3-D vision and range finding techniques

The introduction of third generation robots in automated systems is impeded by the absence of 3-D sensors collecting panoramic range data at medium distance (0–10 meters) in a large volume (up to 100 m³). The work described in the present paper offers a certain number of solutions to this general problem. Our system is built around a 2-D passive machine vision connected to various cameras (VIDICON and CCD). The host computer (HP 1000) pilots numerous sorts of range finders (acoustic and optical).

The concept of multisensory range finders is introduced to allow the best use of each type (active methods). This 3-D vision has been tested in two fields of application:

• -in robotics for the absolute of a mobile robot;

• -in audiovisual for fixing objects or actors in a 3-D synthetic scene.

In some cases the absolute location problem is solved with an opto-electronic remote tracking measurement system. It is the last part of our 3-D machine vision.     

High-precision image position sensing methods suitable for 3-D measurement

A high-precision image position sensing method for a position sensitive detector (PSD) with a kaleidoscopic mirror tunnel (KM-PSM) is introduced. In order to increase the relative resolution of the image position detecting element itself, a hybrid type position sensitive device (R-HPSD) has been invented. A 1-D mark direction sensing method (PM-DSM) which consists of a 1-D image position sensing element, a cylindrical lens and a parallel mirror tunnel is proposed for the realization of a high-precision and miniaturized 3-D position measuring system. Furthermore, an optical range sensing method (RORS), which is suitable for realizing a miniaturized optical range sensing probe based on the triangulation, is proposed.     

Compact three-dimensional camera for robot and vehicle guidance

A very compact 3-D range sensor for robot and vehicle guidance is described. A dual aperture mask in a camera lens is used to transform a small commercially available CCD video camera into a ranging sensor. An active system using a projected laser beam provides a 360° range map which can be used to plan trajectories, either for robots or autonomous vehicles. Good quality panoramic intensity and 3-D images were obtained. Accuracy, speed of acquisition, and cost are discussed. Experimental results are presented.     

Automatic analysis of Young''s fringes in speckle photography and particle-image velocimetry

Laser-speckle photography (LSP) and particle-image velocimetry (PIV) are two closely related optical techniques for the measurement of two-dimensional in-place displacement. LSP is primarily used for the measurement of solid-surface motion, whereas PIV is used in fluid-flow applications. In both cases, data are obtained from the optical-power spectrum of a double-exposure recording of the moving object by measuring the wavelength and orientation of Young's-type fringes corresponding to the displacement magnitude and direction, respectively. Typically, in any single experiment, of the order of 10⁴ fringe patterns must be analysed, and this would be extremely tedious to perform manually. For this reason, a number of schemes have been reported in the technical literature for automated analysis of Young's fringes. In this paper, we examine two techniques, 1-D integration and 2-D autocorrelation, and discuss their application to LSP and PIV. A unified theoretical model is used to show important differences between the two techniques, and resulting implications with respect to analysis procedures are discussed.     

Optical preprocessing for automated inspection

It appears that a simple TV-camera connected to a computer will not produce a very good inspection system. The system is far too ‘anthropomorphic’ and not well enough adapted to the limitations of present-day computers nor to the preprocessing capabilities of present-day optical systems. The taxonomy of preprocessing, utilizing a priori knowledge about an object, is briefly discussed. One particular form of preprocessing is found to be specifically effective—the acquisition of three-dimensional data: 3D data are invariant under changes in illumination. Most often the real concern is over the 3D shape of an object, instead of the 2D distribution of the local reflectivity. Evaluation of 3D data is frequently much easier than that of 2D data and 3D sensors can disply a large dynamical (depth) range. Two new 3D sensor principles, developed at the University of Erlangen, are discussed.     

A generalized iterative approach to curved-ray tomography

A generalized iterative approach to the tomographic reconstruction of strongly refracting fields is proposed. The mappings for existing iterative schemes are recognized to be special cases of a more general form, and this form is shown to possess an arbitrary operator which affects convergence but may be changed without altering the roots of the original mapping. This, therefore, provides the basis for defining new recursive sequences which may converge in cases where the standard iterative schemes are divergent. To illustrate the approach, two enhanced schemes are developed by making particular selections for the arbitrary operator, and a 1-D boundary layer type field is reconstructed from numerically simulated interferometric data.     

