基于时标分解的弹性高超声速飞行器智能控制

考虑弹性高超声速飞行器纵向动力学模型，提出了一种基于时标分解的智能控制方法。考虑刚体状态和弹性模态具有不同的时标特性，采用奇异摄动理论进行快慢时标分解，将模型转换为刚体慢变子系统和弹性快变子系统。针对刚体子系统考虑动力学不确定，基于平行估计模型构造表征不确定逼近效果的预测误差，结合跟踪误差给出复合学习控制策略。针对弹性子系统设计自适应滑模控制稳定弹性模态。通过李雅普诺夫稳定性分析可证系统状态一致终值有界。仿真表明所提出的控制方法能够实现刚弹模态的稳定收敛，且具有更高的跟踪精度、更好的学习性能和更快的收敛速度。     

Quantitative assessment and visualization of flight risk induced by coupled multi-factor under icing conditions

Aircraft icing has been proven to be one of the most serious threats to flight safety. During the analysis of flight risk under icing conditions, quantitative assessment and visualization of flight risk are quite essential as they provide safe manipulation strategies in intricate conditions. However, they are rarely studied. Since the icing flight accidents are the result of the coupling of multiple unfavorable factors, in present study, we have proposed a method to quantitatively assess flight risk induced by multi-factor coupling under icing conditions by Monte-Carlo simulation and multivariate extreme value theory. The results demonstrate that the flight risk probability increases with the rise of unfavorable factors. Besides, a flight risk visualization method named flight safety window has been presented to build the flight risk distribution cloud maps in different complex conditions. The cloud maps show that the icing would give rise to atrophy of the safety scope, and the consequence would be even more severe when coupled with other more unfavorable factors. The proposed methods in this study would be useful in flight risk analysis under icing conditions and can enhance the pilot's situational awareness in selecting correct strategies within the safety zone to avoid unsafe manipulation.     

基于函数链神经网络的深度分类器

目前的宽度学习系统（Broad learning system，BLS）通过所建立的一系列映射节点和增强节点来形成联合节点。因为联合节点与输出层的线性连接，网络权值可以用求解伪逆的方法快速求得，避免了耗时的训练过程，从而成为快速而高效的学习方法。然而在追求高精度结果的过程中，BLS对于增强节点数量的需求过于巨大，容易造成过拟合问题。为此，本文提出了基于函数链神经网络（Functional-link neural network，FLNN）的深度分类器（FLNN based deep classifier，FLNNDC），旨在提供一种更加简单却又不失精度的BLS变体结构。FLNNDC将几个轻量级的BLS子系统堆积成栈式结构，每一个轻量级的BLS子系统随机选择一部分映射节点生成增强节点，而不是全部映射节点。和原宽度结构相比，在几个主流数据集上的实验结果表明本文所提出的FLNNDC分类器具有网络结构更小且学习速度更快的优势。     

无人平台复杂地形探测的视触融合方法

为提高无人平台在复杂环境中的地形探测能力以及解决在小样本数据下识别地形困难的问题，提出了一种无人平台复杂地形探测的视触融合方法。在原始宽度学习的基础上，建立了多模态级联特征节点宽度学习框架。首先进行触觉和视觉初步特征提取和融合特征提取，随后将融合特征矩阵经宽度学习分类器得到地形识别的结果。最后，在自建的视觉-触觉地形 (V-T2)数据集进行了实验验证。结果表明，相比于传统的融合算法，提出的融合算法有很好的准确性和鲁棒性，为无人平台地形探测提供了有效的策略。     

基于深度学习的空间站舱内服务机器人视觉跟踪

为提升舱内跟随服务机器人的任务辅助能力,解决机器人对航天员的视觉跟踪问题,提出了一种基于深度学习和概率模型的人体视觉跟踪算法。利用深度卷积神经网络实现了对穿着多样、姿态任意人体的稳定检测。结合人体检测结果,设计了运动预测概率模型,实现了对指定人员准确、连续的跟踪。算法对包含大多数航天员活动的多个数据集进行了验证。实验结果表明:提出的跟踪算法实现了对穿着多样、姿态任意人体的稳定跟踪,并有效避免了由于穿着相似、遮挡可能造成的误跟踪问题。该算法为空间站舱内跟随服务机器人对航天员的视觉跟踪提供了有效的解决方法。算法基于融合的RGB-D图像,工程上易于构建和实现,也可拓展到其他跟随服务机器人视觉跟踪任务中。     

