基于产学研模式的飞机机务维修实训平台建设

目的探索构建校内生产性实训基地;方法以飞机结构腐蚀与控制实训平台的建设为例;结果阐述校内实现以产学研相结合模式来进行实训平台建设;结论对其它机务维修专业课程的实训平台建设起到引导的作用。     

Interface Familiarity Restores Active Advantage in a Virtual Exploration and Reconstruction Task in Children

Abstract

Active exploration is reportedly better than passive observation of spatial displacements in real environments, for the acquisition of relational spatial information, especially by children. However, a previous study using a virtual environment (VE) showed that children in a passive observation condition performed better than actives when asked to reconstruct in reality the environment explored virtually. Active children were unpractised in using the input device, which may have detracted from any active advantage, since input device operation may be regarded as a concurrent task, increasing cognitive load and spatial working memory demands. To examine this possibility, 7–8-year-old children in the present study were given 5 minutes of training with the joystick input device. When compared with passive participants for spatial learning, active participants gave a better performance than passives, placing objects significantly more accurately. The importance of interface training when using VEs for assessment and training was discussed.     

试析马克思理论体系的两个思想资源

除了常说的三大理论来源以外,马克思的理论体系的形成还有很多丰富的思想资源,其中,基督教和《圣经》传统、古希腊神话和西方文学作品对马克思科学世界观、人生观、价值观的形成产生了深刻影响,马克思对他们采取了批判的借鉴的态度,从中汲取了丰富的营养,成为马克思理论体系重要的思想资源。     

MBD环境下计算机辅助零件分类技术

针对MBD环境下的零件分类问题，研究MBD信息提取、零件编码转换和深度学习分类算法技术，并创建MBD-计算机辅助零件分类系统（MBD-CAPC），实现从输入MBD模型到输出零件分类结果的自动化。     

FAST FEATURE RANKING AND ITS APPLICATION TO FACE RECOGNITION

A fast feature ranking algorithm for classification in the presence of high dimensionahty and small sample size is proposed. The basic idea is that the important features force the data points of the same class to maintain their intrinsic neighbor relations, whereas neighboring points of different classes are no longer to stick to one an- other. Applying this assumption, an optimization problem weighting each feature is derived. The algorithm does not involve the dense matrix eigen-decomposition which can be computationally expensive in time. Extensive exper- iments are conducted to validate the significance of selected features using the Yale, Extended YaleB and PIE data- sets. The thorough evaluation shows that, using one-nearest neighbor classifier, the recognition rates using 100-- 500 leading features selected by the algorithm distinctively outperform those with features selected by the baseline feature selection algorithms, while using support vector machine features selected by the algorithm show less prominent improvement. Moreover, the experiments demonstrate that the proposed algorithm is particularly effi- cient for multi-class face recognition problem.     

基于改进YOLO V5的无人机航拍图像小目标检测方法

本文针对无人机航拍目标检测技术中目标聚集、目标较小及实时性差等问题,将YOLO V5的主干架构进行改进,简化Neck网络,提出了一种提高检测速度又能准确识别的无人机航拍图像检测技术方案。经过仿真实验测试,改进后的YOLO V5网络在保持识别精度的同时,检测速率提升了31%,满足无人机在航拍作业时对于准确性与实时性的要求。     

Coactive design of explainable agent-based task planning and deep reinforcement learning for human-UAVs teamwork

Unmanned Aerial Vehicles (UAVs) are useful in dangerous and dynamic tasks such as search-and-rescue, forest surveillance, and anti-terrorist operations. These tasks can be solved better through the collaboration of multiple UAVs under human supervision. However, it is still difficult for human to monitor, understand, predict and control the behaviors of the UAVs due to the task complexity as well as the black-box machine learning and planning algorithms being used. In this paper, the coactive design method is adopted to analyze the cognitive capabilities required for the tasks and design the interdependencies among the heterogeneous teammates of UAVs or human for coherent collaboration. Then, an agent-based task planner is proposed to automatically decompose a complex task into a sequence of explainable subtasks under constrains of resources, execution time, social rules and costs. Besides, a deep reinforcement learning approach is designed for the UAVs to learn optimal policies of a flocking behavior and a path planner that are easy for the human operator to understand and control. Finally, a mixed-initiative action selection mechanism is used to evaluate the learned policies as well as the human’s decisions. Experimental results demonstrate the effectiveness of the proposed methods.     

基于SE模块改进Xception的动物种类识别

Xception是Inception网络的一种极端化表现,在与Inception v3参数相近的情况下,它能够达到更高的准确度。由于神经网络提取的特征不一定都是有用特征,因此以Xception为基础,将SE(Squeeze and Excitation）模块加入该网络,调整特征通道的权重,使得网络的精确度得到提高。通过实验,融合SE模块的Xception网络训练精确度分别在Oxford-IIIT Pet数据集和CUB_200_2011数据集上提升了1%~1.7%和0.8%~1%,证明了SE模块能够进一步提升Xception的精确度。将改进后的Xception应用到动物种类识别中,根据精确度曲线对实验策略调整改进,最终在测试集上获得95.63%的识别率。     

基于自适应代理模型的结构动力可靠度分析方法

结构动力可靠度分析对于衡量结构的安全状态和工作性能具有重要的指导意义。采用一种同时考虑 动态激励载荷和材料力学性能参数随机性的动力可靠度计算高效方法,通过建立不确定性材料力学性能参数 与条件动力可靠度之间的 Kriging代理模型,推导动力可靠度的解析表达式;提出一个新的学习函数来构建自 适应的 Kriging代理过程,通过复合材料飞行器蒙皮骨架结构在随机振动载荷和随机材料力学性能参数共同作 用下的动力可靠度分析对所提方法的准确性与高效性进行验证。结果表明:所提方法不仅能够避免使用代理 模型对动力可靠度求解时由于使用数值积分所引入的误差,同时能够使用尽可能少的计算成本获得准确的动 力可靠度结果。     

一类非线性系统的一阶D型迭代学习控制

传统的D型迭代学习控制的控制律设计方案依赖于被控系统的相对度.为解决该问题以及相对度增益与高阶微分运算的问题,针对一类具有任意高阶相对度的非线性系统,提出了基于虚拟模型的一阶D型迭代学习控制设计方法.该方法的主要思想是与具有任意高阶相对度的非线性被控系统并联一个一阶子系统,构造一个相对度为1、相对度增益可以任意设计的虚拟模型,在此基础上设计一个一阶D型迭代学习控制律,使得虚拟模型能够实现期望轨迹的完全跟踪,从而实际被控系统在一定误差范围内实现期望轨迹的跟踪.仿真实例验证了所提方法的可行性与有效性.