Real-time object tracking via least squares transformation in spatial and Fourier domains for unmanned aerial vehicles

This paper addresses the problem of real-time object tracking for unmanned aerial vehicles. We consider the task of object tracking as a classification problem. Training a good classifier always needs a huge number of samples, which is always time-consuming and not suitable for realtime applications. In this paper, we transform the large-scale least-squares problem in the spatial domain to a series of small-scale least-squares problems with constraints in the Fourier domain using the correlation filter technique. Then, this problem is efficiently solved by two stages. In the first stage, a fast method based on recursive least squares is used to solve the correlation filter problem without constraints in the Fourier domain. In the second stage, a weight matrix is constructed to prune the solution attained in the first stage to approach the constraints in the spatial domain. Then, the pruned classifier is used for tracking. To evaluate proposed tracker's performance, comprehensive experiments are conducted on challenging aerial sequences in the UAV123 dataset. Experimental results demonstrate that proposed approach achieves a state-ofthe-art tracking performance in aerial sequences and operates at a mean speed of beyond 40 frames/s. For further analysis of proposed tracker's robustness, extensive experiments are also performed on recent benchmarks OTB50, OTB100, and VOT2016.     

Machining of a film-cooling hole in a single-crystal superalloy by high-speed electrochemical discharge drilling

Single-crystal superalloys are typical advanced materials used for manufacturing aeroengine turbine blades. Their unique characteristics of high hardness and strength make them exceedingly difficult to machine. However, a key structure of a turbine blade, the film-cooling hole,needs to be machined in a single-crystal superalloy; such machining is challenging, especially considering the increasing levels of machining efficiency and quality demanded by the aeroengine industry. Tube electrode high-speed electrochemical discharge drilling(TSECDD), a hybrid technique of high-speed electrical discharge drilling and electrochemical machining, provides high machining efficiency and accuracy, as well as eliminating the recast layer. In this study, TSECDD is used to machine a film-cooling hole in a nickel-based single-crystal superalloy(DD6). The Taguchi methods of experiment are used to optimise the machining parameters. Experimental results show that TSECDD can effectively drill the film-cooling hole; the optimum parameters that give the best performance are as follows: pulse duration: 12 ls, pulse interval: 30 ls, peak current:6 A, and salt solution conductivity: 3 m S/cm. Finally, a hole is machined by TSECDD, and the results are compared with those obtained by electrical discharge machining. TSECDD is found to be promising for improving the surface quality and eliminating the recast layer.     

基于Kalman滤波的变体飞行器T-S模糊控制

针对变体飞行器的跟踪控制问题,提出了一种基于Kalman滤波的T-S模糊控制方法。考虑飞行器系统状态不可测,引入惯导数据作为辅助信息,利用Kalman滤波算法融合飞控信息与惯导信息实现状态估计。由于变体飞行器在不同变形结构下气动特性变化较大,为便于控制器设计,采用小扰动线性化方法得到飞行器在不同平衡点处的局部线性模型,并通过状态反馈方法设计局部控制器,局部线性模型和局部控制器通过模糊集和模糊规则聚合成一个连续光滑的全局T-S模糊模型和T-S模糊控制器。通过综合Kalman滤波器与T-S模糊控制器得到一个基于Kalman滤波的T-S模糊控制器。仿真结果表明,该控制器在变形过程中能够实现状态估计,保证飞机的跟踪性能。     

UKF方法在脉冲星自主导航中的应用

针对X射线脉冲星自主导航中的非线性系统滤波问题,将无迹卡尔曼滤波方法应用于自主导航计算过程中。首先,在简述脉冲星导航可行性的基础上,研究了脉冲星自主导航系统的基本原理和实施方案。然后,根据牛顿二体引力模型构建了航天器运动状态方程,根据脉冲到达时间模型建立了系统观测方程,并对二者进行了误差分析和建模。最后,将UKF算法应用于航天器自主导航过程中,仿真结果表明该方法能够实现航天器自主导航信息的解算。     

一种小天体旋转参数估计方法

小天体旋转参数是科学数据,也是小天体测绘,着陆导航的基础数据.研究一种在小天体探测接近段过程中使用的基于图像上特征点跟踪和扩展卡尔曼滤波器的小天体旋转参数估计方法.该方法首先建立小天体旋转关系模型,表示小天体在相机坐标系中的姿态变化;然后定义小天体旋转的状态方程,推到了扩展卡尔曼滤波器的计算过程.通过对观测图像序列上的特征点跟踪,利用扩展卡尔曼滤波器方法得到小天体旋转轴指向和自转角速度估计.实验中,在仿真相机与小天体相机100 km距离上,分析了相机坐标系小天体坐标系之间的四元数初值,图像上特征点跟踪精度,相机的观测指向等因素对小天体旋转参数估计精度的影响.实验结果表明,提出的基于图像特征点跟踪和扩展卡尔曼滤波器的小天体旋转参数估计方法能够得到具有较高精度的估计结果.     

