基于MATLAB的汉语数字语音识别系统

应用动态时间规整（DTW）为识别算法,采用MFCC（MEL频率倒谱系数）为主要语音特征参数,建立了一个汉语数字语音识别系统,其中包括语音信号的预处理、特征参数的提取、识别模板的训练、识别匹配算法;同时,提出利用MATLAB图形用户界面开发环境设计语音识别系统界面,设计简单,使用方便,系统界面友好。     

基于STM32单片机的嵌入式语音识别系统设计

设计了一款以STM32F103为核心的自然语言识别系统,为满足实时语音识别系统对内存资源和运算速度的要求,基于硬件资源合理设计语音处理算法,在嵌入式平台上实现了对孤立词语的语音识别。首先根据背景噪声和语音信号的时域特征差异设定相应门限值,从而实现了对语音信号的端点检测。然后针对语音识别中传统梅尔倒谱系数对语音的高频信息敏感度较低,对语音信号分别提取梅尔倒谱系数(MFCC)与翻转梅尔倒谱系数(IMFCC),结合Fisher准则构造混合特征参数。最后采用动态时间规整算法实现语音识别。因系统体积小、便携性好等特点,易于实现对不同设备的语音控制,有一定的市场前景。     

基于段级特征的快速说话人识别算法及DSP实现

本文在TMS320C6701EVM板的基础上实现一种快速的说话人识别系统.提出一种基于段级语音特征的说话人识别的快速算法,该算法在传统的GMM算法的基础上使用段级语音特征对测试语音进行数据量压缩,以减少计算时间.并基于车比雪夫和不等式提出了基于协方差模型的段级特征的失真测度描述.根据实验选择了段级特征语音段长度,实验表明该算法在不显著影响识别率的基础上有效地减少了算法延迟,提高了识别速度.     

一种改进的VQ算法在说话人识别上的应用

本文介绍了一种基于动态权值改进的矢量量化(VQ)方法。根据不同环境下不同说话人语音特征各阶差异较大的特点.对矢量量化算法进行了改进。经实验结果证明.该方法提高了说话人识别系统的识别率。     

基于HMM的非特定人汉语语音识别系统

设计了基于隐马尔可夫模型(Hidden Markov Model,HMM)的非特定人汉语语音识别系统,主要由录音、训练和识别三大模块构成."录音模块"首先录制一段指定长度的语音信号,然后通过对语音信号的短时能量和过零率进行门限检测,标志出有效语音段并保存."训练模块"利用Baum-Welch算法计算语音样本的MFCC(Mel Frequency Cepstrum Coefficient)参数生成识别用的语音模板."识别模块"利用HMM识别算法比较语音信号和语音模板的相似概率,找到最大值输出,完成语音识别功能.最后,在MATLAB中实现了该语音识别系统.实验结果表明,系统的识别率为60%以上,若结合足够的训练,识别率可以更高.     

基于ROS的语音导航智能小车设计与研究

为提高移动智能小车SLAM自主导航避障性能和运动稳定性,提出一种基于ROS框架的语音导航智能小车设计方法.建立智能小车的运动学模型,完成硬件电路和软件平台的搭建,基于粒子滤波器的FastSLAM算法,结合A*和动态窗口法DWA,编写ROS框架下的Gmapping功能包,实现智能小车的运动实验仿真、路径规划和动态避障.采用语音识别控制策略,利用Gazebo仿真和Rviz可视化,开展智能小车的语音导航试验.对比分析A*结合TEB算法的小车避障耗时及运动平顺性,仿真与实车验证结果均表明,本文方法设计的智能小车避障时间短,移动路径具有连续性,语音导航对环境的适应性强.     

