Human-tracking strategies for a six-legged rescue robot based on distance and view

Human tracking is an important issue for intelligent robotic control and can be used in many scenarios, such as robotic services and human-robot cooperation. Most of current human-tracking methods are targeted for mobile/tracked robots, but few of them can be used for legged robots. Two novel human-tracking strategies, view priority strategy and distance priority strategy, are proposed specially for legged robots, which enable them to track humans in various complex terrains. View priority strategy focuses on keeping humans in its view angle arrange with priority, while its counterpart, distance priority strategy, focuses on keeping human at a reasonable distance with priority. To evaluate these strategies, two indexes(average and minimum tracking capability) are defined. With the help of these indexes, the view priority strategy shows advantages compared with distance priority strategy. The optimization is done in terms of these indexes, which let the robot has maximum tracking capability. The simulation results show that the robot can track humans with different curves like square, circular, sine and screw paths. Two novel control strategies are proposed which specially concerning legged robot characteristics to solve human tracking problems more efficiently in rescue circumstances.     

Nondestructive testing and characterization of residual stress field using an ultrasonic method

To address the difficulty in testing and calibrating the stress gradient in the depth direction of mechanical components, a new technology of nondestructive testing and characterization of the residual stress gradient field by ultrasonic method is proposed based on acoustoelasticity theory. By carrying out theoretical analysis, the sensitivity coefficients of different types of ultrasonic are obtained by taking the low carbon steel(12%C) as a research object. By fixing the interval distance between sending and receiving transducers, the mathematical expressions of the change of stress and the variation of time are established. To design one sending-one receiving and oblique incidence ultrasonic detection probes, according to Snell law, the critically refracted longitudinal wave (LCR wave) is excited at a certain depth of the fixed distance of the tested components. Then, the relationship between the depth of LCR wave detection and the center frequency of the probe in Q235 steel is obtained through experimental study. To detect the stress gradient in the depth direction, a stress gradient LCR wave detection model is established, through which the stress gradient formula is derived by the relationship between center frequency and detecting depth. A C-shaped stress specimen of Q235 steel is designedto conduct stress loading tests, and the stress is measured with the five group probes at different center frequencies. The accuracy of ultrasonic testing is verified by X-ray stress analyzer. The stress value of each specific depth is calculated using the stress gradient formula. Accordingly, the ultrasonic characterization of residual stress field is realized. Characterization results show that the stress gradient distribution is consistent with the simulation in ANSYS. The new technology can be widely applied in the detection of the residual stress gradient field caused by mechanical processing, such as welding and shot peening.     

超声电机直驱的电动物镜控制方法

针对超声电机直接驱动的显微镜物镜为满足系统整体结构紧凑、快速和准确定位的需求,采用小型光电耦合器设计了位置检测传感器,配合遮挡片实现了物镜位置和物镜转换器运动方向的检测。采用传统比例-积分-微分控制(proportion-integral-derivative control,简称PID)和模糊控制提出了并行切换控制策略,将两者结合建立了宏微相融合的预测控制方法,实现了物镜转换器的高精度、快速切换控制。在嵌入式控制系统中实现了所设计的控制算法和策略,并进行了物镜自动切换控制实验。实验结果显示,采用笔者提出的定位结构和控制方法,电动物镜转换器的重复定位误差小于0.015°,定位时间小于3s,满足自动显微镜系统中对电动物镜转换器重复定位精度的要求。     

FRFT在转子起停车状态评估中的应用研究进展

转子起停车过程中由传统的等时间间隔采样所获得的机械动态信号已不再保持原有的周期特性,采用经典的基于傅里叶变换的谱分析方法进行信号处理不再有效。针对转子起停车状态评估中幅值、频率的非平稳特性所带来的问题,引入分数阶傅里叶变换研究了基于起停车信息的故障定性分析和定量识别方法,提出了基于分数阶全息原理的转子起停车故障特征提取方法,并在此基础上实现了基于分数阶主分量原理的转子故障特征模式定量分类。实验验证表明,该方法解决了转子起停车过程中瞬变信息的获取问题,能有效提取出起停车过程振动信号中的典型故障特征,可以实现对转子不同故障类型及不同故障程度的准确分类。     

