增程式电动汽车车速鲁棒自适应跟踪控制

针对增程式电动汽车行驶时因不确定性路面扰动导致车速出现波动的问题,设计自适应鲁棒车速跟踪控制器,使汽车跟踪理想车速,通过Matlab/Simulink仿真软件验证所设计自适应控制器算法的有效性。仿真结果表明,该控制器可以自适应调节控制增益,对路面环境未知扰动的不确定性具有良好鲁棒性。研究成果可为增程式电动汽车的车速跟踪控制设计提供参考。     

盘山公路转弯处车辆监测系统

在盘山路转弯时,驾驶员经常无法看到前方是否有车辆通过。为此设计了一种盘山路弯道车辆警示系统。该系统由磁通传感器SEN-S65检测弯道两边是否有车辆出现,以单片机PIC16F73为核心构成信号处理电路。弯道两边系统采用RF通信方式,当有车辆通过弯道时,由无线数传模块PTR2000向弯道对面系统发出信息;对面系统接收到信息后,控制警示灯闪烁,向驾驶员发出警示。实测表明该系统能够准确地检测出弯道两边是否有车辆通过,并发出警示信息。     

HPXe ionization chambers for γ spectrometry at room temperature

High pressure xenon (HPXe) ionization chambers exhibit many characteristics which make them particularly suitable for industrial γ spectrometry at room or higher temperature. The use of a gas as detection medium allows one to reach very large effective volumes and makes these chambers relatively insensitive to radiation damage. Further, the high atomic number of xenon (Z=54) enhances the total absorption of incident photons and provides, combined to high pressure, a good enough detection efficiency with respect to solid state detectors. Furthermore, such ionization chambers with Frisch grid appear to be very stable over wide periods (e.g. a research prototype has been used aboard MIR orbital station for several years) and temperature range (up to 180°), without maintenance.The characteristics of different prototypes are presented. Their detection efficiency and energy resolution are studied as a function of incident γ ray energy. New developments in electronics and signal processing are also investigated to improve their performances.     

基于自抗扰控制的水面无人艇路径跟踪控制器

针对欠驱动水面无人艇非线性系统,设计了一种参数在线优化的路径跟踪自抗扰控制器。对包含不确定动态的水面无人艇数学模型和参数化路径引导方法做了简要的说明,并引入一种基于混沌局部搜索策略的双种群遗传算法,对自抗扰控制器的参数进行在线优化。通过不同条件下的仿真试验结果说明,本研究设计的路径跟踪控制器对海洋环境的不确定动态、未知扰动以及模型的参数摄动具备良好的稳定性和鲁棒性。     

基于深度学习的人脸妆容迁移算法

人脸妆容迁移是指将参考妆容迁移到素颜人脸上,在保持面部特征不变的同时尽可能展现参考妆容的风格的一种任务。为了进一步实现人脸妆容自动迁移技术,避免现有妆容迁移方法没有充分考虑人与人之间的五官差异而导致提取的人脸信息不足等问题,提出了一种基于深度卷积神经网络的人脸妆容迁移算法。该算法首先自动定位素颜人脸和参考妆容的五官,提取重要部位的特征信息。然后通过妆容传递网络和损失函数,经过深度卷积神经网络自主训练,最终实现了参考妆容向素颜人脸的自动迁移。仿真实验结果表明,与目前的主流算法进行对比,该算法耗时更短、运算性能更具优势,同时在不改变原图五官细节的基础上,妆容迁移效果更为自然。     

