基于YOLOv5s-AntiUAV的反无人机目标检测算法研究

随着无人机的应用领域不断拓展,无人机的“黑飞”给公共安全造成严重损害。为解决侵入式无人机小目标在复杂飞行环境下的错检和漏检问题,提出基于YOLOv5s-AntiUAV的反无人机目标检测算法。首先,引入结合深度超参数卷积的Slim-Neck范式,增强算法特征提取能力并保持计算效率。其次,在骨干和颈部网络引入SPD-Conv模块,提高在低分辨率图像中小目标的检测性能。最后,用Alpha-CIoU替换YOLOv5s算法中的CIoU,增强算法泛用性。YOLOv5s-AntiUAV算法与YOLOv5s、SSD和Faster R-CNN算法在数据集Anti-UAV上的对比实验结果表明,改进算法的mAP@0.5值分别增长了1.1、12.1和4.9个百分点,凸显其实用性。由在VisDrone2019数据集上进行的迁移实验显示,相较于YOLOv5s算法,改进算法mAP@0.5值提升了4.5个百分点,表明其相较于原算法具有更强的鲁棒性。

Synchronous Dynamic Adjusting: An Anti-Collision Algorithm for an RF-UCard System

An RF-UCard system is a contactless smartcard system with multiple chip operating systems and multiple applications. A multi-card collision occurs when more than one card within the reader’s read field and thus lowers the efficiency of the system. This paper presents a novel and enhanced algorithm to solve the multi-card collision problems in an RF-UCard system. The algorithm was originally inspired from framed ALOHA-based anti-collision algorithms applied in RFID systems. To maximize the system efficiency, a synchronous dynamic adjusting (SDA) scheme that adjusts both the frame size in the reader and the response probability in cards is developed and evaluated. Based on some mathematical results derived from the Poisson process and the occupancy problem, the algorithm takes the estimated card quantity and the new arriving cards in the current read cycle into consideration to adjust the frame size for the next read cycle. Also it changes the card response probability according to the request commands sent from the reader. Simulation results show that SDA outperforms other ALOHA-based anti-collision algorithms applied in RFID systems.

Space-borne antenna adaptive anti-jamming method based on gradient-genetic algorithm

A novel space-borne antenna adaptive anti-jamming method based on the genetic algorithm （GA）, which is combined with gradient-like reproduction operators is presented, to search for the best weight for pattern synthesis in radio frequency （RF）. Combined, the GA＇s the capability of the whole searching is, but not limited by selection of the initial parameter, with the gradient algorithm＇s advantage of fast searching. The proposed method requires a smaller sized initial population and lower computational complexity. Therefore, it is flexible to implement this method in the real-time systems. By using the proposed algorithm, the designer can efficiently control both main-lobe shaping and side-lobe level. Simulation results based on the spot survey data show that the algorithm proposed is efficient and feasible.

Anti-aliasing nonstationary signals detecion algorithm based on interpolation in the frequency domain using the short time Fourier transform

To eliminate the aliasing that appeared during the measurement of multi-components nonstationary signals, a novel kind of anti-aliasing algorithm based on the short time Fourier transform （STFT） is brought forward. First the physical essence of aliasing that occurs is analyzed; second the interpolation algorithm model is setup based on the Hamming window; then the fast implementation of the algorithm using the Newton iteration method is given. Using the numerical simulation the feasibility of algorithm is validated. Finally, the electrical circuit experiment shows the practicality of the algorithm in the electrical engineering.

基于GA的磁悬浮球系统的Anti-windup变结构PI控制

磁悬浮球系统是一个开环不稳定的强非线性系统,而且其控制输入-驱动电流必须在限定范围内变化.为了使输出的位移达到期望的位置,采用Anti-windup变结构PI控制器,结合遗传算法优化P、I、α的值,来使小球达到最理想的位置输出.实验结果表明,与常规PI控制器比较,Anti-windup变结构PI控制可以更快速稳定跟踪期望位置,振荡幅度明显减少.

Adaptive Antenna Utilizing Power Inversion and Linearly Constrained Minimum Variance Algorithms

This paper presents a new algorithm based on the power inversion （PI） and the linearly constrained minimum variance （LCMV）. This algorithm is capable of adjusting the weights of the antenna array in real time to respond to and improve the global positioning system （GPS） received signals coming from the desired directions and at the same time to highly suppress the jammers coming from the other directions. The simulation is performed for fixed and moving jammers. It indicates that this structure can give deeper nulls, more than 115 dB depths for fixed jammers and more than 94 dB depths for moving jammers.

OPTIMIZATION OF CORONARY STENT STRUCTURE DESIGN FOR MAXIMIZING THE ANTICOMPRESSION MECHANICAL PROPERTY

Excellent mechanical property of the anti-compression or high collapse pressure has become an essential feature of new coronary stents. How to determine the design parameters of stent becomes the key to improve the stent quality. An integrated approach using radial basis function neural network （RBFNN） and genetic algorithm （GA） for the optimization of anti-compression mechanical property of stent is presented in this paper. First, finite element simulation and RBFNN are used to map the complex non-linear relationship between the collapse pressure and stent design parameters. Then GA is employed with the fitness function based on an RBFNN model for arriving at optimum configuration of the stent by maximizing the collapse pressure. The results of numerical experiment demonstrate that the combination of RBFNN and GA is an effective approach for the mechanical properties optimization of stent.

