A Real Time Self-Tuning Motion Controller for Mobile Robot Systems

This paper proposes an intelligent controller for motion control of robotic systems to obtain high precision tracking without the need for a real-time trial and error method. In addition, a new self-tuning algorithm has been developed based on both the ant colony algorithm and a fuzzy system for real-time tuning of controller parameters. Simulations and experiments using a real robot have been addressed to demonstrate the success of the proposed controller and validate the theoretical analysis. Obtained results confirm that the proposed controller ensures robust performance in the presence of disturbances and parametric uncertainties without the need for adjustment of control law parameters by a trial and error method.      

基于堆叠深度卷积沙漏网络的步态识别

提出了一种基于堆叠深度卷积沙漏网络的步态识别方法。为了解决人体建模中关节点准确定位的问题，采用基于深度卷积的沙漏网络来提取步态图上的关节点坐标，并计算肘关节与膝关节的角度作为运动特征。为了解决行走速度变化带来的影响，采用动态时间规整（Dynamic Time Warping）对特征序列进行距离计算。通过最近邻分类器对结果进行准确分类。该方法在公共CASIA-B数据集与TUM-GAID数据集上进行了验证并与其他方法进行比较，结果表明该方法有较高的识别率。     

基于NB分类器重访概率预测的Web缓存替换策略

Web缓存是用来解决网络访问延迟和网络拥塞问题，缓存替换策略直接影响缓存的命中率。为此，提出一种朴素贝叶斯（NB）分类器重访概率预测的Web缓存替换策略；根据用户之前访问日志，通过分区操作提取多项特征来表示每次访问的对象，并构建特征数据集；训练NB分类器，用来确定缓存中对象被再次访问的概率，为对象分配权重；结合LRU策略来合理删除一些对象。仿真结果表明，提出的策略在保证较高命中率的同时有效降低了执行时间。     

基于注意力机制与文本信息的用户关系抽取

随着社交媒体的发展,用户之间的关系网络对于社交媒体的分析有很大的帮助。因此,该文主要研究用户好友关系检测。以往的关于用户好友关系抽取的研究主要基于社交媒体上的结构化信息,比如其他好友关系,用户的不同属性等。但是,很多时候用户本身并没有大量的好友信息存在,同时也不一定有很多确定的属性。因此,我们希望基于用户发表的文本信息来对用户关系进行预测。不同于以往的潜在好友推荐算法,该文提出了一种基于注意力机制以及长短时记忆网络(long short-term memory,LSTM)的好友关系预测模型,将好友之间的评论分开处理,通过分析用户之间的评论来判断是否具备一定的好友关系。该模型将好友双方信息拼接后的结果作为输入,并将注意力机制应用于LSTM的输出。实验表明,用户之间的评论对于好友关系预测确实有较大的实际意义,该文提出的模型较之于多个基准系统的效果,取得了明显的提升。在不加入任何其它非文本特征的情况下,实验结果的准确率达到了77%。     

Evaluation of physiological workload assessment methods using heart rate and accelerometry for a smart wearable system

Work metabolism (WM) can be accurately estimated by oxygen consumption (VO₂), which is commonly assessed by heart rate (HR) in field studies. However, the VO₂–HR relationship is influenced by individual capacity and activity characteristics. The purpose of this study was to evaluate three models for estimating WM compared with indirect calorimetry, during simulated work activities. The techniques were: the HR-Flex model; HR branched model, combining HR with hip-worn accelerometers (ACC); and HR?+?arm-leg ACC model, combining HR with wrist- and thigh-worn ACC. Twelve participants performed five simulated work activities and three submaximal tests. The HR?+?arm-leg ACC model had the overall best performance with limits of agreement (LoA) of ?3.94 and 2.00?mL/min/kg, while the HR-Flex model had ?5.01 and 5.36?mL/min/kg and the branched model, ?6.71 and 1.52?mL/min/kg. In conclusion, the HR?+?arm-leg ACC model should, when feasible, be preferred in wearable systems for WM estimation.

