Software architecture review by association

During the process of software design, software architects have their reasons to choose certain software components to address particular software requirements and constraints. However, existing software architecture review techniques often rely on the design reviewers’ knowledge and experience, and perhaps using some checklists, to identify design gaps and issues, without questioning the reasoning behind the decisions made by the architects. In this paper, we approach design reviews from a design reasoning perspective. We propose to use an association-based review procedure to identify design issues by first associating all the relevant design concerns, problems and solutions systematically; and then verifying if the causal relationships between these design elements are valid. Using this procedure, we discovered new design issues in all three industrial cases, despite their internal architecture reviews and one of the three systems being operational. With the newly found design issues, we derive eight general design reasoning failure scenarios.     

面向SAR目标识别的深度卷积神经网络结构设计

目的 针对用于SAR （synthetic aperture radar） 目标识别的深度卷积神经网络模型结构的优化设计难题,在分析卷积核宽度对分类性能影响基础上,设计了一种适用于SAR目标识别的深度卷积神经网络结构。方法 首先基于二维随机卷积特征和具有单个隐层的神经网络模型-超限学习机分析了卷积核宽度对SAR图像目标分类性能的影响;然后,基于上述分析结果,在实现空间特征提取的卷积层中采用多个具有不同宽度的卷积核提取目标的多尺度局部特征,设计了一种适用于SAR图像目标识别的深度模型结构;最后,在对MSTAR （moving and stationary target acquisition and recognition） 数据集中的训练样本进行样本扩充基础上,设定了深度模型训练的超参数,进行了深度模型参数训练与分类性能验证。结果 实验结果表明,对于具有较强相干斑噪声的SAR图像而言,采用宽度更大的卷积核能够提取目标的局部特征,提出的模型因能从输入图像提取目标的多尺度局部特征,对于10类目标的分类结果（包含非变形目标和变形目标两种情况）接近或优于已知文献的最优分类结果,目标总体分类精度分别达到了98.39%和97.69%,验证了提出模型结构的有效性。结论 对于SAR图像目标识别,由于与可见光图像具有不同的成像机理,应采用更大的卷积核来提取目标的空间特征用于分类,通过对深度模型进行优化设计能够提高SAR图像目标识别的精度。     

Detection of Geometric Temporal Changes in Point Clouds

Detecting geometric changes between two 3D captures of the same location performed at different moments is a critical operation for all systems requiring a precise segmentation between change and no‐change regions. Such application scenarios include 3D surface reconstruction, environment monitoring, natural events management and forensic science. Unfortunately, typical 3D scanning setups cannot provide any one‐to‐one mapping between measured samples in static regions: in particular, both extrinsic and intrinsic sensor parameters may vary over time while sensor noise and outliers additionally corrupt the data. In this paper, we adopt a multi‐scale approach to robustly tackle these issues. Starting from two point clouds, we first remove outliers using a probabilistic operator. Then, we detect the actual change using the implicit surface defined by the point clouds under a Growing Least Square reconstruction that, compared to the classical proximity measure, offers a more robust change/no‐change characterization near the temporal intersection of the scans and in the areas exhibiting different sampling density and direction. The resulting classification is enhanced with a spatial reasoning step to solve critical geometric configurations that are common in man‐made environments. We validate our approach on a synthetic test case and on a collection of real data sets acquired using commodity hardware. Finally, we show how 3D reconstruction benefits from the resulting precise change/no‐change segmentation.     

基于SOA架构的校园信息化平台的设计和实现

如何利用原有系统的资源,建立开放灵活、低代价的系统已经成为高校信息化建设的关键,面向服务的体系结构（SOA）为解决分布式环境中软件重用,扩展和提高软件开发效率提供了解决方案。文章介绍了SOA的概念和基本结构,针对SOA的实现手段之一Web服务,给出了某高校应用Web服务实现校园信息化平台的模型及系统的总体框架,并分析了实现该平台的一些关键技术。     

Ubuntu下ROS变种ROCOS的系统架构

伴随人工智能等相关技术的进步, 智能机器人技术得到极大提高. 其软件上的日渐成熟和硬件设施的不断完善对研究和开发机器人操作系统和体系结构提出新的需求. 本文针对目前广泛使用的传统机器人操作系统ROS进行介绍和分析, 并对比研究了新型机器人操作系统ROCOS在系统架构方面对传统ROS的改进和优化, 详细介绍了两者的组织框架, 并通过仿真实验实际测试了两者框架的合理性与使用效果, 对后续新型机器人操作系统的研发具有借鉴意义.     