基于Anti-windup的含作动器饱和主动悬架控制方法

作动器饱和特性直接影响主动悬架输出性能,进而影响汽车操纵稳定性和舒适性。在分析作动器饱和输出特性的基础上,建立含作动器饱和的1/4主动悬架模型。采用基于Anti-windup的PID反馈控制方法,改善作动器饱和输出对系统的不良影响。通过Simulink建立主动悬架模型,分别对仅实施PID控制及增加Anti-windup算法时的系统进行仿真研究。结果表明,含作动器饱和时,PID控制器输出作用力极限值大于作动器输出饱和度时,控制系统不能达到最优状态。采用Anti-windup算法后,能够在作动器饱和度小于输出最大作用力时,提高系统收敛速度。

改进YOLOv7的城市小型无人机目标检测方法

针对“低小动”无人机的反制技术已成为低空空域安全防御的重要手段,然而实时检测与准确识别是实施有效反制的前提条件与关键基础。针对城市低空环境下,目标检测算法对不同背景下小尺度无人机目标检测精度低,容易出现漏检误检且易受外界因素干扰等问题,提出了一种基于改进YOLOv7的“低小动”无人机目标检测方法。首先采集大量不同环境、不同背景下的无人机样本构建数据集,并采用ViBe(visual background extractor)算法进行预处理;其次引入坐标注意力机制与SPDConv(space-to-depth convolution)模块改进和优化YOLOv7的网络结构;最后提出融合ViBe和改进YOLOv7的二级检测架构,将改进后的YOLOv7作为网络模型检测经ViBe处理后的图像。依据原图与处理图像的位置大小关系,将检测出的目标坐标映射回归至原图片,从而完成目标检测提取。实验结果表明,所提目标检测方法检测精度达96.5%,较原YOLOv7方法提高了15.8个百分点,显著提升了“低小动”目标的检测精度,能够满足低空无人机的实时精准检测的需求。

Genetic Searching Algorithm for Optimal Runlength-Limited Codes with Error Control

ＧｅｎｅｔｉｃＳｅａｒｃｈｉｎｇＡｌｇｏｒｉｔｈｍｆｏｒＯｐｔｉｍａｌＲｕｎｌｅｎｇｔｈ┐ＬｉｍｉｔｅｄＣｏｄｅｓｗｉｔｈＥｒｒｏｒＣｏｎｔｒｏｌ＊＊ＴｈｉｓｗｏｒｋｗａｓｓｕｐｐｏｒｔｅｄｂｙｔｈｅＣｈｉｎｅｓｅＫｅｙＰｒｏｊｅｃｔｏｆＦｕｎｄａｍ...

基于空域自适应单脉冲的抗干扰雷达多目标参数估计算法

基于空时级联单脉冲的多目标参数高效估计算法(M-STCMP)是一种高效的雷达多目标参数估计算法,但在干扰环境下其参数估计性能将严重下降。文中提出了基于空域自适应单脉冲的抗干扰多目标参数估计(AM-STCMP)算法。首先,将空域自适应单脉冲技术引入M-STCMP算法,自适应形成和差波束抑制干扰信号,提高各目标的信干噪比,并迭代修正自适应和差波束的鉴角曲线;其次,采用空时级联单脉冲技术估计各目标的距离-速度两维参数;最后,采用Relax算法高精度迭代估计各目标的角度-速度-距离三维参数。理论分析和仿真结果验证了AM-STCMP算法的有效性,其参数估计性能接近无干扰情形。

基于改进Gabor算法的遮挡人脸智能识别方法

为提高有遮挡人脸的识别精度,提出基于改进Gabor算法的遮挡人脸智能识别方法。首先,对人脸图像动态范围压缩,并选择反锐化掩模滤波算法展开图像增强处理;其次,利用Gabor滤波器对信息保留较完整、亮度较高的半边脸进行特征提取;最后将提取到的Gabor特征输入到极限学习机中完成遮挡人脸的智能识别。实验结果表明,所提方法对处理遮挡人脸图像具有良好的效果,且其对人脸图像识别具有精准度高、识别时间短等优点。

抗遮挡的行人多目标跟踪算法

为了解决在复杂场景下行人相互遮挡导致跟踪系统精度降低的问题,提出了基于FairMOT的抗遮挡多目标跟踪算法(multiple obeject tracking algorithm with anti-occlusions,AOMOT)。首先通过轻量化平衡模块,解耦不同层次的语义信息,减少检测任务和重识别任务的语义冲突,降低重识别任务的性能提升对检测任务的影响。其次应用自注意力结构提取行人的外观特征,加强局部窗口下的类内特征的区分度,增强行人身份信息的匹配一致性并减少身份标识的频繁切换。最后优化身份关联算法,挖掘低置信度目标中的被遮挡对象,将其重新纳入目标身份关联并更新其重识别特征。实验结果表明,AOMOT相比原有FairMOT在MOT17数据集中高阶跟踪精度提升1.5百分点,身份F1分数提升3百分点,身份切换数量降低32%。