新旧沥青界面融合实测与耗散粒子动力学模拟

为扩充新旧沥青界面再生融合特征的研究手段,多尺度量化剖析新旧沥青界面再生融合速率、融合程度等行为参数,通过动态剪切流变试验,基于细观尺度实测研究新旧沥青界面再生融合规律;通过耗散粒子动力学,基于介观尺度模拟研究新旧沥青界面再生融合机理行为并验证动态剪切流变试验实测结果。分析表明:加热温度、加热时间均与新旧沥青界面再生融合程度呈正相关关系,加热温度与新旧沥青界面再生融合速率呈线性正相关关系,加热时间与新旧沥青界面再生融合速率呈指数负相关关系;加热时间从10~120 min,沥青界面扩散激活能增长了3~5倍;采用新添沥青单一调和的方式再生效果不佳,各试验条件下沥青界面再生融合程度均不足50%,但采用新添沥青与再生剂复合的方式,新旧沥青界面再生融合程度提升2倍左右,并同比降低沥青界面扩散激活能10%~30%;剪切速率可较大幅度影响新旧沥青界面再生融合作用,并且相比于较高加热温度(358~418 K),在低温加热条件下(298~358 K)适当增加剪切速率对改善新旧沥青界面再生融合特征效果更明显;采用的沥青四组分分子结构模型、构建的耗散粒子动力学粗粒化结构模型和选取的模拟计算参数为跨尺度拓展量化表征新旧沥青界面再生融合行为特征与规律提供重要支撑。     

基于模型预测控制的四足机器人斜坡自适应调整算法与实现

为了实现四足机器人在有斜坡地形下的自适应稳定行走,在模型预测控制基础上进行扩展,设计四足机器人在斜坡上的足端位置调整与躯干姿态自适应调整策略。由惯性测量单元(inertial measurement unit,IMU)测得机器人运动时的姿态参数,通过推导的足端轨迹算法,得到足端位置的坐标映射,调整机器人在斜坡上的重心位置;通过设计的“虚拟斜坡”躯干姿态调整算法,实现机器人在上坡过程中躯干姿态的自适应调整。利用实验室的四足机器人物理平台和搭建的实际斜坡地形环境,验证了所提算法的可行性和有效性。机器人平台验证结果表明,所提出的斜坡自适应控制方法提高了机器人在坡面上的稳定裕度,优化了足端运动空间,实现了四足机器人的自适应爬坡调整。     

Effects of geodesic dome trajectories on the specific strength of composite overwrapped pressure vessels: FE modelling

In this study, it was aimed to find out geodesic dome trajectories of composite overwrapped pressure vessels, and investigate the effect of dome profiles on the structural performances. In this context, geodesic paths for 0.2, 0.3, 0.4, 0.5 and 0.6 dimensionless polar opening radii were determined by solving elliptical integrals and filament winding angles were calculated throughout the dome and cylindrical portions. Afterward, finite element analysis was performed to obtain mechanical properties by using the Ansys ACP module. As a result of the current study, it has been concluded that dome profile and filament winding angle are highly dependent on the polar opening radii. When the performance factor was considered, it has been determined that the optimum pressure vessel has 0.6 dimensionless polar opening radii. Moreover, it was observed that minimum equivalent stress, strain, deformation and inverse reverse factors have occurred in the pressure vessel with 0.6 dimensionless polar opening radii. Furthermore, it was showed that effective parameters in the mechanical performance of pressure vessels can be optimized to obtain strengthened and lighter structures.     

Discriminative feature mining hashing for fine-grained image retrieval

With the development of multimedia technology, fine-grained image retrieval has gradually become a new hot topic in computer vision, while its accuracy and speed are limited due to the low discriminative high-dimensional real-valued embedding. To solve this problem, we propose an end-to-end framework named DFMH (Discriminative Feature Mining Hashing), which consists of the DFEM (Discriminative Feature Extracting Module) and SHCM (Semantic Hash Coding Module). Specifically, DFEM explores more discriminative local regions by attention drop and obtains finer local feature expression by attention re-sample. SHCM generates high-quality hash codes by combining the quantization loss and bit balance loss. Validated by extensive experiments and ablation studies, our method consistently outperforms both the state-of-the-art generic retrieval methods as well as fine-grained retrieval methods on three datasets, including CUB Birds, Stanford Dogs and Stanford Cars.     

Simulation study on performance optimization of a prototype scintillation detector for the GRANDProto35 experiment

As a proposed detector, the giant radio array for neutrino detection (GRAND) is primarily designed to dis-cover and study the origin of ultra-high-energy cosmic...