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.     

基于图像识别的机械振动信号特征提取与寿命预测方法研究

滚动轴承作为许多机械设备的关键组件,被广泛应用于机械制造、航空航天等领域,其健康状态直接影响了相应设备的剩余寿命,因此在设备故障预测与健康管理(Prognostics and Health Management, PHM)领域,滚动轴承寿命预测具有很高的研究价值。目前基于数据驱动的轴承寿命预测方法主要利用特征提取并构造健康因子(Health Indicator, HI),然而在这一过程中特征的选择与融合依然依赖于专家先验知识,并且健康因子也很难从复杂的时序数据中进行提取。因此,提出了一种新型的数据驱动寿命预测算法,在特征提取方面,通过连续小波变换(Continuous Wavelet Transform,CWT)将传感器振动信号转换为时频谱图,再通过深度残差网络(Deep residual network, ResNet)结合时空卷积网络(Temporal Convolutional Network, TCN)将时频谱图中的时域频域特征构造成为健康因子,最后完成剩余寿命预测。本研究在PRONOSTIA数据集上与现有的数据驱动算法进行了对比,证明了该算法可以更准确地完成剩余寿命预测。     

基于Hilbert-Huang变换的航天器发射段力学环境分析

文章基于Hilbert-Huang变换,对某载人航天器发射过程中整流罩分离期间测量的力学环境参数进行处理分析,利用HHT的基函数自适应性特点,得到振动时间历程信号的时频域Hilbert谱,准确识别出信号主频率及对应时间范围。此应用案例可为地面力学试验条件制定以及运载接口条件确定提供参考。     

Hybrid integration method for highly maneuvering radar target detection based on a Markov motion model

To detect highly maneuvering radar targets in low signal-to-noise ratio conditions, a hybrid long-time integration method is proposed, which combines Radon-Fourier Transform (RFT), Dynamic Programming (DP), and Binary Integration (BI), named RFT-DP-BI. A Markov model with unified range-velocity quantification is formulated to describe the maneuvering target’s motion. Based on this model, long-time hybrid integration is performed. Firstly, the whole integration time is divided into multiple time segments and coherent integration is performed in each segment via RFT. Secondly, non-coherent integration is performed in all segments via DP. Thirdly, 2/4 binary integration is performed to further improve the detection performance. Finally, the detection results are exported together with target range and velocity trajectories. The proposed method can perform the long-time integration of highly maneuvering targets with arbitrary forms of motion. Additionally, it has a low computational cost that is linear to the integration time. Both simulated and real radar data demonstrate that it offers good detection and estimation performances.     

Multi-faults diagnosis of rolling bearings via adaptive customization of flexible analytical wavelet bases

Multi-faults detection is a challenge for rolling bearings due to the mode mixture and coupling of multiple fault features, as well as its easy burying in the complex, non-stationary structural vibrations and strong background noises. In this paper, a method based on the flexible analytical wavelet transform (FAWT) possessing fractional scaling and translation factors is proposed to identify multiple faults occurred in different components of rolling bearings. During the route of the proposed method, the proper FAWT bases are constructed via genetic optimization algorithm (GA) based on maximizing the spectral correlated kurtosis (SCK) which is firstly presented and proved to be efficient and effective in indicating interested fault mode. Via using the customized FAWT bases for each interested fault mode, the original vibration measurements are decomposed into fine frequency subbands, and the sensitive subband which enhances the signal-to-noise ratio (SNR) is selected to exhibit the fault signature on its envelope spectrum. The proposed method is tested via simulated signals, and applied to analyze the experimental vibration measurements from the running roller bearings subjected to outrace, inner-race and roller defects. The analysis results validate the effectiveness of the proposed method in identifying multi-faults occurred in different components of rolling bearings.     

基于小波神经网络PID的永磁同步电机转速控制

提出了基于小波神经网络PID的永磁同步电机(PMSM)转速控制策略。根据系统运行参数的变化,采用三层前馈式人工神经网络,基于梯度下降纠正误差法在线训练实时更新PID参数值。采用小波神经网络和增量式PID共同构成转速环控制器。建立PMSM数学模型,设计PMSM速度环控制器,构建S函数,对控制算法进行仿真试验,验证了该控制算法的先进性。试验结果表明,所提控制策略比传统PID转速控制具有更好的动态性能和抗干扰能力。     

Real-time mosaicking for infrared videos from an oblique sweeping camera

Image mosaicking is widely used in Geographic Information Systems (GISs) for large-scale ground surface analysis. However, most existing mosaicking methods can only be used in offline processing due to the enormous amounts of computation. In this paper, we propose a novel and practical algorithm for real-time infrared video mosaicking. To achieve this, a novel fast template matching algorithm based on Sum of Cosine Differences (SCD) is proposed to coarsely match the sequential images. The high speed of the proposed template matching algorithm is obtained by computing correlation with Fast Fourier Transform (FFT). We also propose a novel fast Least Squares Matching (LSM) algorithm for inter-frame fine registration, which can significantly reduce the computation without degrading the matching accuracy. In addition, the proposed fast LSM can effectively adapt for noise degradation and geometric distortion. Based on the proposed fast template matching algorithm and fine registration algorithm, we develop a practical real-time mosaicking approach which can produce seamless mosaic image highly efficiently. Experiments on synthetic and real-world datasets demonstrate that the proposed algorithm is not just computationally efficient but also robust against various noise distortions.     

飞行器助推段振动环境分析

对飞行器助推段非平稳随机振动信号采用局部平稳模型进行分析,把此段非平稳随机信号的时变谱分解成2个独立的过程:决定时变均方值的确定性过程和决定频率特性的平稳随机过程.利用小波分析对确定性信号和随机信号进行分离,用信号的全部带宽计算非平稳信号的时变均方值,缩短了平均时间,减少了偏置误差;用信号的整个持续时间确定非平稳信号的谱特性,提高了频率分辨率,减少了随机误差.对分离出的平稳随机振动自功率谱密度进行了归纳,得到了飞行器助推段的振动环境.     

基于动力补偿的液压并联运动平台控制策略

针对所研制的飞行模拟器用大负载液压六自由度并联运动平台,在对系统闭环动力学模型进行详细分析的基础上,依据系统数学模型存在外扰力、摩擦力及不确知参数等因素影响的控制特性,利用系统闭环动力学模型的动力补偿特性,采用六维动力补偿器,提出一种PD控制器加小波神经网络补偿器进行在线动力学补偿的实时控制策略,并通过实验验证了其控制的有效性.结果表明:该方法具有良好的跟踪特性,能够提高系统响应快速性、运动精度及抗负载扰动能力,很大程度上克服了系统的动力耦合及参数时变和未知力扰动带来的影响,为多自由度运动系统的高性能实时控制开辟了崭新途径.