PKS体系下基于知识图谱的政务问答机器人研究

提出了一种基于国产PKS环境上政务知识库问答机器人。该系统基于多源异构数据，利用深度学习模型，以流水线工作方式，由Schema设计、知识抽取、知识融合、知识存储、知识更新5部分组成，自动完成知识库建立，同时自动对用户问题进行分类，意图识别，利用知识图谱检索、分析计算技术形成答案提供给用户。该机器人系统能够大大提高政务工作人员的工作效率。     

基于知识的物料传送系统的控制模式

本文讨论作者开发的一个基于知识的机器人物料传送仿真系统 ,其中利用知识法则对传送机器人的控制模式进行转换 ,使系统在输入零件速率随机变动的情况下 ,能自动工作于输出零件流为最大的优化工作状态 ,从而实现对物料传送系统的自动寻优控制 .文章阐述了利用仿真实验获取系统有关知识和建立知识库的方法 ,同时通过对该系统的动画仿真研究 ,证实了知识库的有效性     

基于领域知识的3维动态场景目标识别和配准

为了增强遥操作系统的临场感,使操作者更好地融入远程工作环境,提出了基于领域知识的3维动态场景目标识别和配准算法.首先,通过离线解析和切分虚拟样机CAD（计算机辅助设计）模型构建包含多视角点云特征和装配约束的领域知识库.其次,通过动态采集场景点云并计算CVFH（clustered viewpoint feature histogram）和FPFH（fast point feature histogram）特征,利用领域知识库中多视角点云特征和点云CVFH特征进行比对实现目标识别,并由FPFH特征经过两步配准实现目标姿态确定.最后,利用装配约束知识库实现以遥操作机器人工作状态变化为驱动的指导信息的精确配准和实时推送.实验结果表明,该系统不但可以有效地指导远程机器人完成维修操作,还可以提高遥操作的精度和效率.     

一个基于知识库及自动推理机制的化学反应系统

提出了一种基于知识库及自动推理机制的化学反应系统,详细阐述了该自动推理系统的技术架构与算法实现.主要收集了无机化学知识,分析并提取相关化学反应规则,建立了相应的化学知识库,实现了由添加反应物到化学反应生成物的一系列自动推理过程.通过稀硫酸和铁反应实例表明,该系统对开发智能的虚拟实验平台,以及学生运用软件自主学习化学知识,掌握化学规律与现象,具有重要作用.     

基于机器视觉的遥控器装配系统研究与设计

为实现遥控器按键装配生产线自动化,结合机器视觉技术与SCARA机器人的精确定位与控制,采用Roberts算子进行工件边缘提取,确定模板匹配方法。在得出图像坐标与机器人末端抓手坐标的转换函数关系后,计算模板位置与抓取位置之间的差值,在工件装配过程中进行差值补偿,实现自动装配。经测试,抓取和装配精度达到±0.1 mm,完全符合规定的装配速度和精度要求。系统实现了遥控器按键的智能装配,优化了现场工作流程,提高了工作效率。与其他自动化系统相比,该系统更具智能性和便捷性。系统的设计思路可用于各类自动装配,能够有效保证装配过程的安全性、可靠性及稳定性,是未来智能装配的发展趋势。     

智能工厂搬运系统的设计

本文针对智能工厂生产模式的实际需求,研制基于搬运机器人的智能工厂搬运系统,整个系统包含两大设备:装配搬运机器人和外围设备.外围设备包括供料站、冲压站和仓库区.装配搬运机器人主要负责物料的转移和搬运,而外围设备负责物料的处理加工和存储.具体细分,可将系统分成装配搬运机器人、供料站、冲压站和仓库区这四部分,形成一体化流水线...     

基于接触状态感知发育的机器人柔性装配方法

为了提高工业机器人装配的实时性、自适应性和鲁棒性,借鉴人类后天感知学习方式,提出一种基于接触状态感知发育的柔性装配方法.采用机器人末端的位姿和力/力矩来描述装配接触状态,结合支持向量数据描述和改进极限学习机对接触状态感知发育,形成可自我更新成长的经验知识库,预测机器人的装配动作,完成柔性装配任务.为验证所提出方法的有效性,以小型断路器卡合装配为例进行实验,实验结果表明,采用接触状态感知发育可实现装配经验知识库的自我更新,完成机器人的柔性装配,验证了所提出方法的可行性和有效性.     

零件图纸自动阅读与识别研究初步——视图分割

智能装配机器人的研究是世界性的研究项目之一，为了实现智能装配机器人对零件的操作，首先必须做到对零件图纸的自动阅读与识别。本文介绍了零件图纸的自动阅读与识别的处理过程，着重介绍了这一处理过程中的视图分割原理与采用的技术。     

智能网络管理中一种基于Petri网的知识验证技术1

知识库是智能网络管理系统中的重要组成部分,知识库的维护要求知识库系统必须具备知识的验证功能.智能网管系统中知识组织的动态性决定了知识的结构性验证必须正确、有效和及时.本文以有色Petri网为基础,针对智能网管系统中知识验证的主要内容,以一种扩充的有色Petri网作为描述和分析工具,建立了相应的知识表示模型和知识验证模型,提出了一种简洁的、基于有色Petri网的知识验证方法,并给出了相应的判定准则.     

