融合WiFi、激光雷达与地图的机器人室内定位

由于室内环境受多径效应影响,单一WiFi定位效果不佳;激光雷达(LiDAR)虽然测距定位精度高,但在室内存在大量单一、重复的场景结构(如走廊)时,往往会由于无法提取有效特征进行匹配而造成大量错误定位。因此,该文提出基于卡尔曼滤波框架的WiFi、激光雷达与地图的融合定位新方法。其中,滤波器的状态定义为机器人当前与历史时刻的位置序列。滤波器的观测值由两部分组成,一部分为该文所提基于多环路分割地图下信号强度加权匹配的WiFi指纹定位结果;另一部分来自激光雷达在单一重复场景中计算出来的高精度相对定位结果(如横向定位)。利用场景地图中的先验参考位置,可将该横向定位结果转变为机器人位置的线性约束。最后,利用卡尔曼滤波器实现机器人高精度的融合定位结果。实验中,针对两种典型的单一、重复的室内场景,分别采用2维与3维激光雷达对该文算法进行验证。实验结果表明,由于激光横向定位精度可达厘米级,结合厘米级地图可以极大提高机器人定位精度。与单一WiFi定位算法相比,利用激光雷达计算出来的相对定位结果结合场景地图,平均定位误差可降低70%～80%,在满足机器人实时定位需求情况下,实现定位精度与稳定性的显著提升。     

基于RSSI的Wi-Fi室内定位常用算法对比

WiFi定位是室内定位的方法之一,其观测值主要为WiFi的信号强度(received signal strength indication,RSSI),利用信号强度进行室内定位的方法基本上可以分为两种,一种根据模型计算接收节点与发射节点的距离,估计接收节点的位置;另一种是根据事先采集的信号强度指纹进行空间匹配。通过在不同场景下对两种定位算法的比较,基于信号强度指纹匹配的定位精度明显高于基于估计接收节点的定位精度,但与指纹匹配法相比,位置估计法更加灵活。     

基于激光雷达的有界区域快速全局定位算法

室内全局定位技术是自动驾驶车辆和机器人的核心组成部分,也是当今的研究热点.基于该时代背景,本文提出了一种基于激光雷达的有界区域快速全局定位算法,将全局地图与局部地图的特征向量进行对比匹配,实现了雷达在有界区域内的精确定位.利用仿真实验对算法的有效性和定位精度进行了分析,发现其平均位置误差比为0.791％,平均朝向误差为...     

基于自组织可增长映射的移动机器人仿生定位算法研究

为提高移动机器人在同步定位和地图构建(SLAM)中的定位精度,该文提出一种基于自组织可增长映射 (GSOM)的仿生定位算法。该方法将位置细胞的激活特性和神经网络输出层神经元建立响应连接,通过GSOM神经网络构建空间的拓扑地图,利用感知距离信息实现位置细胞的激活响应从而估计机器人位置,以此还原机器人的运行路径。实验结果表明细胞间隔R对定位精度有较大影响,选取合适的细胞间隔能有效地减少神经网络的学习时间,提高定位精度,该文算法平均误差在0.153 m以内,定位精度达到90.243%,均优于原有算法。经验证该文算法建立的模型能够实现机器人的空间位置表征,提高了机器人在实验场景下的定位精度,表现出良好的位置估计性能。     

基于机器人激光定位的一种改进AMCL算法

高效的定位算法是实现机器人自主运动的前提,由于激光模型受复杂环境的限制,传统自适应蒙特卡罗定位(AMCL)算法提供的位姿精度有限.提出一种增加扫描匹配(SM)和离散傅里叶变换(DFT)的优化AMCL算法,将传统AMCL的加权均值输出作为SM的初始值,通过构建激光雷达观测点与先验地图的匹配函数模型,利用高斯牛顿的方法优化求解,最终通过DFT滤波滤除位置处的小抖动,提升了系统的稳定性和鲁棒性.通过运动中的绝对定位实验和重复定位,实验验证了优化算法优于传统AMCL算法,优化算法有效提高了系统定位精度,同时保证了鲁棒性.     

