卡尔曼滤波在MEMS惯性姿态测量中的应用

由于测控成本和有效载荷的限制,一般采用微机电系统(MEMS)惯性传感器来测量小型无人机的飞行姿态。在MC9S12XS128单片机上通过嵌入式软件编程实现了卡尔曼滤波算法,并在JZJ-1型自准直仪转台上对MEMS加速度计和陀螺仪的输出信号进行了数据融合试验,较好地解决了MEMS惯性测量系统的零漂和机械振动干扰问题。     

基于改进卡尔曼滤波在车辆组合导航中的研究

针对车辆在长时间行驶的状态中,组合导航经常会出现滤波精度不高的问题,提出一种采用遗传算法对卡尔曼滤波器进行在线调节的自适应数据融合方法,根据位置误差系数和卡尔曼滤波器的新息统计信息,实时调整参数值,将卡尔曼滤波器调整到运行过程中的最优状态。通过MATLAB软件对该方法进行仿真试验,模拟INS/GPS组合导航系统运行情况,表明这种方法和传统卡尔曼滤波算法相比较导航精度明显提高。     

U-卡尔曼滤波在状态估计中的应用

本文介绍了新的滤波估计方法-U卡尔曼滤波.UKF能给出最佳估计的至少二阶近似.本文通过一个时间序列的仿真实例说明UKF比EKF的估计精度要高.     

卡尔曼滤波在滴定分析法中的应用

本文研究卡尔曼滤波在电位滴定分析法中的应用。介绍了滴定分析模型、卡尔曼滤波算法及程序框图。卡尔曼滤波对于提高那些反应较慢或电极响应迟缓的滴定体系的分析准确度有很大帮援。考察了采样间隔、滴定速度及系统误差等参数对结果准确度的影响。将本法应用于HAc/NaOH、KI/AgNO_3及CaCl_2/EDTA三种下同的滴定体系,分析结果的准确度均大大提高。     

卡尔曼滤波在足球机器人定位中的应用

在动态环境中的自我定位是移动机器人的一个关键问题。机器人要完成复杂的任务,首先要感知周围环境,得到机器人的位姿,才能完成后续的决策及规划。卡尔曼滤波是一种最优化回归数据处理算法,已广泛应用于机器人导航、传感器数据融合、导弹追踪。本文将卡尔曼滤波应用于足球机器人定位中。     

卡尔曼滤波预测在客流量统计中的应用

运动物体的跟踪是现代科学技术里的重要一环,在众多领域都有着广泛的应用,比如说雷达检测跟踪系统。运动目标的跟踪分为预测和匹配两方面,其中预测尤为重要。文中运用向量卡尔曼滤波的方法对在5W红外LED光源照射下的人体的运动进行实时跟踪预测,为后续匹配工作提供高准确率的保障。在实验室模拟环境下进行实验,实验结果表明预测的误差能够控制在10个像素点以内。在对公交车客流视频进行处理后,有非常高的准确率,实时性也很好。     

图像处理技术在形貌测量中的应用

本文以微分干涉显微法测量表面形貌为例,分析影响精度的各种因素,重点介绍图像处理技术在测量中应用的原理、实现方法及实际效果.证明应用图像处理技术,对改善二维、三维图像质量,提高检测精度的可行性.     

卡尔曼滤波及其在时间序列预测中的应用

根据时间序列预测的特点和要求,分析了传统时间序列预测方法的不足,提出了将卡尔曼滤波应用于时间序列预测。推导了基于卡尔曼滤波的ARMA模型参数实时更新算法,并采用功率谱密度分析方法确定预测模型的形式与阶数。最后,通过对光纤陀螺随机漂移建模进行了实证研究。     

卡尔曼滤波技术在捷联惯导系统初始对准中的应用

阐述了卡尔曼滤波技术在捷联惯导系统初始对准中的应用,建立了捷联惯系统初始对准的误差模型和卡尔曼滤波模型观测方程,进行了卡尔曼滤波仿真,并分析了提高采样频率对系统的估计精度和收敛速度的影响。仿真结果表明,此方法算法简单,能有效缩短初始对准时间,对准稳态精度较高,是一种理想的初始对准方案。     

