结合边缘特征先验引导的深度卷积显著性检测

针对当前基于深度学习的显著性检测算法缺少利用先验特征和边缘信息，且在复杂场景中难以检测出鲁棒性强的显著性区域的问题，提出了一种结合边缘特征，利用先验信息引导的全卷积神经网络显著性检测算法。该算法利用三种被经常用到的先验知识结合边缘信息形成先验图，通过注意力机制将提取的先验特征与深度特征有效融合，最终通过提出的循环卷积反馈优化策略迭代地学习改进显著性区域，从而产生更可靠的最终显著图预测。经过实验定性定量分析，对比证明了算法的可靠性。     

神经网络全局势函数在多相催化中的应用

当今的多相催化研究需要新的技术和方法从原子尺度上表征活性中心结构和反应中间体。本文作者课题组近期开发了理论模拟新技术来探索催化剂活性位点结构，即基于神经网络势函数的大规模原子模拟（LASP）软件中实现的全局神经网络势函数计算方法。本文介绍了该方法可以显著降低催化体系的计算代价，而维持与密度泛函理论同一级别的计算精度，从而解决多相催化中的许多复杂问题。本文对神经网络势函数方法的实现细节和目前已实现的应用场景进行了详细介绍。神经网络势函数可以用来预测材料晶体结构，理解高压氢化条件下TiO₂表面的结构演化和确定三元氧化物ZnCrO晶相中合成气制甲醇活性位点。最后文章分析了神经网络势函数的局限性和今后可能的三个研究方向，即材料性质预测、多元素体系神经网络势函数构造和化学反应拟合。     

计算机辅助汉字书写教学技术研究综述

传统的汉字书写教学方法具有一定的局限性，为此人们开始关注计算机辅助的汉字书写教学技术。它主要研究汉字书写规范的智能化、自动化评判方法以及可视化的反馈形式。对近20年来计算机辅助汉字书写教学技术的发展情况进行了综述；详细介绍了数字化展示、书写评判、水平评测等几个研究分支的特点、关键问题以及主要成果；结合相关技术的发展和需求的变化，提出了汉字书写数字化教学技术在未来的发展方向。     

多约束下距离加权最优滑模末制导律

为解决多约束下制导炮弹的精确制导问题,采用带有相对距离权函数的最优滑模末制导律,将权函数引入到最优制导律中,通过改变制导炮弹的运动轨迹、运动时间,进而增强制导精度。针对单权函数难以同时满足制导精度与导引头视线角、过载等约束的问题,采用不同权函数的分段加权方法解决加权最优末制导引起的制导问题。结合滑模变结构控制理论,设计分段加权最优滑模末制导律,增强制导系统的抗干扰能力。仿真验证结果表明,该末制导律既能解决过载、导引头视线角、落角等多约束情况下的精确制导问题,同时又具有一定的鲁棒性。     

Reflecting on the preparation of guidelines for strategic environmental assessment (SEA) of nuclear power programmes

In this paper, the ‘guidelines for strategic environmental assessment (SEA) of nuclear power programmes’ by the International Atomic Energy Agency are introduced. This includes a reflection on their preparation process and contents as well as consultation feedback. The preparation process started with two meetings of international nuclear and SEA experts and the creation of a writing team which prepared an initial set of draft guidelines. This was followed by various consultation exercises. The guidelines are organised along an allocation of tasks within a tiered system of energy related policies, plans, programmes and projects. Whilst consultation showed that there was agreement on the approach to most issues, no consensus was present on the extent to which economic and social issues should be fully integrated with environmental considerations. Strong support was given to the way quality review is designed, going beyond focusing on the main SEA reports to cover procedural and participatory aspects next to elements of a comprehensively tiered decision making framework, the ability to influence decisions as well as the quality (expertise and experience) of those involved in conducting the SEA.