淡水湖泊水下未知物体形状的反演

将逆散射问题转化为由远场算子F、测试远场数据f和未知物体边界Γ构成的算子方程F(Γ)=f.利用Levenberg Marquardt方法和A.Kirsh引入的F在Γ上导数,求解该方程,从而确定未知物体边界Γ.     

Ultrabright γ-ray emission from the interaction of an intense laser pulse with a near-critical-density plasma

An efficient scheme for generating ultrabright γ-rays from the interaction of an intense laser pulse with a near-critical-density plasma is studied by using the two-dimensional particle-in-cell simulation including quantum electrodynamic effects. We investigate the effects of target shape on γ-ray generation efficiency using three configurations of the solid foils attached behind the near-critical-density plasma: a flat foil without a channel (target 1), a flat foil with a channel (target 2), and a convex foil with a channel (target 3). When an intense laser propagates in a near-critical-density plasma, a large number of electrons are trapped and accelerated to GeV energy, and emit γ-rays via nonlinear betatron oscillation in the first stage. In the second stage, the accelerated electrons collide with the laser pulse reflected from the foil and emit high-energy, high-density γ-rays via nonlinear Compton scattering. The simulation results show that compared with the other two targets, target 3 affords better focusing of the laser field and electrons, which decreases the divergence angle of γ-photons. Consequently, denser and brighter γ-rays are emitted when target 3 is used. Specifically, a dense γ-ray pulse with a peak brightness of 4.6×10²⁶ photons/s/mm²/mrad²/0.1%BW (at 100 MeV) and 1.8×10²³ photons/s/mm²/mrad²/0.1%BW (at 2 GeV) are obtained at a laser intensity of 8.5×10²² W/cm² when the plasma density is equal to the critical plasma density n_c. In addition, for target 3, the effects of plasma channel length, foil curvature radius, laser polarization, and laser intensity on the γ-ray emission are discussed, and optimal values based on a series of simulations are proposed.     

初始压入位置对Ni基单晶合金纳米压痕影响研究

针对Ni基单晶合金建立初始压入γ 相的γ /γ' 模型和初始压入γ'相的γ'/γ 模型, 采用分子动力学方法模拟金刚石压头压入两种模型的纳米压痕过程, 计算两种模型[001]晶向硬度. 采用中心对称参数分析两种模型(001)相界面错配位错对纳米压痕过程的影响. 结果显示: 弛豫后, 两种模型(001)相界面错配位错形式不同, 其中γ'/γ 模型(001)相界面错配位错以面角位错形式存在; 压入深度在0.930 nm 之前, 两种模型(001)相界面错配位错变化不大, 压入载荷-压入深度及硬度-压入深度曲线较符合; 压入深度在0.930 nm之后, γ'/γ 模型(001)相界面错配位错长大很多, 导致相同压入深度时γ'/γ 模型比γ /γ'模型压入载荷和硬度计算结果小; 压入深度在2.055 nm之后, γ /γ'模型(001)相界面错配位错对γ 相中位错进入γ'相有阻碍作用, 但仍有部分位错越过(001) 相界面进入γ' 相中, γ'/γ 模型(001)相界面处面角位错对γ' 相中位错进入γ 相有更明显的阻碍作用, 几乎无位错越过(001) 相界面进入γ 相中, 面角位错的强化作用更明显, 所以γ'/γ 模型比γ /γ'模型压入载荷上升速度快.     