卡尔曼滤波在机载监视系统中的应用研究

机载监视系统是现代航空领域很重要的一个研究内容,通过它可以实时观察记录飞机的多项信息,主要对监视系统的伺服控制单元进行了研究设计。因受飞机姿态变化、空中风阻力、机械振动、负载扰动、观测噪声等因素影响,导致电机旋转不稳,成像仪随之不稳,造成显示屏画面抖动,影响扫描监视效果。考虑实际应用中常规PID控制器对电机控制的局限性,提出了在PID控制算法上加入卡尔曼滤波,改进伺服控制效果。结果表明:新的控制方法对干扰有抑制作用,控制效果明显改善。     

声学多普勒流速剖面仪ADCP及其在水文测流中的应用

本文阐述了声学多普勒流速剖面仪ADCP的流量测量原理,并通过对ADCP流量测量仪与常规流速仪的比较,说明了ADCP流量测量的特点及其测量过程,并对ADCP流量测量中可能出现的问题进行了总结归纳,指出了应用中应注意的问题。     

多普勒测流仪在海洋水文测量中的应用

根据海洋水文测量的实际,对多普勒测流仪和常规海流计的海流测量方式做了比较分析,验证了多普勒测流仪的诸多使用优点,并对多普勒测流仪的工作原理、工作特性和使用方法进行了详细的论述。     

ADCP的引进及在淮河流量测验中的应用

介绍了ADCP的测流原理和在水文系统使用的优点，通过淮滨站实际使用的例子说明使用ADCP的紧迫性和必要性，提出ADCP使用注意事项和管理规程。     

走航式ADCP在澜沧江流量测验中的应用

通过走航式ADCP在澜沧江流量测验的实际应用,阐述了ADCP流量测量的过程及特点,明确了其适用环境以及测验使用条件.对走航式ADCP与常规缆道流速仪法测得的流量比测,验证了ADCP的测流精度,相比具有更高的测验效率,省工省时.说明ADCP测流系统技术先进,功能齐全、性能稳定、安全可靠、安装方便,数据采集自动化程度较高,在测量现场能直接地获取各类测流数据,满足澜沧江流量测验,为ADCP在云南及西南河流流量测验的推广应用抛砖引玉.     

Analysis and compensation of ADCP current direction error caused by ambient magnetic field

Internal magnetic compasses of acoustic Doppler current profilers (ADCPs) are easily disturbed by significant amount of hard/soft iron in some moorings or ships, causing the significant current direction error. In this paper, with the consideration of the roll and pitch of ADCPs mounted on moorings or ships, we deduce and derive the quantitative relationship between the magnetic deviation and its induced current direction error. According to the relationship and the characteristic model of the magnetic deviation, the compensated coefficients are calibrated on the spot and applied to compensate the subsequent current observation. Compared with the current direction without the magnetic deviation affection, the measured current direction accuracy evaluated by the root mean square error improves remarkably. From the experimental result, the approach is proved to be effective for correcting the fixed-point current direction measurement of ADCPs, which are affected by the magnetic deviation and in the small tilt condition.     

The importance of ADCP alignment with GPS in moving-boat streamflow measurements

This paper addresses the importance of the alignment of an acoustic Doppler current profiler (ADCP) with a global positioning system (GPS) in moving-boat streamflow measurements. It presents a mathematical analysis of the discharge bias induced by a misalignment angle. A small misalignment angle may cause a significant bias in a transect discharge. The bias consists of non-directional and direction components. The directional bias is proportional to the ratio between boat velocity and water velocity. In a normal condition of ADCP streamflow measurements, however, the directional bias in transect discharges can be approximately cancelled in the average discharge of reciprocal transects, even if heading-dependent errors are involved in the misalignment. This paper also presents a trial-and-error method for estimating the misalignment angle. We analyzed the transect discharge data obtained from a field measurement on the Yangtze River at the Huangling Temple hydrology station located about 5 km downstream from the Three Gorges Dam to gain insights on the effect of misalignment. Results of this case study suggest that the data for transect discharges must be processed to remove directional bias prior to the Type A evaluation of the random uncertainty of the measured discharge. Otherwise, the estimated Type A uncertainty would be false and misleading.     

