Visual object tracking via online sparse instance learning

Sparse representation has been attracting much more attention in visual tracking. However most sparse representation based trackers only focus on how to model the target appearance and do not consider the learning of sparse representation when the training samples are imprecise, and hence may drift or fail in the challenging scene. In this paper, we present a novel online tracking algorithm. The tracker integrates the online multiple instance learning into the recent sparse representation scheme. For tracking, the integrated sparse representation combining texture, intensity and local spatial information is proposed to model the target. This representation takes both occlusion and appearance change into account. Then, an efficient online learning approach is proposed to select the most distinguishable features to separate the target from the background samples. In addition, the sparse representation is dynamically updated online with respect to the current context. Both qualitative and quantitative evaluations on challenging benchmark video sequences demonstrate that the proposed tracking algorithm performs favorably against several state-of-the-art methods.     

Analytical Stability Bound for a Class of Delayed Fractional-Order Dynamic Systems

Delayed Linear Time-Invariant (LTI) fractional-order dynamic systems areconsidered. The analytical stability bound is obtained by using Lambertfunction. Two examples are presented to illustrate the obtainedanalytical results.     

Modeling spherical explosions with aluminized energetic materials

This paper deals with the numerical solution and validation of a reactive flow model dedicated to the study of spherical explosions with an aluminized energetic material. Situations related to air blast as well as underwater explosions are examined. Such situations involve multiscale phenomena associated with the detonation reaction zone, the aluminium reaction zone, the shock propagation distance and the bubble oscillation period. A detonation tracking method is developed in order to avoid the detonation structure computation. An ALE formulation is combined to the detonation tracking method in order to solve the material interface between detonation products and the environment as well as shock propagation. The model and the algorithm are then validated over a wide range of spherical explosions involving several types of explosives, both in air and liquid water environment. Large-scale experiments have been done in order to determine the blast wave effects with explosive compositions of variable aluminium content. In all situations the agreement between computed and experimental results is very good.     

无人动力伞带跟踪微分器的高度控制方法

为全面改善无人动力伞飞行高度控制性能,提出一种带跟踪微分器的积分先行控制(TD-IPD)方法。以线性二次型最优控制引出的积分先行控制器为基础,考虑到微分环节的干扰问题,引入跟踪微分器将信号从干扰中分离出来,并用最速控制综合函数fsun(·)减小达到稳态时所用步数;由于积分环节易饱和,采用饱和补偿方法对积分环节结构进行修改,从而提高积分环节抗饱和能力;以某动力伞的高度控制为对象,通过Matlab仿真验证了所提方法的有效性。     

Diversity of photonic differentiators based on flexible demodulation of phase signals

We theoretically propose a multifunctional photonic differentiation （DIFF） scheme based on phase demodulation using two cascaded linear filters. The photonic D1FF has a diversity of output forms, such as the 1 st order intensity DIFF, the 1 st order field DIFF and its inversion, and the 2nd-order field DIFF, depending on the relative shift between the optical carrier and the filter＇s resonant notches. As a proof, we also experimentally demonstrate the DIFF diversity using a phase modulator and two delay interferometers （Dis）. The calculated average deviation is less than 7% for all DIFF waveforms. Our schemes show the advantages of flexible DIFF functions and forms, which may have different optical applications. For example, high order field differentiators can be used to generate complex temporal waveforms, and intensity differentiators are useful for the ultra-wideband pulse generation.     

High-bandwidth tracking control of piezo-actuated nanopositioning stages using closed-loop input shaper

An integrated control strategy for piezo-actuated nanopositioning stages is proposed in this paper. The aim is to achieve high-speed and high-precision tracking control of nanopositioning stages. For this purpose, a direct inverse compensation method is firstly applied to eliminate the hysteresis nonlinearity without involving inverse model calculation. Then, an inside-the-loop input shaper is designed to suppress the vibration of the compensated system. A Smith predictor is introduced to prevent the potential closed-loop instability caused by the time delay of the inside-the-loop input shaper. Finally, a high-gain feedback controller is employed to handle the disturbances and modeling errors. To demonstrate the effectiveness of the proposed control method, comparative experiments are carried out on a piezoelectric actuated stage. The results show that the proposed control approach increases the tracking bandwidth of the stage from 22.6 Hz to 510 Hz.     

LTE位置管理中信令开销的分析

为了克服2G和3G移动通信网络位置管理方案的缺陷,4G长期演进(Long Term Evolution,LTE)采用了基于跟踪区列表(Tracking Area List,TAL)的位置管理方案。基于TAL的位置管理方案的性能取决于TAL分配方案。考虑到本地移动设备(User Equipment,UE)的活动区域相对固定,该文提出一种嵌入式马尔科夫链模型,用于分析本地UE的基于TAL的位置管理方案的信令开销。推导得到位置更新开销和寻呼开销的数学公式。利用这些公式,可搜索得到能使信令开销最低化的最佳TAL分配方案。     

利用模糊理论对单脉冲雷达天线副瓣的自动识别及跟踪

本文讨论了根据单脉冲雷达的特点及跟踪过程，运用模糊综合评判、模糊数据库、模糊控制的原理对单脉冲雷达系统的天线进行副瓣自动识别及跟踪的方法，经过仿地具试验初步验证该方法是可行的。     

光学PICK UP技术概要

光学ＰＩＣＫ ＵＰ是ＣＤ－ＲＯＭ驱动器的关键部件，它能够把光盘上的信息凹坑和非凹坑的长度和跳变沿转换成电信号，并产生聚集误差信号和道跟踪误差信号。文中简单介绍了光学ＰＩＣＫ ＵＰ的工作原理，结构和有关技术。     

雷达目标特性数据自动采集技术研究

本文讨论了一种用于雷达目标识别系统的雷达目标自动数据采集系统,该采集系统包括数据采集电路和目标检测跟踪电路,在系统软件的控制下,采集电路根据目标检测跟踪电路得到的目标空间位置信息在观测空域对需采样的目标进行开窗筛选、拾取目标视频回波数据。采集系统可以自动地对多个目标连续采集,形成带有数据标志的回波数据流,由此建立目标原始回波特性数据库供分析、处理、识别。