三维粒子测速系统中遗传匹配法的设计

提出了基于遗传算法的三维粒子图像匹配方法及实际中的优化设计。该方法以视差为编码、结合SSD(sum of square difference)与SAD(sum of absolute difference)法对结果进行评估,并通过单点、多点的双交叉、变异与序列的混沌化处理达到对分析空间的搜索;最后,局部处理结合全局唯一性迭代检测进一步增加了匹配结果的可信度。利用优化前后所提算法对空间粒子图进行了对比及误差分析。结果表明:所提算法适用于粒子测速系统的立体匹配,能够给出较准确的视差信息。     

基于神经网络的移动机器人路径规划方法

针对动态环境下移动机器人路径规划,提出了一种基于递归神经网络的实时路径规划方法。利用神经网络表示机器人的工作空间,每个神经元都只有局部侧连接。目标点位置神经元具有全局最大的正活性值,该活性值通过神经元之间的局部侧连接逐渐衰减地传播到整个状态空间,障碍物及其周围区域神经元活性值则被抑制为零。目标点全局地吸引机器人,障碍物局部地将机器人推开实现避障,从而能够在动态环境下产生最优规划路径。仿真结果表明该方法具有较好的环境适应性和实时性。     

最短路问题的闭环DNA算法

提出了不等长闭环DNA分子的概念,由此推广了闭环DNA计算模型。给出了固定端点的最短路问题闭环DNA算法,该算法首先对每条弧进行了三组DNA编码,再用有目的的终止技术合成固定端点的所有链,然后通过接入实验和电泳实验得到最短路,并通过检测实验输出所有最短路径。得出了算法的复杂性,为说明算法的有效性给出了一个算例。最后讨论了最短路问题闭环DNA算法在变权网络、自由终点或固定中间点的最短路问题中的应用,并给出了相应的解决方法。由此说明该算法具有广泛的适应性。     

面向农产品安全的关键质量链分析

针对农产品质量安全控制问题,在传统过程能力分析的基础上,将农产品食品链视为一个过程网络,应用PERT/CPM的基本思想,提出了基于质量损失的关键质量链分析评价方法。最后以菠菜产品的种植加工进行了实例分析。     

标定方程AX=ZB的两种新解法

提出了同时确定手眼关系以及机器人坐标系到世界坐标系关系的两种算法。和现有算法相比,给出的算法分别基于对偶四元数和矩阵直积理论,均可一次计算出标定方程的旋转部分和平移部分,不存在误差的传递问题。仿真结果表明了算法的有效性,这些结论对于类似问题的求解具有一定的参考价值。     

覆盖网对域间路由可靠性的影响研究

覆盖网可以利用终端间的多路径路由提升Internet路由的性能和可靠性。如何选取高质量的覆盖网路径,并且准确地对其进行性能评估是有待解决的核心问题。对此,首先提出一种对覆盖网路由的评估方法,该方法从路由可靠性角度进行分析,运用可靠度(RD)这一指标,从多条覆盖网路径的分离度、长度、质量等方面进行综合评估。在此基础上,提出基于广度优先遍历队列(BFSQ)的分布式的启发式算法优化覆盖网多路径的选择。仿真结果表明该算法可以通过少量底层拓扑信息的获取,达到有效提升多路径覆盖网路由可靠性及性能的目的。     

A Research on Path Optimization for Business Structure of Chinese Investment Banks

Chinese investment banks, as a part of the financial market, play an important role in the development of the national economy. With the development of financial market after China＇s entry into WTO, the investment banks confront with immensely international competition. Meanwhile, Chinese investment banks are facing with problems such as elementary business structure, small capitalization and faultiness of governance, which lead to their less competitive power. This article conducts research from the perspective of path optimization of business structure of investment banks. Through economic analysis of path optimization of business structure, we conclude path optimization of business structure of domestic investment banks will lay a solid foundation for enhancing their competitive power. Whereafter this article constructs optimized path for business structure of Chinese investment banks and analyzes the influence factors of the optimized path. At last, this article advises on how to enhance the business structure of Chinese investment banks.     

Participatory Sustainability Impact Assessment: Scientific Policy Advice as a Social Learning Process

This paper deals with the development of scientific policy advice by providing a methodology to foster a social learning process. The methodology, called participatory Sustainability Impact Assessment (pSIA), aims to structure complex problem situations, to clarify interests and basic assumptions, to interpret scientific studies, to develop impact assessment, and to explore sensitivity of uncertainty and lack of information. In pSIA workshops the participants are supposed to build conceptual models with different modelling methods, like System Dynamics, Value Chain Analysis, and Morphological Analysis. A case study is presented that describes a workshop series with political as well as academic actors, applying the pSIA approach to an impact assessment of Economic Partnership Agreements between the European Union and the Eastern and Southern Africa Region.

基于动态RCS的无人机航迹实时规划方法研究

为了提高无人机的生存能力,针对不确定飞行环境的无人机航迹规划问题展开研究,提出了一种基于RCS的无人机航迹实时规划方法,建立了相关计算模型,提出合理的代价函数,并采用多阶段分析博弈评估算法对航迹进行优化。仿真结果表明,基于动态RCS的无人机航迹实时规划方法能够比较好的解决无人机的航迹规划问题。     

基于路集的多阶段任务系统可靠性分析模型

提出了一种基于路集的多阶段任务系统可靠性分析模型,将模型推广到更为一般的任意结构系统。该模型考虑了部件的工作配置关系,把多阶段任务系统的可靠性分析转化为系统的状态转移概率。同时给出了系统状态转移概率求解算法,最后给出了一个应用实例。