面向MBD模型的特征信息自动识别算法研究

针对特征识别技术对产品制造信息识别的缺失以及匹配算法的复杂性问题，提出了一种基于MBD模型的特征信息自动识别算法。首先，通过EXPRESS和Java映射，构建STEP文件解析器解析MBD模型信息。其次，建立基于图和提示的几何特征识别算法，用表征面邻接图作为匹配算法的数据结构，以提示作为搜索与特征库进行匹配。然后，提出基于关系树的产品制造信息识别方法识别产品制造信息。最后，以轴和机匣作为测试对象，在Intellij IDEA平台中对所提方法进行了验证，分析了匹配算法时间复杂度。结果表明，该方法能够识别几何特征信息和产品制造信息，本匹配算法时间复杂度为O(n)，优于时间复杂度为O(n²)的传统匹配算法。     

仿壁虎爬壁机器人的语音识别及语音回放系统研制

随着人们对工业生产便捷性的需求日渐增强，人工智能技术一直保持着发展的动力与活力，因此与语音识别技术相结合的机器人控制拥有良好的发展前景。这里将语音识别技术应用于仿生壁虎爬壁机器人的控制中，研制一种仿壁虎爬壁机器人的语音识别及语音回放系统。语音识别方式采用模板匹配法以适合爬壁机器人使用者的语音识别系统输入命令的特点，降低了控制系统对存贮容量的要求，同时提高了处理速度。语音回放模块使用结构简单的APR9600语音录放芯片，可获得较好的音质效果。     

一种基于动态剔除和场景匹配的 Robust SLAM 方法

针对动态环境中移动物体和结构变形引起激光雷达自主定位精度下降的问题,本文提出了一种Dynamic Lego-loam方法。为减小动态点误匹配给激光雷达里程计带来的误差,该方法首先在里程计精解算之前,提出了一种基于先验信息的点云粗配准算法用以剔除动态点,提高了激光雷达里程计的匹配精度。然后,针对环境中的动态变化带来的误差累积和建图重影问题,利用场景匹配的方法优化了传统基于半径的闭环检测方法。大范围采用基于半径的粗搜索快速定位至局部场景,小范围构建区域高度差描述符精确匹配至最相似历史帧,最终实现了精准的闭环检测并提高了动态环境中的建图精度。实验结果表明,在动态环境下,Dynamic Lego-loam算法相较于Lego-loam算法自主定位精度提高了63%。     

应用动态时间规整与矢量量化的语音识别算法

提出了一种基于动态时间规整(DTW)的改进平均最小距离识别算法,改善了孤立词识别的鲁棒性并提高了识别率。同时对矢量量化(VQ)算法分析了不同码本大小下的识别率,并比较了各种算法的运算时间。通过在MatLab上实现特定人孤立词小词汇量语音识别,实验的结果表明:基于DTW算法的改进平均最小距离法识别率显著提高;码本较大时VQ算法的识别率最高;VQ算法的识别率一般高于DTW算法且运行时间短。     

Fault diagnosis of motor drives using stator current signal analysis based on dynamic time warping

Electrical motor stator current signals have been widely used to monitor the condition of induction machines and their downstream mechanical equipment. The key technique used for current signal analysis is based on Fourier transform (FT) to extract weak fault sideband components from signals predominated with supply frequency component and its higher order harmonics. However, the FT based method has limitations such as spectral leakage and aliasing, leading to significant errors in estimating the sideband components. Therefore, this paper presents the use of dynamic time warping (DTW) to process the motor current signals for detecting and quantifying common faults in a downstream two-stage reciprocating compressor. DTW is a time domain based method and its algorithm is simple and easy to be embedded into real-time devices. In this study DTW is used to suppress the supply frequency component and highlight the sideband components based on the introduction of a reference signal which has the same frequency component as that of the supply power. Moreover, a sliding window is designed to process the raw signal using DTW frame by frame for effective calculation. Based on the proposed method, the stator current signals measured from the compressor induced with different common faults and under different loads are analysed for fault diagnosis. Results show that DTW based on residual signal analysis through the introduction of a reference signal allows the supply components to be suppressed well so that the fault related sideband components are highlighted for obtaining accurate fault detection and diagnosis results. In particular, the root mean square (RMS) values of the residual signal can indicate the differences between the healthy case and different faults under varying discharge pressures. It provides an effective and easy approach to the analysis of motor current signals for better fault diagnosis of the downstream mechanical equipment of motor drives in the time domain in comparison with conventional FT based methods.     