A new interpolation scheme for detiding vessel-mounted ADCP data in tidal reach of Yangtze Estuary

Spatiotemporal fitting by the least squares method is commonly applied to separate the mean flow (runoff) and tidal current from vessel-mounted ADCP data in tidal reach. To analyze this technique in an estuary region with interaction of the runoff and tide, three sets of 29-h periods vessel-mounted ADCP data in Yangtze Estuary is tested. A diverse set of basis functions is studied and a nodes determination method, named gradient algorithm, is proposed for comparison purposes. The Green function together with a nodal configuration determined by gradient algorithm is the best option. In general, the semi-major axis of the principal tidal ellipse (M2) is parallel to the riverbank and the phase of shoal waters is ahead of that of thalweg, in keeping with shallow-water tide wave dynamics. Because the tidal currents in Yangtze Estuary are explained by shallow-water tide wave dynamics, the use of the Green function and gradient algorithm in the spatiotemporal fitting by least squares technique is a promising scheme for detiding vessel-mounted ADCP data in shallow-water tide wave dynamics systems.     

动态卡尔曼滤波在导航试验状态估计中的应用

阐述了GPS动态试验的新方案,使用两个精度相差一个数量级的GPS接收平台,通过匀速运动车辆的DGPS及GPS的滤波对比试验,验证了卡尔曼滤波器的有效性.并针对传统EKF(extended Kalman filtering)滤波器动态滤波性能较差的缺陷,引入了一种基于非线性思想的动态无导数卡尔曼滤波器,并对其状态方差阵及随机噪声方差阵Cholesky分解更新公式做了改进,避免了导数的运算,加快了滤波速度,有效地确保方差矩阵平方根的正定性从而抑止了发散.将这种新的卡尔曼滤波器应用于实际动态定位状态估计问题上.试验结果表明:比起传统卡尔曼滤波器,新的卡尔曼滤波器有较高的精度,实用性更强.     

Effects of non-homogeneous flow on ADCP data processing in a hydroturbine forebay

Accurate modeling of the velocity field in the forebay of a hydroelectric power station is important for both power generation and fish passage, and is able to be increasingly well represented by computational fluid dynamics (CFD) simulations. Acoustic Doppler Current Profiler (ADCP) are investigated herein as a method of validating the numerical flow solutions, particularly in observed and calculated regions of non-homogeneous flow velocity. By using a numerical model of an ADCP operating in a velocity field calculated using CFD, the errors due to the spatial variation of the flow velocity are quantified. The numerical model of the ADCP is referred to herein as a Virtual ADCP (VADCP).Two applications of the VADCP are modeled in the numerical analyses presented. Firstly the virtual measurement error of the VADCP is calculated for a single instrument adjacent to the short converging intake of a powerhouse. Secondly, the flow discharge through the forebay is estimated from a transect of VADCP instruments at different distances from the powerhouse. The influence of instrument location and orientation are investigated for both cases.A velocity error of up to 94% of the reference velocity is calculated for a VADCP modeled adjacent to an operating intake and is shown to decrease with distance from the powerhouse. Qualitative agreement is observed between the calculated VADCP velocities and reference velocities by a horizontal offset distance of 18 m upstream of the powerhouse.     

关于走航式ADCP测流盲区计算问题的探讨

就普遍关心的ADCP测流盲区问题,介绍了其成因和计算公式,并结合实例,对走航式ADCP在实际测流过程中的盲区及盲区面积的计算,进行了较为深入地探讨。