The affect of contact force sensations on user performance in virtual assembly tasks

The development of a realistic virtual assembly environment is challenging because of the complexity of the physical processes and the limitation of available VR technology. Many research activities in this domain primarily focused on particular aspects of the assembly task such as the feasibility of assembly operations in terms of interference between the manipulated parts. The virtual assembly environment reported in this research is focused on mechanical part assembly. The approach presented addresses the problem of part-to-part contacts during the mating phase of assembly tasks. The system described calculates contact force sensations by making their intensity dependent on the depth of penetration. However the penetration is not visible to the user who sees a separate model, which does not intersect the mating part model. The two 3D models of the part, the off-screen rendered model and the on-screen rendered model are connected by a spring-dumper arrangement. The force calculated is felt by the operator through the haptic interface when parts come in contact during the mating phase of the assembly task. An evaluation study investigating the effect of contact force sensation on user performance during part-to-part interface was conducted. The results showed statistically significant effect of contact force sensation on user performance in terms of task completion time. The subjective evaluation based on feedback from users confirmed that contact force sensation is a useful cue for the operator to find the relative positions of components in the final assembly state.     

基于COM＋的协同集成设计环境下用户管理模型

近年来,计算机支持的协同设计(CSCW)已经成为一项热门的研究话题。本文提出了协同设计环境下用户管理的模型,探讨了不同逻辑网络结构,特别是在对等网络结构下,用户动态加入和退出的算法。通过使用COM 技术实现了客户和服务器位置的透明性、角色的划分和用户之间的安全交互等功能。实验结果进一步验证了设计的正确性和有效性。     

Integrating specification of human-computer interface with Jackson system development

Procedures for integrating task analysis and design of the human-computer interface (HCl) with a structured system development method, Jackson system development (JSD), are described. Process structure diagrams (PSDs), the major modelling notation of JSD, described tasks, and task specifications were evaluated for cognitive complexity. Task complexity analysis helped allocation and design of human tasks and to produce computer task support specifications. The cognitive dimension of task analysis highlighted the need for task support actions, especially information display support for working memory. Dialogue specification for a direct manipulation interface design was based on the JSD entity/event model from which the interaction metaphor, screen design, and permissible manipulations were derived. Further PSDs were constructed to specify interface object management processes. Preliminary evaluation of the method was carried out using JSD analysts. Results showed the method proved easy to learn even for non-HCl specialists.     

基于PCTBTA任务模型的用户界面开发方法

为了适应普适计算环境中用户、设备、使用环境和开发平台的多样性,基于模型的方法被应用于用户界面开发过程中,试图在抽象层次上描述界面,通过模型转换,使其适用于不同的平台.然而,由于目前基于模型的用户界面开发方法（model-based user interface development,简称MBUID）中所采用任务模型的局限性,致使生成的界面难以满足动态环境下用户的可用性需求.提出一种基于任务模型的用户界面开发框架,旨在建模和生成有效、高效、令用户满意的用户界面.在可用性方面,为了准确描述普适计算环境中用户任务,提出一种基于感知控制理论的任务分析方法（perceptual-control-theory-based task analysis,简称PCTBTA）,将使用上下文信息引入到任务分析过程中,并且在较高的抽象层次上反映交互的内容,给可用性设计提供任务空间;在技术方面,为PCTBTA任务模型向界面模型的转换提供技术支持.最后,通过实例说明所提出方法的可行性,并通过与其他方法在可用性和性能方面的比较,表明该方法的有效性.     

Toward a formal specification of menu-based systems

As software systems continue to increase in sophistication and complexity, so do the interface requirements that support the corresponding user interaction. To select the proper blend of ingredients that constitutes an adequate user interface, it is essential that the system designer have a firm understanding of the interaction process, i.e., how the selected dialogue format interacts with the user and with the underlying task software. To promote such an understanding, this paper presents a model that characterizes one particular dialogue format: menu-based interaction. This model is actually a sequence of models, hierarchically structured, in which each successive model builds on its predecessor by introducing additional characterization elements. The first model describes the minimal set of elements inherent to any menu-based interface; successive models extend this minimal set by introducing task actions, incremental history sequences, and frame-associated memory. These principal model elements enable the characterization of fundamental, menu-based operations like computational and decision processes, user response reversal, and user directed movement. Moreover, because the principal model elements correspond directly to “real world” objects, an intuitive as well as formal understanding of menu-based interaction can be achieved. Effectively, the model elements and the hierarchical structure imposed by these elements provide an ideal environment for characterizing and classifying menu-based systems at various levels of sophistication.     