CsI(Tl)晶体的$lt;i$gt;α$lt;/i$gt;/γ 波形数字化实时甄别

基于CsI（Tl）探测器对α/γ 粒子的波形甄别能力,采用电荷比较法设计了一种波形实时甄别系统. 介绍了实时甄别系统的设计原理,利用⁶⁰Co-γ源、²⁴¹Am-α源对实时系统进行了甄别实验,探究了不同参数对甄别效果的影响,并给出了最优甄别效果下的参数设置. 研究表明,设计的数字化实时波形甄别系统体积小,能准确、实时地甄别开α/γ粒子,最佳品质因子大于1.4,事件计数率可达3× 10⁵/s. 关键词： 数字化波形甄别 实时 电荷比较法 CsI(Tl)晶体     

Phase-and spin-dependent manipulation of leakage of Majorana mode into double quantum dot

We present a phase-and spin-dependent manipulation of leakage of a Majorana mode into a double quantum dot. We study the density of states(DOS) to show the effect of phase change factor on the Majorana leakage into(out) of a double quantum dot. The DOS is derived from the Green's function of the quantum dot by the equation of motion method, and exhibits a formant structure when φ = 0, 2π and a resonance shape when φ = 0.5π and 1.5π. Also, it changes more strongly under the spin-polarized coefficient than the non-polarized lead. Such a theoretical model can be modified to explore the spin-dependent effect in the hybrid Majorana quantum dot system.     

3,22-二羟基何伯烷的单晶结构及波谱学数据

利用三异丁基硼氢化锂（L-selectride）试剂还原3-氧代-22-羟基何伯烷得到单一构型的3α,22-二羟基何伯烷和3β,22-二羟基何伯烷.在此基础上,通过X-射线单晶衍射实验确认了3β,22-二羟基何伯烷的绝对构型,并完整归属了该类化合物的¹H和¹³C NMR数据.本文的方法亦可用以制备其他使用常规方法难以分离的α-和β-异构体.     

外电场辅助化学气相沉积方法制备网格状β-Ga2O3纳米线及其特性研究

利用外电场辅助化学气相沉积（CVD）方法,在蓝宝石衬底上制备出了由三组生长方向构成的网格状β-Ga₂O₃纳米线.研究了不同外加电压大小对β-Ga₂O₃纳米线表面形貌、晶体结构以及光学特性的影响.结果表明：外加电压的大小对样品的表面形貌有着非常大的影响,有外加电场作用时生长的β-Ga₂O₃纳米线取向性开始变好,只出现了由三组不同生长方向构成的网格状β-Ga₂O₃纳米线;并且随着外加电压的增加,纳米线分布变得更加密集、长度明显增长.此外,采用这种外电场辅助的CVD方法可以明显改善样品的结晶和光学质量.     

稠苯芳杂环及其烷基质子化学位移的计算

本文提出了予测稠苯芳杂环及其烷基链上质子化学位移的计算方法。 将稠苯芳杂环化合物用凯库勒式表示,计算式为为需考虑的苯环内的乙烯基效应。σ_mi,ci为各苯环的环流效应。σ_1,Hc为各芳杂环的屏蔽效应,对杂环上质子它就是该单独芳杂环上相应质子的δ值,对苯环上质子要将它分解为各结构因素的效应,即:σ_1,He=(1/2)^d-1δ_x=y(或σ_z)+σ_c-c·σ_m,H. σ_x-y与σ_z为杂原子或其基团的屏蔽效应,σ_c=c为存在于芳杂环中的乙烯基的效应,σ_m,Hc为芳杂环的环流效应,d为对不同质子所考虑的键数。有取代基时需考虑取代基的效应。计算环上烷基质子的公式为:δ=σ_p,CH3+ασ_c,CH3+βσ_t,CH3+σ_l,G σ_l,G为稠苯芳杂环基的某级效应。     

1,2-环己二酮基态光解离反应的理论研究

通过密度泛函理论计算研究了1, 2-环己二酮(α-CHD)基态光解离反应的势能面. 通过IRC方法确定了5个主要的反应通道, 相应的产物分别为P₁(c-C₅H₈O+CO), P₂(2C₂H₄+2CO), P₃(CH₂CHCH₂CH₂CHO+CO), P₄(2CH₂CO+C₂H₄)和P₅(CH₃CHCO+CH₂CHCHO). 获得了反应过程中反应物、产物、中间体和过渡态的结构参数. 详细阐述了这些通道的反应过程, 分析了其反应机理, 总结出最优的反应路径为α-CHD→c-C₅H₈O+CO. 理论分析与实验结果相符. 获得的结果为进一步进行与1, 2-环己二酮相关的研究提供有价值的信息.     

