Kuchnicki, S., Cuitiño, A., Radovitzky, R., 2006. Efficient and robust constitutive integrators for single-crystal plasticity modeling. International Journal of Plasticity 22 (10), 1988–2011]. wherein we construct an explicit formulation for the multiscale physics models. This update is based on the model of Kuchnicki et al. (in preparation) using a power law representation for the plastic slip rates. The existing implicit form of the model provides qualitative matching with experiments at quasi-static strain rates. The model is recast in an explicit form and applied first to a high quasi-static strain rate to verify that the two forms of the model return similar predictions for similar input parameters. The explicit model is also applied to several high strain rates, showing that it captures characteristic features observed in experimental tests of high-rate deformations, such as the drop in stress immediately after yield that is present in split Hopkinson pressure bar (SHPB) experiments. This test provides qualitative evidence that the model is suitable for high-strain-rate applications. The utility of the model is further demonstrated by a one-dimensional simulation of a SHPB test. Finally, a test case modeling pressure impact of a Tantalum plate using 600,000 elements is shown. The simulations show that the explicit model is capable of recovering the salient features of the experiments while integrating the constitutive update in a robust manner.     

Computation of Kullback–Leibler Divergence in Bayesian Networks

Kullback–Leibler divergence KL(p,q) is the standard measure of error when we have a true probability distribution p which is approximate with probability distribution q. Its efficient computation is essential in many tasks, as in approximate computation or as a measure of error when learning a probability. In high dimensional probabilities, as the ones associated with Bayesian networks, a direct computation can be unfeasible. This paper considers the case of efficiently computing the Kullback–Leibler divergence of two probability distributions, each one of them coming from a different Bayesian network, which might have different structures. The paper is based on an auxiliary deletion algorithm to compute the necessary marginal distributions, but using a cache of operations with potentials in order to reuse past computations whenever they are necessary. The algorithms are tested with Bayesian networks from the bnlearn repository. Computer code in Python is provided taking as basis pgmpy, a library for working with probabilistic graphical models.     

多个磁流变弹性体自调谐式吸振器的联合控制研究

遗传算法是一种自适应、启发式、全局优化的搜索算法,在结构振动控制等复杂控制领域得到了广泛的应用,本文针对磁流变弹性体自调谐式吸振器难以建立准确的控制模型的特点,通过改进基本遗传算法,设计和实现了一种同时控制多个吸振器的联合控制算法,该方法在协调控制时,根据多个吸振器协调控制对控制量的要求采用指定位交叉产生新个体,与通过随机位交叉产生的新个体相比,这样能以较大的概率产生更健壮的新个体,从而使算法更快收敛.将该算法应用到两个吸振器联合控制的实验系统中,在吸振器移频范围内,被减振系统各观测点均有较好的减振效果,当吸振器与主系统质量比为1∶19.8时,减振效果最高约12dB,主被动减振效果之差最佳时达7dB.     

无线传感器网络多跳路由协议研究

目的：无线传感器网络发展迅速,但传感器的高能耗问题成为制约其发展的主要瓶颈,高效节能的路由协议设计成为研究热点。方法：针对目前无线传感器网络常用的LEACH路由协议存在的簇首能耗过分集中、簇首分布不均衡问题,提出了改进的路由协议EEACRA,在总结、分析LEACH路由协议现有问题的基础上,给出了EEACRA路由协议的簇首选取门限值、簇首位置调整算法和基于能量代价最小的簇间多跳路由算法的实现方法,同时给出了具体的实现EEACRA协议的工作流程和关键算法。在MATLAB环境下对LEACH路由协议和EEACRA路由协议进行了仿真,对比了不同能耗降低措施对网络能耗降低的贡献。结果：仿真结果表明EEACRA路由协议的网络稳定期较LEACH路由协议有较大的改善。结论：证明了改进的路由协议EEACRA可以有效地提高网络的稳定期。     

无人机地面运动控制与仿真

为了解决无人机纠偏刹车系统中实际存在的问题,对无人机地面运动进行了研究,采用前轮转弯、方向舵、差动刹车联合使用模式,以无人机的侧偏角、侧偏距为反馈信号,设计了无人机地面运动纠偏控制律,通过遗传算法的全局优化过程实现其纠偏参数寻优,在Simulink平台建立了无人机纠偏数学模型;试验结果表明,建立的无人机地面运动纠偏数学模型正确有效,能够模拟出无人机地面运动的真实情况;基于遗传算法的控制策略能够实现一定范围内的无人机地面运动纠偏,提高跑道的利用率。     

Unsupervised classification of slip events for planetary exploration rovers

This paper introduces an unsupervised method for the classification of discrete rovers’ slip events based on proprioceptive signals. In particular, the method is able to automatically discover and track various degrees of slip (i.e. low slip, moderate slip, high slip). The proposed method is based on aggregating the data over time, since high level concepts, such as high and low slip, are concepts that are dependent on longer time perspectives. Different features and subsets of the data have been identified leading to a proper clustering, interpreting those clusters as initial models of the prospective concepts. Bayesian tracking has been used in order to continuously improve the parameters of these models, based on the new data. Two real datasets are used to validate the proposed approach in comparison to other known unsupervised and supervised machine learning methods. The first dataset is collected by a single-wheel testbed available at MIT. The second dataset was collected by means of a planetary exploration rover in real off-road conditions. Experiments prove that the proposed method is more accurate (up to 86% of accuracy vs. 80% for K-means) in discovering various levels of slip while being fully unsupervised (no need for hand-labeled data for training).     

应用近红外光谱和多算法融合方法分析烤烟的香型风格特征

以全国17个主要烤烟产地省份中收集的3 914个烟叶样品的近红外光谱为实验对象,其中浓香型、中间香型、清香型烟叶光谱分别865条、1 403条、1 646条,应用近红外光谱和多算法融合方法分析其香型风格特征。在以产地进行初步划分烟叶香型以及认可过渡型和非典型香型类型的基础上,选取基于主成分及Fisher准则的投影法(PPF)、偏最小二乘判别(DPLS)、支持向量机(SVM)作为单分类器,得到各个算法第1和2判别分析结果;应用PPF-DPLS-SVM融合和各算法第1和2判别分析结果,将预测验证样品的分析结果详细划分为典型、过渡型、非典型香型样品(分别为493,392,115个);其中典型香型烟叶样品的判别准确率达到92.7%,较未进行典型样品划分时PPF,DPLS,SVM单算法的识别准确率分别提高了30.2%,15.4%,16.6%。样品数据来源于全国主要烤烟产地,数据量大,代表性较好,分析结果具有一定普遍性;提出的多算法融合分析方法大幅度提高了通过客观数据判别烤烟香型的准确率;同时,将烤烟香型细划分为典型、过渡型和非典型香型的方式,对烤烟烟叶原料的科学合理利用以及烟叶原料的模块化工业加工等有指导作用。     

Characteristics of Good Meta-Heuristic Algorithms for the Frequency Assignment Problem

Most writers on frequency assignment algorithms have described the details of a single algorithm, and evaluated the algorithm on selected data sets. There has been relatively little emphasis on describing the common features that are important if an algorithm is to have good performance. This paper describes the key features, with particular emphasis on algorithms for weighted fixed spectrum problems. The use of algorithms handling weighted constraints has become increasingly common in recent years. The advantages and disadvantages of weighting constraints are demonstrated.