翻转课堂融入线上教学的应用探索——以“Photoshop图像处理”课程为例

在新的教学环境下,开展翻转课堂教学面临着巨大的挑战。结合教学实践,分析实施翻转课堂教学存在的问题,在教学设计、教学形式和教学评价等方面提出相应的对策。大部分学生对翻转课堂内容及形式总体满意,但课程设计方面存在一些不足,需在教学实践中不断完善。     

Propulsionless planar phasing of multiple satellites using deep reinforcement learning

This work creates a framework for solving highly non-linear satellite formation control problems by using model-free policy optimisation deep reinforcement learning (DRL) methods. This work considers, believed to be for the first time, DRL methods, such as advantage actor-critic method (A2C) and proximal policy optimisation (PPO), to solve the example satellite formation problem of propellantless planar phasing of multiple satellites. Three degree-of-freedom simulations, including a novel surrogate propagation model, are used to train the deep reinforcement learning agents. During training, the agents actuated their motion through cross-sectional area changes which altered the environmental accelerations acting on them. The DRL framework designed in this work successfully coordinated three spacecraft to achieve a propellantless planar phasing manoeuvre. This work has created a DRL framework that can be used to solve complex satellite formation flying problems, such as planar phasing of multiple satellites and in doing so provides key insights into achieving optimal and robust formation control using reinforcement learning.     

基于DDPG算法的无人机集群追击任务

无人机的集群化应用技术是近年来的研究热点，随着无人机自主智能的不断提高，无人机集群技术必将成为未来无人机发展的主要趋势之一。针对无人机集群协同执行对敌方来袭目标的追击任务，构建了典型的任务场景，基于深度确定性策略梯度网络（DDPG）算法，设计了一种引导型回报函数有效解决了深度强化学习在长周期任务下的稀疏回报问题，通过引入基于滑动平均值的软更新策略减少了DDPG算法中Eval网络和Target网络在训练过程中的参数震荡，提高了算法的训练效率。仿真结果表明，训练完成后的无人机集群能够较好地执行对敌方来袭目标的追击任务，任务成功率达到95%。可以说无人机集群技术作为一种全新概念的作战模式在军事领域具有潜在的应用价值，人工智能算法在无人机集群的自主决策智能化发展方向上具有一定的应用前景。     

Modeling irregular small bodies gravity field via extreme learning machines and Bayesian optimization

Close proximity operations around small bodies are extremely challenging due to their uncertain dynamical environment. Autonomous guidance and navigation around small bodies require fast and accurate modeling of the gravitational field for potential on-board computation. In this paper, we investigate a model-based, data-driven approach to compute and predict the gravitational acceleration around irregular small bodies. More specifically, we employ Extreme Learning Machine (ELM) theories to design, train and validate Single-Layer Feedforward Networks (SLFN) capable of learning the relationship between the spacecraft position and the gravitational acceleration. ELM-base neural networks are trained without iterative tuning therefore dramatically reducing the training time. Analysis of performance in constant density models for asteroid 25143 Itokawa and comet 67/P Churyumov-Gerasimenko show that ELM-based SLFN are able learn the desired functional relationship both globally and in selected localized areas near the surface. The latter results in a robust neural algorithm for on-board, real-time calculation of the gravity field needed for guidance and control in close-proximity operations near the asteroid surface.     

卷积神经网络和峭度在轴承故障诊断中的应用

针对传统智能诊断方法依靠专家知识和人工提取数据特征工作量大的问题,结合深度学习方法在特征提取和处理大数据方面的优势,研究了一种基于卷积神经网络和振动信号峭度指标的滚动轴承故障诊断方法。该方法将深度学习应用于轴承故障诊断,提取滚动轴承正常状态、内圈故障、外圈故障和滚动体故障4种状态的振动信号,将振动信号分段处理得到峭度指标,使用数据到图像的转换方法将峭度指标转换为灰度图,送入卷积神经网络模型完成故障分类。在进行滚动轴承故障诊断的实验时,所提的模型诊断准确率达到99.5%,高于传统支持向量机(SVM)算法的95.8%。