基于CSI相位矫正的室内指纹定位技术研究

针对室内定位服务的迫切需求,提出了基于信道状态信息相位作为指纹信息的室内定位方法,搭建了室内WiFi定位平台,实现了室内环境下的高精度位置估计。在离线阶段获取WiFi的信道状态信息,包括振幅、相位等参数,利用线性变换的方法对采集到的CSI相位信息进行预处理,建立鲁棒的信号指纹数据库。在线阶段提出了改进的加权KNN算法,对初次估计坐标进行二次匹配,从而完成设备及人员的定位。实验结果显示,平均定位精度达到了0.63 m,相较于传统的室内定位技术,定位精度有了明显的提升。     

基于信号幅值分布的室内指纹定位算法

室内定位算法是基于位置服务领域研究的难点之一。针对室内定位应用场合和精度问题,提出了利用K-means和KNN融合算法对覆盖率广的WiFi信号进行指纹定位。WiFi指纹定位主要问题是前期指纹库数据的精确以及后期数据的匹配效果。首先,对WiFi信号的概率分布进行研究,弥补了一直以来由于K-means是无监督学习带来的k值的难以选取的缺陷,提高了指纹库的精确性,同时确保数据实时性。后期在线数据处理利用KNN分类算法进行后期在线定位过程的准确性。经多个实验场景测试结果表明,该算法在室内定位精度上3 m定位精度概率保持在78.4%,4 m精度保持在93.6%,基本上保证了室内定位精度的要求。     

基于区域限制的WiFi/PDR融合实时定位算法

针对室内定位系统在实际应用场景中算法复杂度高、运算量较大的问题,文中设计并实现了一种基于EKF的WiFi/PDR融合定位系统。在WiFi指纹定位部分提出了基于自适应滑动窗口的指纹匹配方法,通过邻近状态的RSSI欧式距离解算得到搜索窗口,以动态调整指纹库的匹配范围,从而实现了定位结果的快速收敛。在融合定位阶段,结合EKF与PDR的系统特性来解决时间配准问题。以WiFi数据更新为基准,利用EKF算法进行数据融合,在融合数据不同步时由PDR直接输出定位结果。实验结果表明,该定位系统具有良好的运行效果与稳定性,所提方法在实际定位场景中的平均定位误差为2.27 m,并在80%的情况下能够达到3 m的定位精度。     

基于点云极化表征与孪生网络的智能车定位

基于3维激光雷达(LiDAR)的智能车定位在地图存储空间与匹配效率、准确率等方面仍存在诸多问题。该文提出一种轻量级点云极化地图构建方法：采用多通道图像模型对3维点云进行编码生成点云极化图,利用孪生网络结构提取并训练点云极化指纹,结合轨迹位姿信息构建点云极化地图。还提出一种基于点云极化地图匹配的智能车定位方法：采用孪生网络对查询指纹与地图指纹进行相似度建模实现快速的地图粗匹配,采用基于2阶隐马尔可夫模型(HMM2)的地图序列精确匹配方法获取最近的地图节点,通过点云配准计算车辆位姿。使用实地数据集和公开的KITTI数据集进行测试。实验结果表明,地图匹配准确率高于96%,定位平均误差约为30 cm,并对不同类型的LiDAR传感器与不同的场景具有较好的鲁棒性。     

基于三维激光雷达的室外环境定位算法研究

针对使用激光雷达进行室外定位的问题,提出一种基于三维立体激光视觉的特征匹配定位算法,以提高激光视觉在室外环境的位置特征可靠性和定位精度。该算法首先从原始点云数据中提取三维视觉的物类信息;再对提取的物体类进行特征提取,根据其个体特征进行分类并剔除非地标类,根据与相邻类的位置关系提取群体特征;最后通过与特征地图进行匹配实现全局定位。通过数据集仿真验证,该算法能够准确地提取有效类和位置特征,平均每帧的类提取率达到90%以上,对95%以上的场景都能成功定位,且定位误差不随地图范围增大而积累,具有稳定的定位性能和较高的定位精度。     

GEMNET: a generalized, shuffle-exchange-based, regular, scalable,modular, multihop, WDM lightwave network