Design and implementation of an Internet based effective controlling and monitoring system with wireless fieldbus communications technologies for process automation--an experimental study

The network requirements of control systems in industrial applications increase day by day. The Internet based control system and various fieldbus systems have been designed in order to meet these requirements. This paper describes an Internet based control system with wireless fieldbus communication designed for distributed processes. The system was implemented as an experimental setup in a laboratory. In industrial facilities, the process control layer and the distance connection of the distributed control devices in the lowest levels of the industrial production environment are provided with fieldbus networks. In this paper, the Internet based control system that will be able to meet the system requirements with a new-generation communication structure, which is called wired/wireless hybrid system, has been designed on field level and carried out to cover all sectors of distributed automation, from process control, to distributed input/output (I/O). The system has been accomplished by hardware structure with a programmable logic controller (PLC), a communication processor (CP) module, two industrial wireless modules and a distributed I/O module, Motor Protection Package (MPP) and software structure with WinCC flexible program used for the screen of Scada (Supervisory Control And Data Acquisition), SIMATIC MANAGER package program ("STEP7") used for the hardware and network configuration and also for downloading control program to PLC.     

基于多目视觉的布匹瑕疵检测系统硬件部分设计

为实现布匹瑕疵检测的自动化,设计了基于多目视觉的布匹瑕疵在线检测系统.主要针对系统的性能要求,对系统的硬件部分进行了设计,除了对系统的主要硬件设备进行选型和设计外,还在验布机的基础上,设计了系统的机械传动部分,保证了布匹在一定的转速下,系统能够获得稳定清晰的图像,进而实现瑕疵的检测和定位.     

Multi-stage diagnostic system for reciprocating compressor using DTW technique

The reciprocating compressor has widely been used in the industry because of low cost to mass production and high efficiency of performance. If an abnormal compressor is produced, high service cost occurs due to the increased service claim rate (SCR). A great deal of research has been conducted for real-time diagnosis of the compressor while it is being made in the manufacturing line. However, since the compressor structure is quite complicated, it is still difficult to find the cause of the compressor fault. Even with outstanding diagnostic tools, it is hard to screen an abnormal compressor at one time. We have developed a multi-stage diagnosis scheme to monitor the abnormal compressor. In this diagnosis scheme, the dynamic time warping (DTW) technique is employed to diagnose the fault compressor. The LAN based network is applied for connecting from first diagnosis PC to second diagnosis PC. The multi-stage diagnosis manufacturing line is established and the DTW technique has been successfully implemented in the diagnostic system. The multi-stage DTW system screens most of the abnormal compressor. The gage R&R is conducted to secure the reliability of the system.     

基于CAN总线的观察型无人遥控潜水器

研制了一种基于CAN总线的遥控潜水器(ROV)的监控系统.利用模块化的设计方法,实现了系统的软硬件结构;详细介绍了CAN总线应用层协议的具体格式及实现方法.     

一种嵌入式系统软件水印识别器的VLSI实现

提出了一种基于分支软件水印方案识别器的VLSI实现.硬件形式的识别器一般难以获取嵌入式系统内部的函数返回地址,通过改进获取指纹信息分支函数(FBF)入口的方法,使得只需要根据系统的程序指针和指令字就可以完成FBF函数入口的识别,使原有方法更适用于嵌入式系统.最后针对所支持的指令集进行了仿真测试,并对设计实现进行了逻辑综合.性能分析表明,硬件形式识别器识别速度远远超过软件的识别速度.     

标准型上移轨道门平衡系统的优化设计

本文就标准型上移轨道门平衡系统的设计原理进行了详细探讨,在此基础上建立了参数设计优化数学模型,编制了优化设计程序,并通过工程实例说明了优化设计的全过程。     

一种变化光线下的拼图机器人识别系统

文章改进了图像分割和模板匹配技术，提高了系统识别的准确性和鲁棒性，设计并实现了一种适应光线变化的拼图机器人识别系统。文章最后给出了实验与仿真，证明了识别系统的可行性。