A software architecture centric self-adaptation approach for Internetware

Being one of the basic features of Internetware, self-adaptation means that the software system can monitor its runtime state and behavior and adjust them when necessary according to pre-defined policies. Focusing on the three fundamental issues of self-adaptation, including the scope, operability and trustworthiness, a software architecture （SA） centric approach for Internetware＇s self-adaptation is presented in this paper. All of the self-adaptive actions, i.e. monitoring, analyzing, planning and executing, are performed based on SA. In detail, runtime state and behavior of Internetware are represented and changed in the form of runtime soft- ware architecture. The knowledge for self-adaptation is captured, organized and reasoned in the form of SA so that automatic analysis and decision-making are achieved.     

User‐Centric Service Map for Identifying New Service Opportunities from Potential Needs: A Case of App Store Applications

Today, service innovation is just as important as product innovation. The ideation of service innovation, vis‐à‐vis product innovation, is user‐oriented, rather than maker‐oriented. Thus, capturing and understanding user context is key to being able to offer personalized and relevant services, and further identify new service opportunities. Therefore, a user‐centric approach is needed in new service development, especially in the era of ubiquitous service. There exists a rich literature on how to incorporate customers into new service development, but most has focused only on their expressed needs, and failed to identify potential needs. In response, this paper proposes a ‘user‐centric service map’, which first visualizes the portfolio of existing services based on the dictionary of potential needs and then helps to investigate vacuums that can provide a concrete shape for new service opportunities. The suggested approach is composed of three parts: first, constructing a potential needs dictionary, second, developing a service map, and finally, identifying new service opportunities. A case study of Apple's App Store services is conducted to verify the feasibility and utility of the proposed approach and offer strategic implications.     

深度神经网络在多场景车辆属性识别中的研究

单一的车辆属性识别已无法满足现有的交通系统,为了提高在实际监控中车辆检测定位的可靠性,利用深度神经网络的思想建立了一种能够在近景监控场景和交通监控场景两种不同场景下识别车辆属性的模型,主要包括车辆类型和颜色两种属性类别。以YOLOv3神经网络为基础,对其进行改进,降低网络深度的同时保证准确率,将车辆类型和颜色属性进行分级训练,提高模型检测速度,此外,创建了AttributesCars车辆属性数据集完成数据准备工作。实验结果表明,所提方法在平均准确率为95.63%的前提下可以满足视频的实时性要求,并且在两种不同场景下均取得了不错的成绩,适用于多场景车辆属性识别。     

The IKBALS project: Multi-modal reasoning in legal knowledge based systems

In attempting to build intelligent litigation support tools, we have moved beyond first generation, production rule legal expert systems. Our work integrates rule based and case based reasoning with intelligent information retrieval.When using the case based reasoning methodology, or in our case the specialisation of case based retrieval, we need to be aware of how to retrieve relevant experience. Our research, in the legal domain, specifies an approach to the retrieval problem which relies heavily on an extended object oriented/rule based system architecture that is supplemented with causal background information. We use a distributed agent architecture to help support the reasoning process of lawyers.Our approach to integrating rule based reasoning, case based reasoning and case based retrieval is contrasted to the CABARET and PROLEXS architectures which rely on a centralised blackboard architecture. We discuss in detail how our various cooperating agents interact, and provide examples of the system at work. The IKBALS system uses a specialised induction algorithm to induce rules from cases. These rules are then used as indices during the case based retrieval process.Because we aim to build legal support tools which can be modified to suit various domains rather than single purpose legal expert systems, we focus on principles behind developing legal knowledge based systems. The original domain chosen was theAccident Compensation Act 1989 (Victoria, Australia), which relates to the provision of benefits for employees injured at work. For various reasons, which are indicated in the paper, we changed our domain to that ofCredit Act 1984 (Victoria, Australia). This Act regulates the provision of loans by financial institutions.The rule based part of our system which provides advice on the Credit Act has been commercially developed in conjunction with a legal firm. We indicate how this work has lead to the development of a methodology for constructing rule based legal knowledge based systems. We explain the process of integrating this existing commercial rule based system with the case base reasoning and retrieval architecture.     

基于CBR的复杂结构井钻井过程智能决策支持系统

由于复杂结构井钻井过程存在大量复杂和不确定性因素,建立精确的数学模型来解决钻井决策问题十分困难,提出了基于实例推理技术构建钻井过程智能决策支持系统.在传统的实例相似度计算模型基础上,针对不同类型的属性设计了一个相似度综合计算模型,有效地解决信息的不精确性.详细描述了系统的总体结构、实例表示与组织以及推理模型的设计.     

