用非线性反馈控制混沌Roessler系统到达任意目标

本文设计了识别不确定Roessler系统未知参数的观测器，提出了控制Roessler系统中混沌的非线性反馈控制策略。数值模拟结果表明：观测器可以有效地标识未知参数；选取不同的目标参数，既可以使Roessler系统稳定在不同周期轨道上，也可以稳定在任意目标点上。     

Design and implementation of 2DOF spherical inverted pendulum

1.INTRODUCTION Theinvertedpendulumsystemisaclassicalbench markfortheeducationandresearchintheautomation andmechanicalengineeringfield.Manyeducatorsfo cusontheinvertedpendulumcontrolsystembecause theexperimentsexhibittheadvantageandvalidityof moderncontroltheoryveryintuitively.Researchers alsohopetoverifynewalgorithmsonthesystem. Thebasicobjectiveoftheinverted pendulumsystem iskeepingthebalanceofthependulumatthe‘invert ed’status[1～3].Wehavecomparedtheperformance ofthreemoderncontrolm…     

一种用于自动标引系统的主题词自动切分方法

该文着重介绍一种知识指导下的主题词自动切分方法，该方法能灵活处理自然主题词、大幅度跳跃及关联词的切分问题，其字面切分精度达９９．３３％，综合切分精度达９８．９１％。     

一阶非线性非完整系统的改进的Routh方程

导出了一阶非线性非完整系统的改进的Ｒｏｕｔｈ方程并举例说明其应用，从而提供了建立非完整系统的运动微分方程的一种新方法。该方程的特点在于其第一组方程中只包含系统的独立的广义动量，而第二组方程中不包含任何广义动量。同时，本文结果也是对黄昭度先生的“改进的Ｒｏｕｔｈ方程”一文中的某些错误的修正。     

一维非线性映象新模型的分叉,窗口和混沌研究

本文针对几个一维非线性映象自已设计的新例子，分析了其中的混沌运动，获得了新的特点，目的是更加完善混沌运动机理，有助于其它混沌现象的研究．     

基于D-S推理的模糊模式识别方法

针对战场环境下数据融合系统的ESM设备获得的辐射源信息在时间上的冗余性,提出了一种基于D-S证据推理的模糊模式识别方法。该方法通过对观测样本信任度(模糊隶属度)的重新分配,有效地解决了辐射源信息不确定性问题,提高了辐射源的正确识别率。最后通过单一的模糊识别和D-S证据组合的模糊识别两种方法的仿真对比实验验证了后者的有效性。     

Face Recognition Based on Support Vector Machine and Nearest Neighbor Classifier

Support vector machine (SVM), as a novel approach in pattern recognition, has demonstrated a success in face detection and face recognition. In this paper, a face recognition approach based on the SVM classifier with the nearest neighbor classifier (NNC) is proposed. The principal component analysis (PCA) is used to reduce the dimension and extract features. Then one-against-all stratedy is used to train the SVM classifiers. At the testing stage, we propose an al-     

Lyapunov稳定性逆定理的另一种证明

Lyapunov方法是研究非线性控制问题的重要手段,而Lyapunov稳定性逆定理又是应用此方法的主要定理,但其证明却非常烦琐.利用直接取指数函数的方法,简化了Lyapunov稳定性逆定理中关键函数G的构造,给出了Lyapunov稳定性逆定理的另一种简单的证明方法.     

用于图像目标识别的神经网络方法

本文提出了一种全局最优的神经网络（ＦｕｌｌＤｏｍａｉｎＯｐｔｉｍｕｍＮｅｕｒａｌＮｅｔｗｏｒｋ）模型用于目标识别。通常所设计的神经网络不能保证全局最优，使得网络不一定收敛到期望样本点上。本文的模型采用了先设计稳定点、再构造吸引域的方法，提高了网络的识别正确率及速度。针对图像识别中矢量维数大的实际，提出了一种不变性方法，使得样本维数下降而分类距离保持不变。同时又证明了网络的收敛性、收敛速度及映射保距等。计算机模拟结果表明，网络对噪声或缺损图均能正确识别。     

Weight measurement using image-based pose analysis

Image-based gait analysis as a means of biometric identification has attracted much research attention. Most of the existing methods focus on human identification, posture analysis and movement tracking. There have been few investigations on measuring the carried load based on the carrier＇s gait characteristics by automatic image processing. Nevertheless, this measurement is very useful in a number of applications, such as the study of the carried load on the postural development of children and adolescence. In this paper, we investigate how to automatically estimate the carried weight from a sequence of images. We present a method to extract human gait silhouette based on an observation that humans tend to minimize the energy during motion. We compute several angles of body leaning and determine the relationship of the carried weight, the leaning angles and the centroid location according to a human kinetic study. Our weight determination method has been verified successfully by experiments.