船载高精度星敏感器安装角的标定

星敏感器测量船体姿态精度与星敏感器与甲板之间的安装角标定精度密切相关。本文介绍了船载星敏感器的相关坐标系及安装角的定义,建立了船载星敏感器蒙气差修正模型,提出了一种船载高精度星敏感器安装角标定方法。船进坞坐墩时,船载经纬仪通过拍摄方位标确定航向,星敏感器通过星图识别获得视场内星点的赤经、赤纬,构成地心惯性系参考矢量,经岁差、章动、极移、船位等修正,得到各恒星在惯导地平系下的参考矢量。然后,根据蒙气差模型对星点逐个修正俯仰角,重构惯导地平系下的参考矢量。最后,依据姿态确定算法原理,解算星敏感器安装矩阵,求解安装角。实验表明,使用该方法可使方位角、俯仰角的标定精度达10"以内。该方法有效发挥了星敏感器指向精度高的优点,改善了程序自动化程度,提高了船体姿态测量精度。     

Effects of Si on the Microstructures and Mechanical Properties of High-Chromium Cast Iron

Effect of Si on the microstructures and mechanical properties of high-chromium cast iron was investigated. The eutectic carbides are refined greatly and a transformation of matrix from austenitic matrix to pearlite is observed with increase in Si content from 0.5 to 1.5 wt.%. The refinement of eutectic microstructure is attributed to the decrease in the eutectic temperature, while the transformation from austenite matrix to pearlite is associated with the increase in solubility of carbon in the matrix. In the pearlite matrix, two types of pearlite are observed: one with lamellar pearlite, distributing at the periphery, and the second one with granular pearlite at the center. The density of secondary carbides precipitated from the matrix increases greatly with addition of Si from 0.5 to 1.5 wt.%, which is associated with more carbon and chromium elements confined in the matrix in the alloy containing 1.5 wt.%. More rod-like particles are observed in the alloy containing 0.5 wt.% Si, while the morphology of secondary carbides of alloy containing 1.5 wt.% is granular. The mechanical properties are improved with a 7% increase in tensile strength from 586 to 626 MPa and impact toughness from 5.8 to 7.3 J cm^?2.     

Effect of Fe on Microstructure and Properties of 8xxx Aluminum Conductor Alloys

The effect of Fe contents (0.3-0.7 wt.%) on the microstructure, electrical conductivity, mechanical and creep properties of 8xxx aluminum conductor alloys was investigated. Results revealed that the as-cast microstructure of 8xxx alloys was consisted of equiaxed α-Al grains and secondary Fe-rich intermetallics distributed in the interdendritic region. The extruded microstructure showed partially recrystallized structure for 0.3% Fe alloy but only dynamically recovered structures for 0.5 and 0.7% Fe alloys. With increasing Fe contents, the ultimate tensile strength and yield strength were remarkably improved, while the electrical conductivity was slightly decreased. Moreover, the creep resistance was greatly improved, which is attributed to the larger volume fraction of fine intermetallic particles and smaller subgrain size in the higher Fe-containing alloys. The creep threshold stress was found to increase from 24.6 to 33.9 MPa with increasing Fe contents from 0.3 to 0.7%, respectively. The true stress exponent values were close to 3 for all three experimental alloys, indicating that the creep mechanism of 8xxx alloys was controlled by dislocation glide.     

基于自校正Kalman滤波的液膜密封端面摩擦状态监测技术

声发射技术是液膜密封端面摩擦状态的有效检测方法,但是受工业背景噪声的影响,难以分离出声发射信号中所需信息。针对此问题,采用基于ARMA模型的自校正Kalman滤波技术处理声发射信号。该滤波器能在系统模型参数和噪声特性未知的情况下,收敛于稳态最优卡尔曼滤波器,因此滤波后的声发射信号的所需特征信号更突出,有利于液膜密封端面摩擦状态的检测。建立RBF神经网络,以时域、频域和时频域特征值作为输入进行网络训练,实现密封端面摩擦状态模式识别。实验结果证明,该监测方法能实时有效地识别端面摩擦状态,识别结果与电涡流直接测量得到的结果一致。     

Optimization of algal lipid extraction by mixture of ethyl acetate and ethanol via response surface methodology for biodiesel production

The effects of extraction time, extraction temperature, solvent to biomass ratio and solvent composition on lipid yield from lyophilized Chlorococcum sp. biomass using a mixture of ethyl acetate and ethanol (EAE), a new proposed solvent, were studied. Subsequently, the process conditions of extraction by EAE were optimized using Box-Behnken design (BBD). The results revealed that the extraction temperature had the greatest effect on lipid extraction efficiency, followed by volume ratio of ethyl acetate to ethanol (EA/E) and extraction time. The largest lipid extraction yield of 15.74% was obtained under the following extraction conditions: 40mL solvents per gram of biomass for 270 min with gentle stirring at 80 °C by EAE with an EA/E of 1.0. Furthermore, palmitic acid, stearic acid, oleic acid, and linoleic acid were the most abundant fatty acids in the lipids extracted, indicating the great potential of the proposed lipid extraction procedure for microalgae-based biodiesel production.