Practitioner Summary: Work with high energy demand can impair employees’ health and life quality. Three models were evaluated for estimating work metabolism during simulated tasks. The model combining heart rate, wrist- and thigh-worn accelerometers showed the best accuracy. This is, when feasible, suggested for wearable systems to assess work metabolism.     

Participatory design of a robot for demonstrating an epileptic seizure

ABSTRACT

We present the methodology and results of participatory design of a robot for presenting an epileptic seizure and a scenario of the educational workshop using this robot. Children with epilepsy encounter stigma and stereotypes and may receive inadequate aid when having an epileptic seizure. The goal of the larger project was to use the prototype device in a series of workshops for improving teachers' actions during an epileptic seizure and their attitudes towards epileptic students. In this paper, we show how various design goals for an educational robot were accomplished to fit the needs of all identified stakeholders, particularly people with epilepsy.

We used a co-design (participatory design) approach through a series of meetings participated by members of the association Polish Association for People Suffering from Epilepsy, students and faculty members of the biomedical engineering and robotics departments, teachers, psychologists and medical specialists (epileptologist, neurologist).

These meetings created an opportunity for a better understanding of the (functional and nonfunctional) requirements and resulting tradeoffs and led the participants to find appropriate solutions. Participation of people with epilepsy in the design process allowed them to deal with the potentially stereotyped representation of themselves. The prototype robot, therefore, combined goals of various stakeholders, such as an accurate presentation of an epileptic seizure, lightweight, ease of use and control, while preserving the dignity of people with epilepsy.

As a result of the co-designing process, an inexpensive robot was created and used in a series of 10 pilot workshops with 217 participants, mainly teachers of primary and middle schools. Teachers improved their understanding of epilepsy and suggested further improvements to the system.     

A tripartite approach to operator-error evaluation in ballast water management system operation

This paper presents a new methodology developed to quantitatively analyze and prioritize the contributions of Human Factors (HFs) in the human-machine-interaction (HMI) within a complex sociotechnical system such as a Ballast Water Management (BWM) System. The methodology is a combination of the Human Factor Analysis Classification System (HFACS), Analytic Hierarchy Process (AHP), and a modified version of the Theory of Inventive Problem Solving (TRIZ) known as the Radial Dynamics Model (RDM). The methodology (HFACS-AHP-RDM) is based on a five-step algorithm, with which data from experts’ judgment was analyzed. A human-error and system risk minimization hierarchy was subsequently proposed to improve human performance and minimize the likelihood of an unwanted event such as the discharge of harmful aquatic organisms and pathogens (HAOPs). The result from the study in order of hierarchy showed fatigue, training and complex automation to be the HFs with the greatest impacts on BWM operations. Minimizing their impact, therefore, will have the greatest positive contribution on the performance of the system.Relevance to industryThe study's outcome shall help decision makers in prioritizing limited resources (e.g. time and money) allocation to resolving only issues related to HFs with the greatest impact on BWM System's performance. The new methodology could also be applicable in assessing the relative impacts of subjective criteria like HFs in complex sociotechnical systems other than BWM Systems.     

高温深井通风降温技术的适用条件研究

随着矿山开采深度的不断增加，“三高一扰动”恶劣开采环境问题日益突出，给矿井通风降温工作带来极大困难。对于深井矿山，仅采用通风降温难以满足深部开采的需要，系统研究高温深井矿山通风降温技术的适用性问题，对矿山深部开采具有重要的理论意义和工程适用价值。首先从热力学角度，揭示风流在通风线路中的热交换规律；其次利用差分法原理计算深井筒风流温度，并依此推导出巷道和回采工作面风流温度变化趋势；然后结合采场安全生产允许温度进行反推，最终获取通风降温条件下的可采极限深度计算公式，并选取广西铜坑矿锌多金属矿体作为工程应用试验区进行论证分析。结果表明：基于风流的热交换模型可以推导出风流在井筒、巷道及工作面的温度计算公式，该公式与低温梯度、风流流经路径长度有关；假定工作面温度达到安全开采允许最高温度，可反推出该条件下矿山可采极限深度和巷道通风极限长度（仅采用通风降温措施时）；基于铜坑矿区锌多金属矿的实际条件，代入相关数据，验证计算所得极限开采深度符合实际，即所推导公式是可行的。