GEMNET is a generalization of shuffle-exchange networks and it can represent a family of network structures (including ShuffleNet and de Bruijn graph) for an arbitrary number of nodes. GEMNET employs a regular interconnection graph with highly desirable properties such as small nodal degree, simple routing, small diameter, and growth capability (viz. scalability). GEMNET can serve as a logical (virtual), packet-switched, multihop topology which can be employed for constructing the next generation of lightwave networks using wavelength-division multiplexing (WDM). Various properties of GEMNET are studied     

Thermodynamics of copper matte converting: Part III. Steady-state volatilization of Au,Ag, Pb,Zn, Ni,Se, Te,Bi, Sb,and As from slag,matte, and metallic copper

A comprehensive compilation of various thermodynamic data required for a complete analysis of copper matte converting reactions is presented. The data comprise estimated free energies of formation for such gases as SeO, SeS, TeO, TeS, BiO, BiS, SbO, SbS, AsO, and AsS, as well as activity coefficients in dilute copper alloys and vapor pressures of various elements and compounds. The volatilization of minor elements in steady-state reactors comprising gas and several condensed phases is mathematically formulated, and a parameter which governs the volatilization in such reactors is defined and named volatilization constant. The vapor pressures of various volatile species are calculated thermodynamically for the Noranda Process reactor by assuming equilibrium conditions. The volatilization constants of various minor elements are expressed explicitly as functions of oxygen and sulfur activities. Formerly Associate Professor, Department of Metallurgical Engineering, University of Utah, Salt Lake City, UT     

Bits, symbols, bauds, and bandwidth

The use of clear, concise, and unambiguous language in telecommunications engineering is vital to communicate a desired meaning and understanding. Terminology should be based on rigorous engineering principles and traceable to well known and dependable sources. This article deals with four everyday terms commonly encountered in the popular semi-technical press as well as in serious engineering periodicals. The misuse of these and other terms can lead to low-balled cost proposals, court litigation, and patent infringement cases. Imagine how an engineering student can be confused when she/he typically encounters bandwidth measured in bits per second     

Novel planar, square-shaped, dielectric-waveguide, single-, and dual-mode filters

This paper presents a novel concept of a square-shaped dielectric-waveguide resonator, which can easily realize a dielectric-waveguide cross-coupled filter or a dielectric-waveguide dual-mode filter, using the conventional printed circuit board (PCB) process. This planar dielectric-waveguide resonator has a higher Q value than a microstrip resonator. The physical mechanisms of both single- and dual-mode filters are elucidated. Some new coupling structures are developed for undergoing the PCB process. Filter design procedures are detailed. Design curves for the coupling coefficients of internal- and external-coupling structures are determined by full-wave finite-element-method electromagnetic calculations. Measurement results are highly consistent with theory for all trial filters. This study offers an effective means of producing low-cost high-performance planar circuit filters.     

PBX技术.市态.营销.选购

本文介绍了ＰＢＸ交换机的技术及市场需求态势。并提出ＰＢＸ产品的营销要素及其用户在选购时应遵循的原则。     

Dilemmas,uncertainty, risks,and board performance

This paper traces the origins of the words risk, uncertainty, governance and development and argues that a deeper understanding of each allows more effective decision taking in the boardroom. It argues that the role of the board of directors is to balance and rebalance continuously their irresolvable dilemma — ‘how do we drive our enterprise forward while keeping it under prudent control?’ It argues that it is the board’s role to focus on uncertainty, rather than risk, and this requires a different set of intellectual skills from board members to be able to cope with monitoring a range of diverse scenarios. This is crucial for a board to develop stronger ways of both leading their organisation and of ensuring the connectedness of the learning within and between the board and the operational unit’s risk taking. It advocates that to achieve this a board must develop new ways of learning — especially of thinking strategically and becoming more sensitised to the dynamics of their changing external environments. This will take them well beyond the comforts of their specialist managerial disciplines and into the true world of directing. The paper demonstrates the importance of continuous learning in reducing organisational uncertainty and risk. It uses the ‘Learning Board’ process and model as the central processor of both strategic and operational learning to create an annual rhythm for the board’s year.