基于知识本体的智能答疑系统知识库——以“大学计算机基础”课程为例

答疑系统的智能化程度是影响精品课程持续性发展的重要因素。文章提出基于知识本体建立智能答疑系统知识库,可以自动生成相关知识领域较全面的问题和相应答案,使学生输入问题后能很快在知识库中找到答案,大大提高学生对于答疑系统的利用率,也可以增强学生对于精品课程的兴趣。     

Management of complex structural engineering objects in a relational framework

To structure the development of an integrated building design environment, the global representation of the design data may best be organized in terms of hierarchies of objects. In structural engineering design we deal with large sets of independent but interrelated objects. These objects are specified by data. For an engineering design data base the system must be able to model the objects composing the design as well as to manage effectively the design data. The data base management system therefore needs to have some knowledge of the intended use of the data, and must provide an abstraction mechanism to represent and manipulate objects. Much recent research in engineering data bases focuses on object management for specific tasks but gives little attention to the shareability of the underlying information. This paper describes an architecture for the management of complex engineering objects in a sharable, relational framework. Potential application of this approach to object management for structural engineering analysis and design is discussed.     

三给地形中地物模型构造方法

根据地物模型与地形三角网关系,将各类繁多地物模型划分为两大类。第一类地物是依赖于地形三角网地物。第二类地物是独立于地形三角网地物。在首先构造地形Delaunny三解网的基础上,本文采取了对三角片纹理实施单独控制的方法。实验结果表明,本文提出的对地物以对象形式实施单独控制的方法,可以充分满足地物表现的需要。     

Concurrent intelligent rapid prototyping environment

Rapid prototyping technology enables rapid production of complex objects directly from a computer-aided design model without involving any tooling or conventional part programming. This has created a new set of problems associated with part design, process planning, support design and value engineering analysis of rapid prototyping parts. In this paper, a methodology for resolving these problems is described, which uses concurrent engineering, distributed blackboard, value engineering, knowledge-based and feature-based technologies. The functionality, design methodology and knowledge representation techniques of a concurrent intelligent rapid prototyping system for stereolithography form the main focus of this paper.     

A RFBSE model for capturing engineers’ useful knowledge and experience during the design process

Reuse of designers’ knowledge and experience of solving problems during the engineering design process holds the key to increase efficiency of decision making in future projects. An important part of this useful knowledge and experience is the interpretation of data and information about design objects and processes as well as the generation of new information for decision-making. However, previous studies on knowledge representation models have mainly focused on developing a structure to describe the knowledge about design objects and design processes while a systematic method that can effectively integrate knowledge about design objects and knowledge about problem-solving strategies is still missing. To fill this gap, a RFBSE knowledge representation model for capturing useful design knowledge and experience for future reuse is developed and evaluated in this study. This paper describes the key elements of this model, explains the rationale of using particular elements, and discusses the evaluation of the model using an engineering design example.     

From automation to tangible interactive objects

Automation led to many innovations for a long time, most of them were developed during the twentieth century. It was commonly thought as a layer on top of a mechanical system. It promoted system management over low-level control. The more information technology evolves, the more it takes a fundamental part in our lives. This article describes a paradigm shift where automation will no longer be an add-on, and where software supports the definition, implementation and operationalization of functions and structures of products from the beginning of the design process. Any design today starts by using computer-aided design tools that enable us to easily draw, modify and fine-tune any kind of system. We can fully develop an airplane and literally fly it as a complex piece of software. Usability and usefulness can be tested before anything physical is built. Consequently, human-centered design (HCD) is now not only feasible but also can drive the overall engineering of products. We have started to design products from outside in, i.e., from usages and purposes to means. We even can 3D print mechanical parts from the software-designed parts with ease. In human–computer interaction, specific research efforts are carried out on tangible objects, which define this inverted view of automation. We now design and develop by using information technology to do mechanical things, and therefore redefine the essence of a new kind of cognitive mechanical engineering. This article is about the revolution that is currently happening in engineering and industrial design due to the immersive influence of computers in our everyday life, and the expansion of HCD.     

Knowledge visualization for evaluation tasks

Although various methods for the evaluation of intelligent systems have been proposed in the past, almost no techniques are present that support the manual inspection of knowledge bases by the domain specialist. Manual knowledge base inspection is an important and frequently applied method in knowledge engineering. Since it can hardly be performed in an automated manner, it is a time-consuming and costly task. In this paper, we discuss a collection of appropriate visualization techniques that help developers to interactively browse and analyze the knowledge base in order to find deficiencies and semantic errors in their implementation. We describe standard visualization methods adapted to specifically support the analysis of the static knowledge base structure, but also of the usage of knowledge base objects such as questions or solutions. Additionally, we introduce a novel visualization technique that supports the validation of the derivation and interview behavior of a knowledge system in a semi-automatic manner. The application of the presented methods was motivated by the daily practice of knowledge base development.     

基于本体的整体与部分关系的形式化表示及其错误检查研究

整体/部分WP(whole/part)关系在计算机领域建模中无处不在.在一个WP关系中,由于整体对象具有管理和控制部分对象的功能,因此WP关系存在着许多其它关系没有的特征.但本体的研究中,存在着形式化程度不够的问题,人们常用自然语言解释这些特征,然而这样的解释通常是不准确的.另一方面,由于知识处于不断的增加与更新之中,难免造成知识库中知识的错误,导致知识库维护困难.因此本文首先基于本体对WP关系的特征进行形式化描述,在此基础上,根据WP关系的特征提出了一系列相关知识的形式判定规则.