平行行为的反垄断规制

长期以来,传统寡头市场中大企业协调一致的平行行为的反垄断规制一直困难重重,也由此产生了一系列相互交错的制度工具。这些制度工具包含多种相互关联的基本概念,也反映了反垄断规制对于平行行为的复杂的理论立场。随着算法时代的来临,反垄断法将面对数量更多、更为稳固的平行行为。必须进一步将缺乏合意证据的“协议”、滥用共同支配地位行为、算法合谋等视为一个同源的整体——平行行为,并明确反垄断法相应的制度工具及其发展变化,才能更好地实现对平行行为的规制。

Particle Swarm Optimization Applied to Some Anti-Windup Problems

The particle swarm optimization (PSO) algorithm is introduced to deal with some open anti-windup problems, i.e., determining the initial condition when applying the iterative algorithm to enlarge the estimate of the domain of attraction, determining the design point in the delayed anti-windup scheme, and determining the design point and the weighting factors in the multi-stage anti-windup scheme. Therefore, the corresponding PSO-based algorithms are proposed. Unlike the traditional methods in which the free design parameters can only be selected by trial and error with the available computational results, the PSO-based algorithms provide a systematic way to determine these parameters. In addition, the algorithms are easy to be implemented and are very likely to find the desirable parameters that further improve the anti-windup closed-loop performances. Simulation results are presented to validate the effectiveness and advantages of the proposed method.

基于遗传模糊PID算法的倒立摆稳摆抑噪抗扰实验与分析

倒立摆稳摆控制原理应用的实际工程需求牵引,使倒立摆成为检验各类控制算法的理想替代实验研究平台。因此,将倒立摆摆杆的姿态稳定控制和抑噪抗扰作为研究内容,以理论分析、编程实现、仿真实验和实物实验验证作为研究手段,聚焦控制器设计的优化。首先,引入模糊控制对PID参数调节进行优化,改固定参数调节为在线参数调节,旨在提高系统的鲁棒性与抗扰性;然后,运用遗传算法的量化因子和比例因子完成调节参数的在线寻优,通过设计适应度函数和选择算子,经过选择、交叉、变异不断演化迭代,最终获取最佳适应度个体、找到最优解,通过仿真验证遗传模糊PID算法突出的抑噪抗扰优势;最后,基于该控制策略进行倒立摆实物实验,验证了其指令跟踪有效性的结论。该研究表明:遗传模糊PID算法具有良好的指令跟踪和抑噪抗扰性能,具有一定的工程应用价值。

基于模糊神经网络PID的PMSM速度控制算法研究

针对永磁同步电机(PMSM)控制系统中存在的内部参数难以确定、动态变化适应性差等问题,提出了一种模糊神经网络PID控制算法,实现对PMSM速度的精确控制。首先,该控制算法采用模糊神经网络对PID控制参数进行自适应整定,并利用反向传播(BP)算法动态调整模糊神经网络的参数。然后,利用Matlab/Simulink仿真了不同频率和大小的负载干扰下,该控制算法对PMSM响应速度和控制精度的影响,结果显示速度波动维持在极小范围,最大响应时间小于13ms;最后,在电机实验平台上进行了测试,实验结果表明该控制算法具有较好的响应速度和抗干扰能力,且对正弦波信号具有良好的跟随效果,可以适应不同的工况。

A collision recovery algorithm using optimal Q parameter based on BIBD in RFID

In this work,an optimal Q algorithm based on a collision recovery scheme is presented. Tags use BIBD-( 16,4,1) codes instead of RN16 s. Therefore,readers can make a valid recognition even in collision slots. A way of getting the optimal slot-count parameter is studied and an optimal Q algorithm is proposed. The theoretical and simulation results show that the proposed algorithm can improve reading efficiency by 100% more than the conventional Q algorithm. Moreover,the proposed scheme changes little to the existing standard. Thus,it is easy to implement and compatible with ISO 18000-6C.

数控压机伺服控制系统复合控制器I-ABC与PID优化

为了提高数控压机伺服控制系统的控制精度,针对伺服控制系统运行控制过程构建数学模型,在人工蜂群算法基础上融入了云分析模型,之后采用改进人工蜂群算法(Improved Artificial Bee Colony,I-ABC)调节比例-积分-微分(Proportional Integral Differential,PID)参数,建立了一种复合控制方法。研究结果表明:以I-ABC进行PID控制时,可以使幅度差降低到0.4%,相位差基本在-0.54°之内,系统加载精度也获得了明显提升,表现出了更优的跟踪性能。以I-ABC进行PID控制时,能够对多余力起到明显抑制作用,响应速度也获得明显提升,可以有效满足系统的准确控制要求。在系统内加入干扰信号,引入I-ABC实施PID调节可以减小系统超调量,还可以获得更短调节时间,使系统获得更强抗干扰性能。