Abbreviation: SEA: Strategic Environmental Assessment; EIA: Environmental Impact Assessment; IAEA: International Atomic Energy Agency     

基于Kinect的自适应跑步机行走速度估计方法

自适应跑步机是康复医疗和人体工程学的研究热点,也是虚拟现实运动输入设备的重要组成部分。针对自适应跑步机上人体位置相对地面几乎无变化,很难通过简单的位置差分获得到人体运动速度的问题,提出了一种适应性广、无标记点、非接触式的行走速度估计方法。将Kinect采集到的人体关节点位置数据通过四元数标定、高斯滤波和三次样条插值处理后,用步长修正算法计算出行走时的步态时空参数。基于步态时空参数估计使用者在自适应跑步机上行走时的速度。在固定速度的跑步机上,通过将速度估计值与所设定的跑步机实际速度进行对比,验证了速度估计算法有效性,该速度估计算法可以适用于自适应跑步机的控制算法开发。     

An enhanced state estimation for linear parabolic PDE systems with mobile sensors

This paper studies an enhanced state estimation problem of distributed parameter processes modeled by a linear parabolic partial differential equation using mobile sensors. The proposed estimation scheme contains a state estimator and the guidance of mobile sensors, where the spatial domain is decomposed into multiple subdomains according to the number of sensors and each sensor is capable of moving within the respective subdomain. The state estimator is desired to make the state estimation error system exponentially stable while providing an performance bound. The mobile sensor guidance is used to enhance the transient performance of the error system. By the Lyapunov direct technique, an integrated design of state estimator and mobile sensor guidance laws is developed in the form of bilinear matrix inequalities (BMIs) to meet the desired design objectives. Moreover, to make the performance bound as small as possible, a suboptimal enhanced state estimation problem is formulated as a BMI optimization one, which can be solved via an iterative linear matrix inequality algorithm. Finally, numerical simulations are given to show the effectiveness of the proposed method.     

考虑行走弹性影响下的车辆换挡过程分析

为研究行走弹性对工程车辆行走换挡过程的影响，建立了考虑驱动轮轮胎变形引起的车辆行走弹性影响的整车行走动力学模型，该模型适用于换挡过程分析。计算与分析了典型装载机动力换挡过程，将考虑或不考虑行走弹性时计算得到的整车加速度和瞬时冲击度的变化过程分别进行了对比，利用实车试验对单一换挡过程进行了验证；并将考虑行走弹性时升挡过程中变速箱输入输出转速变化过程、离合器滑摩功率特性的计算结果与不考虑行走弹性时换挡过程的计算结果进行了对比。研究结果表明：与不考虑行走弹性的情况相比，行走弹性变形使得换挡过程中整车加速度的变化更加平缓；换入挡离合器有两次滑摩过程，冲击度更小，换挡时间更长；离合器接合时的冲击度与其接入时间有关；滑摩功更小。     

Exploring the biomechanics of walking and crowd “flow”

Crowd movement simulation models are generally based on aggregated speed and flow data collected more than 50 years ago. There appears to be no validated modelling capability to include the impact of recent and future changes in population demographics, resulting from an ageing population and increasing obesity rates. New analytical approaches and data gathering are required to successfully model crowd movement and safety for current and future generations. This study carried out (a) a review of the primary components of crowd movement, demographics and analytical techniques, (b) prototype experiments to investigate age-related aspects of space and potential points of contact and (c) a new predictive model for crowd flow analysis based on pedestrian biomechanics and anthropometric data. The model uses the physical space taken up by the biomechanical walking process and the spatial buffer between points of potential contact with other pedestrians to predict the speed of movement at different levels of congestion. The new analytical model was used to predict single file speeds (for people with different demographics in congested space), which compared well with published experimental data. The next steps for model development for wider “flows” and additional experiments to provide data sets for wider demographics are also proposed.     

Nonlinear model-based track guidance of user-defined points at the vehicle front

Preventive pedestrian protection systems are validated by means of fully automated driving tests reproducing safety-critical traffic situations on a proving ground. In order to assess these preventive safety systems, a precise and reproducible collision of a pedestrian dummy with a specific point at the vehicle front, e.g., the left corner of the vehicle, must be ensured. Hence, a track guidance of this specific point is required. Beyond the state of the art a new nonlinear model describing the lateral deviation of any point at the vehicle front to a predefined path is proposed in this paper. Based on this model the method of input–output linearization is used to design a flexible lateral guidance system for an easy application in different vehicles. Furthermore, the closed-loop stability is proven and experimental results are presented.