Device topological thermal management of β-Ga_2O_3 Schottky barrier diodes

The ultra-wide bandgap semiconductor β gallium oxide(β-Ga_2 O_3) gives promise to low conduction loss and high power for electronic devices. However, due to the natural poor thermal conductivity of β-Ga_2 O_3, their power devices suffer from serious self-heating effect. To overcome this problem, we emphasize on the effect of device structure on peak temperature in β-Ga_2 O_3 Schottky barrier diodes(SBDs) using TCAD simulation and experiment. The SBD topologies including crystal orientation of β-Ga_2 O_3, work function of Schottky metal, anode area, and thickness, were simulated in TCAD, showing that the thickness of β-Ga_2 O_3 plays a key role in reducing the peak temperature of diodes. Hence, we fabricated β-Ga_2 O_3 SBDs with three different thickness epitaxial layers and five different thickness substrates. The surface temperature of the diodes was measured using an infrared thermal imaging camera. The experimental results are consistent with the simulation results. Thus, our results provide a new thermal management strategy for high power β-Ga_2 O_3 diode.     

Laser Doppler velocimetry measurement at a hard-to-reach intake port of a two-stroke engine

To understand better the complex scavenging process in ported two-stroke engines, optical set-ups were designed for velocity measurements at the exit of an intake port of a motoring and firing single-cylinder propane-fueled two-stroke engine by laser Doppler velocimetry. The radial velocity component was measured at the center of one port at engine speeds of 600, 900, and 1200 rpm. Laser beams entered the engine from the top through a quartz window and the light scattered by the seed particles was collected at 90° through the exhaust ports and a side window. The noise produced by the piston head was analyzed and separated from the signal generated by the seed particles. Analysis of the results from this engine showed that, in general, differences in both magnitude and the shape of the plotted results were observed when the engine was firing. A backflow into the intake system was observed at 600 rpm; this backflow decreased in strength, shifted in crank angle at 900 rpm, and eventually was eliminated at an engine speed of 1200 rpm.     

Enhancement of subgap conductance in a graphene superconductor junction by valley polarization

We theoretically study the differential conductance of a graphene/graphene superconductor junction, where the valley polarization of Dirac electrons is considered in the nonsuperconducting region. It is shown that the subgap conductance will increase monotonically with the valley-polarization strength when the chemical potential μ is near the Dirac point μ≤ 3?(? is the superconducting gap), whereas it will decrease monotonically when μ is far away from the Dirac point, μ≥ 5?.The former case is induced by the specular Andreev reflection while the retro-reflection accounts for the later result. Our findings may shed light on the control of conductance of a graphene superconductor junction by valley polarization.     

多孔介质中非等温互溶流体驱替过程的格子Boltzmann研究

利用格子Boltzmann方法,在孔隙尺度上对多孔介质内非等温混溶驱替过程进行数值研究,定量分析温度粘性膨胀系数（β_T）以及路易斯数（Le）对界面不稳定性和驱替效率的影响.结果表明：随着β_T的增大,界面不稳定性增强,驱替效率降低.当β_T>0时,随着Le的增大,界面不稳定性减弱,驱替流体与被驱替流体之间的界面趋于平缓,指尖残余率减小,驱替效率增大.当β_T<0时,Le对于驱替效率的影响相反.     

自驱动Janus微球近壁运动特性实验与数值模拟研究

自驱动Janus微球是形状规则但表面构成不同的特殊活性颗粒. 针对微米级Pt-SiO₂型Janus 微球近壁面自驱动现象, 实验测得了微球的自驱动速度V_Janus, 并观察到微球运动过程中与垂直方向存在一偏转仰角ψ, 且ψ角随H₂O₂溶液浓度的增大呈减小趋势. 在此基础上, 建立自驱动Janus微球的数值模型, 通过模拟得到了微球在不同浓度H₂O₂溶液中的偏转仰角ψ及距底面的高度δ, 模拟与实验一致. 利用这些数据进一步讨论了壁面效应对微球旋转特征时间τ_R的影响. 这一工作对于理解Janus 微球的运动机理及发展相关应用具有重要意义.