面向室内场景的惯性磁感应定位方法

针对卫星拒止的室内导航问题,在现有磁信标定位技术的基础上,通过分析低频旋转磁场的特点,建立了一种不受传感器姿态和磁信标磁矩信息影响的磁感应矢量夹角观测模型,并结合MEMS惯性导航的误差模型提出了一种以磁感应矢量夹角的正余弦值为量测信息的惯性磁感应动态定位方法,避免了磁传感器姿态误差和磁矩误差对定位结果的影响。利用实验室的双轴磁信标实验系统进行静态测试,验证了磁感应定位模型具有不受传感器姿态、环境中遮挡物以及磁矩影响的特点;通过数值仿真的方式对基于无迹卡尔曼滤波(UKF)的惯性磁感应定位方法进行验证,结果表明该方法能有效地抑制惯性导航的发散,且位置最大误差小于0.75 m,满足大多数室内场景下的高精度导航定位需求。     

叶片磨抛机器人力/位混合控制的设计与实现

为提高航空类发动机叶片的自动化磨抛精度,减小复杂曲面叶片加工轨迹控制误差,采用基于六维力传感器的机器人力/位混合控制策略,实现机器人磨抛轨迹的在线修正。搭建以Staubli机器人和ATI六维力传感器为核心部件的叶片磨抛验证平台,通过C++开发上位机,采集磨抛过程中六维力传感器信息并进行Kalman滤波。通过示教确定机器人运动轨迹,对机器人运动轨迹与力传感器信息进行采集分析,确定基于力/位混合控制可以实现机器人运动轨迹的在线修正,为复杂曲面的叶片磨抛轨迹控制提供一种解决方案。     

利用增量相位和星光仰角增强XNAV

针对X射线脉冲星导航系统(XNAV)中过程噪声统计特性难以准确获取,对其不当假设导致滤波器估计性能不佳的问题,提出基于自适应差分卡尔曼滤波器(ADDF)的多信息融合算法。为了降低导航误差,在传统脉冲星计时观测的基础上,增加恒星星光仰角及两个时刻间的相位增量观测量,共同增强XNAV。首先,分别建立计时观测模型、相位增量模型及星光仰角模型;然后将多信息测量模型集成到卫星轨道动力学方程中,以建立ADDF滤波模型;最后对所提方法进行仿真验证。实验结果表明,在相同的初始状态和初始噪声误差条件下,与传统X射线脉冲星导航算法相比,多信息融合算法能将导航位置估计精度提高70%以上,位置估计误差降低到 200 m 左右,速度估计精度提高40%以上,且ADDF性能优于无迹卡尔曼滤波器。     

基于ICKF的永磁同步电机无传感器控制方法

针对在永磁同步电机(Permanent magnet synchronous motor,PMSM)中安装传感器带来的高成本、体积增大、可靠性降低和易受环境干扰等问题,提出采用迭代容积卡尔曼滤波(Iterative cubature Kalman filter,ICKF)算法来估计电机转速和转子位置,并将其应用于永磁同步电机无传感器控制。首先,建立了PMSM在α-β坐标系下的离散数学模型。其次,在Matlab/Simulink环境下分别建立了基于容积卡尔曼滤波(Cubature Kalman filter,CKF)和ICKF的PMSM转速、电流双闭环的无传感器矢量控制系统仿真模型,并进行了给定转速和加负载两种工况的仿真验证。最后,基于TMS320F28335芯片搭建了硬件实验验证平台。仿真分析与实验结果均表明,迭代容积卡尔曼滤波算法应用于永磁同步电机无传感器控制中抗负载变化干扰性好、电机运行稳定、电机转速和转子位置估计精度高,可满足对电机精确控制要求较高的应用场合,具有重要参考价值和推广意义。