融合MT2503与MEMS传感器的惯性导航定位算法

为提高惯性导航室内定位算法的精度与连续性,提出一种融合MT2503与MEMS传感器的惯性导航定位算法,算法以MT2503芯片作为定位终端,并将加速度计传感器、陀螺仪,磁力计等传感器与其进行融合,通过加速度计传感器解算步长、步幅、步频,通过陀螺仪与磁力计来识别定位终端微动偏移量,最后在初始位置上累加定位终端位移得出定位终端实时位置。实验证实通过零速修正和卡尔曼滤波对误差进行校正,有效的解决了MEMS(micro-electro mechanical system)定位算法中存在的导航解算误差累积问题,提升了MEMS惯性导航室内定位算法的精度。     

基于Golay序列调制的视觉诱发电位脑机接口系统

脑机-接口是一种新型的人机交互方式,不依赖于外周神经和肌肉,而是直接将脑电信号翻译成计算机的控制命令。针对传统脑机-接口系统识别准确率低、可识别目标少等问题,提出了一种基于Golay序列调制的视觉诱发电位脑机接口系统。首先受试者必须注视着屏幕上闪烁的目标,通过脑电放大器对人脑表皮层产生的信号进行采集、放大、A/D转换等预处理操作,然后经过实时分析软件对预处理后的脑电信号进行分类识别,取出大脑特定获得下的特征;最后将其转换为计算机的控制命令,并将输出结果反馈给用户。系统选取了Golay序列对刺激目标进行调制,且运用模板匹配法对目标进行识别。实验结果表明,该方法能有效的识别刺激目标,且获得了93.75%的识别准确率。     

反符号权重模板下DTCNN的动态分析及应用

对ＤＴＣＮＮ中非对称的反符号权重函数的稳定性情况进行了分析，扩大了参数的范围，分析了域值变量对此系统的影响，得到一些新的稳态特性，为反对称模板的硬件设计提供了条件。     

Optimal purchasing process for electricity and renewable energy credits with price and demand uncertainty

This paper optimizes the electricity and renewable energy credit (REC) purchasing process for energy distribution. Electricity is traded in deregulated time-sequential markets at fluctuating prices. Optimal electricity purchasing under price and demand uncertainty is a challenging task for electricity distributors, and the recently implemented renewable portfolio standards (RPS) further complicate the purchasing process. Government regulatory decisions concerning the RPS require distributors to purchase corresponding certificates, namely RECs, equivalent to a certain percentage of their electricity sales. This paper formulates and optimizes the joint purchasing process for electricity and RECs. It also analyzes the effect of RPS policy on electricity distributors.     

二维纳米胶体球模板和Cu纳米颗粒阵列的制备

直径为500 nm的纳米球刻蚀模板是通过自组装法制备的.整个实验揭示了自组装过程的影响机制,并得到形状规则,排列均匀一致的单层胶体晶体掩模板.利用离子束溅射得到纳米环阵列,SEM图可以看出纳米环阵列大小均匀,完全按照预期的点阵结构排列.     

一种关系DBMS数据更改功能的实现策略

本文介绍了数据库管理系统 NITDB 的数据更改(插入、删除、更新)功能及其实现技术。提出了一种充分利用查询处理的子系统来提高数据更改处理的效率,使处理过程简单有效的方法,以及应用推迟更新技术来解决更新和删除过程中有关问题的思想。文中还讨论了系统实现视图更新所采用的一种“宏扩展”方法。     

Fabrication of porous alumina templates with a large-scale tunable interpore distance in a H_2C_2O_4-C_2H_5OH-H_2O solution

Highly ordered porous alumina templates with a large-scale tunable interpore distance (100―445 nm) have been successfully fabricated under an electric field of 40―180 V by modifying oxalic acid solution with adequate alcohol. The results under our experimental conditions show that the phenomena of burning and breakdown during the high-field anodization process can be avoided by adding a proper amount of alcohol to the oxalic acid solution. An excellent linear relationship between interpore distance and anodization voltage is obtained under 40―170 V, and the maximum anodization voltage that could be used to avoid burning and breakdown is 180 V.     

基于自适应初始搜索点预测的目标跟踪算法

为了得到恰当的初始搜索点以使得目标跟踪算法避开背景干扰并缩短搜索距离,提出了一种自适应初始搜索点预测的算法。该算法通过对坐标变换参数的变化率进行Kalman滤波来更好地预测初始搜索点;更重要的是,该算法有效地在线估计Kalman滤波器中的模型噪声功率,而非先验地对它们的取值做出假设,因而能够在没有任何人工干预的情况下动态地根据不同的目标运动状况和搜索精度进行实时调整。大量实景视频流上的实验结果均证实了该算法显著提高了跟踪稳定性,并且大幅降低了计算量。     

Steady-State Analysis of Target Tracker with Constant Input/Bias Constraint

Navigation and surveillance applications require tracking constant input/bias targets. When the target＇s trajectory follows a constant input/bias constraint, model mismatching caused by conventional tracking algorithms can be handled by a delayed update filter （DUF）. The statistical convergence and stability properties of the delayed update filter were studied to insure the rationality of its steady-state analysis. A steady-state filter gain was then designed for a constant-gain DUF to reduce the computations without much performance loss. Simulations demonstrate the potential of the constant-gain DUF, and the CGDUF is nearly 60% faster than the DUF without much loss in steady-state tracking accuracy.