基于Mecanum轮配送车的视觉与磁带双重导引技术研究

针对卫星总装过程中提出的物料配送需求,设计并实现了以Mecanum轮全方位移动平台为基础,可在狭窄配送通道内快速、灵活移动的智能物料配送车。结合总装现场实际环境条件,配送车采用了视觉导引与磁带导引相结合的智能导引策略,有效消除了光照强度变化、地面裂纹、电磁信号干扰等环境因素对车道线采集的影响,最终实现总装物料从库房至多工位的自动配送。     

Numerical investigation of flow mechanisms of tandem impeller inside a centrifugal compressor

This study numerically investigated a single stage centrifugal compressor ‘‘Radiver’’ with a wedge diffuser and several tandem-designed impellers to explore the flow phenomena within the tandem impeller and the potential to enhance compressor performance. The results demonstrate that tandem design and clocking fraction (λ_s) significantly affects the compressor performance. The compressor stage with tandem impellers of Series A of boundary layer growth interruption alone are observed to have a widely operating range but efficiency and total pressure ratio penalty compared with that of conventional impeller. The tandem impeller with at least the same impeller efficiency as the conventional design is considered as a critical design criteria so that further modification process based on the flow characteristic of tandem impeller is necessary. In order to restrain the inducer wake and exducer shock losses, parameters modification of blade angle and thickness distributions are necessary and the modified tandem impeller of Series B is obtained. The modified tandem impeller with 25% clocking arrangement shows an 8.45% stall margin increase and maintains the total pressure ratio and efficiency as the conventional design, which proves the potential of tandem impeller to improve compressor stage performance. It is noteworthy that the tandem impellers of Radiver have not shown obviously balanced exit flow field and the fundamental mechanism of stall margin extending of tandem impeller lies on the improved impeller/diffuser matching performance resulting from the incidence angle variation at diffuser inlet.     

Visual inspection of aircraft skin: Automated pixel-level defect detection by instance segmentation

Skin defect inspection is one of the most significant tasks in the conventional process of aircraft inspection. This paper proposes a vision-based method of pixel-level defect detection, which is based on the Mask Scoring R-CNN. First, an attention mechanism and a feature fusion module are introduced, to improve feature representation. Second, a new classifier head—consisting of four convolutional layers and a fully connected layer—is proposed, to reduce the influence of information around the area of the defect. Third, to evaluate the proposed method, a dataset of aircraft skin defects was constructed, containing 276 images with a resolution of 960 × 720 pixels. Experimental results show that the proposed classifier head improves the detection and segmentation accuracy, for aircraft skin defect inspection, more effectively than the attention mechanism and feature fusion module. Compared with the Mask R-CNN and Mask Scoring R-CNN, the proposed method increased the segmentation precision by approximately 21% and 19.59%, respectively. These results demonstrate that the proposed method performs favorably against the other two methods of pixel-level aircraft skin defect detection.     

结构光三维视觉系统研究

介绍了结构光三维视觉原理,在广义坐标系中采用投影和透视变换理论建立了结构光三维视觉模型。在此基础上,提出了基于非线性最小二乘法的模型参数标定方法,并建立了结构光三维视觉系统,给出了实验结果。     

基于自联想网络的发动机传感器解析余度技术

本文提出了一种基于自联想神经网络的传感器解析余度技术。在这种网络中,冗余传感器的信息被压缩、重组进入网络的第一部分,网络的第二部分将压缩信息恢复出来。基于数据融合原理,若一个传感器发生故障,其它传感器仍可提供足够的信息代替发生故障的传感器。理论分析和用于涡轴发动机的仿真结果表明,这种特殊结构的自联想网络具有良好的过滤噪声和故障信号的作用,特别适合于用作不易建模的复杂对象的传感器信号重构     

线性模型跟踪控制系统的模型匹配设计

研究了用完全模型匹配方法求解模型跟踪控制的问题。对通常的模型匹配解法进行了扩展,并所据此引入广义最小方差控制思想,提出一种模型跟踪和误差的新方法。最后,对该方法在惯性导航系统中的应用进行了仿真研究。验证了其可行性和有效性。     

贝叶斯景象匹配技术

景象匹配技术以其极高的末制导精度在导航系统中起着关键性的作用。经典的景象匹配定位方法只能提供象素级的定位精度,为了获得更高的精度可采用ＦＦＴ匹配算法、最小二乘匹配算法等。该文首次提出了一种新的贝叶斯景象匹配算法。其基本思想在于：匹配定位位置可看作为随机量,先利用匹配图像间的信息求其条件概率分布,再利用图像中的大量样本来求解其贝叶斯估计值。同时研究了基于贝叶斯估计的景象匹配技术的原理及算法,并将其与     

交互外形设计中的有理插值方法

在CAGD中,构造光滑曲线通过给定的二维或三维点集是非常重要的问题。广泛釆用的方法是Hermite技术,但是依照这种方法,插值曲线的形状过多地依赖于切矢的模长,且形状的修改不能依几何的方式交互完成。本文提出了一种结合交互技术和插值技术于一体的曲线插值方法,插值基函数为有理函数,在此基础上讨论了形状参数在几何修形中的作用,     

超声钻削加工工具更换问题的探讨

介绍了适用超声钻削加工的半波长振动系统结构、针对该振动系统,给出了工具更换后变幅杆与工具之间进行阻抗匹配的方法,依据各部分的边界条件,按一维问题,导出了加工中工具更换后的阻抗匹配数学模型,在超声钻削加工应用中具有一定的实用价值     

图像序列中机动目标三维运动和结构的计算

综合视觉运动分析中的2类处理方法,选取图像中的角点作为特征点,在理论上证明了图像序列的光流场可以近似地用角点的位移场代替。利用已有文献中的建模思想,详细推导出递归计算机动目标三维运动和结构的非线性计算模型,采用广义卡尔曼滤波(EKF)递归地计算图像序列中机动目标的三维运动和结构。合成图像序列和真实图像序列实验结果表明该算法能取得较好的效果。