Architecture‐driven assistance for fault‐localization tasks

Finding software faults is a problematic activity in many systems. Existing approaches usually work close to the system implementation and require developers to perform different code analyses. Although these approaches are effective, the amount of information to be managed by developers is often overwhelming. This problem calls for complementary approaches able to work at higher levels of abstraction than code, helping developers to keep intellectual control over the system when analyzing faults. In this context, we present an expert‐system approach, called FLABot, which assists developers in fault‐localization tasks by reasoning about faults using software architecture models. We have evaluated a prototype of FLABot in two medium‐size case studies, involving novice and non‐novice developers. We compared time consumed, code browsed and faults found by these developers, with and without the support of FLABot, observing interesting effort reductions when applying FLABot. The results and lessons learned have shown that our approach is practical and reduces the efforts for finding individual faults.     

Evolution styles: foundations and models for software architecture evolution

As new market opportunities, technologies, platforms, and frameworks become available, systems require large-scale and systematic architectural restructuring to accommodate them. Today’s architects have few techniques to help them plan this architecture evolution. In particular, they have little assistance in planning alternative evolution paths, trading off various aspects of the different paths, or knowing best practices for particular domains. In this paper, we describe an approach for planning and reasoning about architecture evolution. Our approach focuses on providing architects with the means to model prospective evolution paths and supporting analysis to select among these candidate paths. To demonstrate the usefulness of our approach, we show how it can be applied to an actual architecture evolution. In addition, we present some theoretical results about our evolution path constraint specification language.     

An IT management assessment framework: evaluating enterprise architecture scenarios

Enterprise architecture, EA, is an established approach for the model-based and holistic management of IT. The scope of EA is however wide and the predominant EA frameworks suggest the creation of broad and detailed models. IT management cannot control all areas spanned by the present frameworks for EA models. In order to ensure well-informed decisions, IT management has a series of questions that need to be answered. This paper proposes an assessment framework that can be used to identify relevant questions for assessments of EA and EA scenarios, within the areas of EA that IT management can control. Three top dimensions in the proposed framework are presented: IT organization, IT systems, and Business organization. The framework further includes sub dimensions for identifying questions. An application of the assessment framework, as it was applied to assess EA scenarios in a power company, is also described.     

基于语义推理的DSS模型研究与应用

基于语义网理论建立基于语义推理的DSS模型。该模型引用W3C提出的资源描述框架（RDF）以描述业务领域中各要素,形成领域本体。并对领域中业务数据进行语义标注,构建事实库。在所构建事实库的基础上,系统根据预设规则进行推理,以达到决策支持的目的。最后基于以上方法对施工项目设备推荐领域进行实例建模,建立设备推荐系统。该系统将Semantic Web与Ontology理论引入到施工项目领域,以解决施工项目过程中存在的设备选型推荐问题,并作为语义Web理论在该领域内应用的一次尝试。     

An information architecture for a performance management framework by collaborating SMEs

Recently, much attention has been given to performance management within collaborative environments resulting in a wide variety of performance measurement/management frameworks. However, practically all of these frameworks present a common gap regarding the information treatment needs that support these frameworks. This issue is even more important in the case of Small and Medium-sized Enterprises (SMEs) due to their special characteristics. For that purpose, this paper presents an approach called Global Performance Management (GPM) Framework composed of a methodology, an information architecture and a technological solution, focusing mainly on the information architecture. The methodology guides SMEs when defining, developing, deploying and operating the GPM framework. The information architecture consists of a set of elements that support the tasks of collection, treatment and analysis of the used information, which complement the methodology to build a real model for the GPM framework. Finally, the technological solution bridges the engineering and conceptual environment, with the real-world operational and executable environment, following the path defined by the methodology. The information architecture was applied to two European pilots, achieving satisfactory results, which are highlighted in the paper as well as the main barriers encountered and solutions provided.     

Disturbance observer–based prescribed adaptive control for rate‐dependent hysteretic systems

In this paper, a disturbance observer–based prescribed adaptive control approach is proposed for ultra‐high‐precision tracking of a class of hysteretic systems with both high‐order matched and mismatched disturbances. Considering the adverse effects of asymmetric and rate‐dependent hysteresis nonlinearities, a polynomial‐based rate‐dependent Prandtl‐Ishlinskii model is first developed to characterize their behaviors, and inverse model based compensation is also constructed. Furthermore, the resulting inverse compensation error is analytically given, and a novel disturbance observer with adaptive control techniques is designed to handle the bounded disturbances, including the inverse compensation error and the high‐order matched and mismatched disturbances. Comparative experiments on a multiaxis nano servo stage are finally conducted to demonstrate the effectiveness of the proposed control architecture, where substantial performance improvement over existing results are achieved on various tracking scenarios.