bread

：以matlab工具箱的形式给出了静态及动态未知环境下移动机器人地图构建的仿真研究平 台。通过分析polaroid6500声纳、电子罗盘、编码器等传感器的工作原理建立了参数可调的传感器 模型。这些传感器为机器人提供位姿及环境信息。在分析已有基于hough变换的线段提取算法的基 础上提出了抑止。杂线段。现象的CIHT[Clus~r Inhibiting Hough Transform]算法．采用cIHT 对声蚋信息作环境特征提取，便于开展基于环境特征的未知环境地图构建算法方面的研究工作。(Abstract：A simulation system ofmapping for sonar-based mobile robot was developed易0 in static and dynamic unknown environments in a foFin ofmatlab toolbox．Sensor models with tunable parameters were established by interpreting and simulating the mechanism of encoder。digital compass．and polaroid 6500 sonar．These sensors provide the information of robot pose and environments．A novel C1HT[Cluster Inhibiting Hough Transfo删J algorithmfor sonar data was proposed by ana~zing existing line segments detection methods based on Hough Transform，which would make the research of feature—based mapping of unknown environment more convenient．)

基于matlab的IIR滤波器实验的源程序及实验结果图形(IIR filter based on the matlab source code and experimental results of graphics)

图像特征提取 特征提取 matlab 特征提取 图像特征 (Image feature extraction feature extraction feature extraction matlab image features)

pca eigenface application

标定程序 用在立体视觉 (calibrate stereomatching)

KALMAN滤波与平滑原理及matlab经典源码，内附滤波平滑处理结果图形，比较适合初学(KALMAN filtering theory and matlab classic source, more suitable for beginners)

Equations which describe the TCS/MM method

an adaptive median filter

使用MATLAB，ReliefF算法应用于特征提取，适用于shp属性数据。(Using MATLAB, the ReliefF algorithm is applied to feature extraction and is suitable for SHP attribute data.)

多目标优化 相对传统多目标优化方法, PSO在求解多目标问题上具有很大优势。首先, PSO的高效搜索能力有利于得到多目标意义下的最优解 其次, PSO通过代表整个解集的种群按内在的并行方式同时搜索多个非劣解,因此容易搜索到多个Pareto 最优解 再则, PSO的通用性使其适合于处理所有类型的目标函数和约束 另外, PSO 很容易与传统方法相结合,进而提出解决特定问题的高效方法。就PSO 本身而言,为了更好地解决多目标优化问题,必须解决全局最优粒子和个体最优粒子的选择问题(Compared with the traditional multi-objective optimization of multi-objective optimization method, PSO in solving multi-objective problem has a great advantage. First, PSO is conducive to the efficient search capabilities are more objective sense of the optimal solution Secondly, PSO representative of the entire solution set through the population by way of the inherent parallel search multiple non-inferior solution, and this can easily search for the most number of Pareto optimal solution Furthermore, PSO' s versatility make it suitable for handling all types of objective function and constraints addition, PSO is easy to integrate with the traditional method, and then propose an efficient way to solve specific problems. The PSO itself, in order to better address the multi-objective optimization problems, the need to address the global best particle and the individual selection of the optimal particle)