On the use of shared task models in knowledge acquistion,strategic user interaction and clarification agents

In this paper, three different roles of a shared task model as an intermediate representation of a task are presented and illustrated by applications developed in cooperation with industry. First the role of a shared task model in knowledge acquisition is discussed. In one of the two applications, decision support in the domain of soil sanitation, one of the existing generic task models for diagnostic reasoning provided a means to structure knowledge acquisition. In the second application, diagnosis of chemical processes, the acquisition process resulted in a shared task model for diagnostic reasoning on Nylon-6 production. Secondly, the role of a shared task model in designing user interaction is addressed. Three levels of interaction are considered of importance: interaction at the object level, at the level of strategic preferences, and at the level of task modification. In an application in the domain of environmental decision making, this led to the design of a user interface based on the acquired shared task model, within which all three levels of interaction were available to users. Finally, the role of shared task models within a multi-agent system including a clarification agent is addressed. Two software agents were designed that each share a task model with the user: one for a diagnosis task, and one for a clarification task. The shared model of the clarification task reflects the shared task model of diagnosis; clarification includes clarification of the overall diagnostic reasoning process. The multi-agent architecture presented has been developed to support a user both at the level of the diagnostic task he or she is performing and at the level of clarification. The architecture has been applied to the diagnosis of chemical processes.     

强跟踪延迟滤波算法及其在感应电机无速度传感器控制中的应用

在强跟踪滤波(Strong track filter, STF)算法和延迟扩展Kalman滤波(Schmidt extended Kalman filter, SEKF)算法的基础上, 提出了强跟踪延迟滤波(Strong track Schmidt filter, STSF)算法, 结合感应电机降阶模型建立了电机状态估计算法, 将其应用于感应电机无速度传感器控制系统中, 并与扩展Kalman滤波(Extended Kalman filter, EKF)、SEKF和STF三种算法的状态估计性能作比较. 仿真和实验结果表明, STSF算法在估计精度、跟踪速度、抑止噪声等方面均优于EKF算法, 并且计算复杂度显著降低, 能有效在线估计电机转速和磁链.     

融合潜在狄利克雷分布与元路径分析的用户相关性度量方法

用户相关性度量是异构信息网络研究的基础与核心。现有的用户相关性度量方法由于未充分开展多维度分析和链路分析,其准确性尚存在提升空间。为此,提出了一种融合狄利克雷分布（LDA）与元路径分析的用户相关性度量方法。首先利用LDA进行主题建模,通过分析网络中节点的内容来计算节点的相关性;然后,引入元路径来刻画节点间关系类型,通过关联度量（DPRel）方法对异构信息网络中的用户进行相关性测量;接着,将节点的相关性融入到用户相关性度量计算中;最后,采用IMDB真实电影数据集进行实验,将所提方法和嵌入LDA主题模型的协同过滤推荐方法（ULR-CF）、基于元路径的相关性度量方法（PathSim）进行了对比分析。实验结果表明,所提方法能够克服数据稀疏性弊端,提高用户相关性度量的准确性。     

人机交互中基于时间约束的人体工效模型

针对人机交互领域速度-准确度折中关系的预测中任务完成精确度的预测模型较为欠缺的问题,提出了一种基于时间约束的精确度模型预测方法。该方法采用了人机交互研究中常用的受控实验测试分析法,研究了在计算机用户界面中要求用户在给定的时间内完成任务时,任务完成的精确度与给定的时间约束之间的折中关系,用以衡量完成时间约束任务的人体工效。实验中设计了一系列受时间约束的轨道滑动任务,实验环境中自变量包括轨道长度、轨道宽度以及规定的在轨道中滑动的时间,因变量为任务完成的精确度,采用在轨道中滑动时轨迹的纵向偏差表示。通过对30位被试者实验数据的分析发现,任务完成的精确度与轨道宽度以及滑动速度(表示为轨道长度/规定的滑动时间)之间构成线性的关系,在此基础上采用最小二乘方回归法建立了一个基于时间约束的任务完成精确度的量化模型;该模型与真实实验数据集的拟合优度达到了0.857。     

基于动态分类的Markov用户浏览预测模型

针对多Markov链用户浏览预测模型分类算法的时间复杂度过高问题,提出一种基于动态分类的Markov用户浏览预测模型。该模型通过学习提取用户浏览特征,利用这些特征对用户浏览路径进行分类,实现预测并动态更新用户浏览特征。实验结果表明,该模型可明显降低用户浏览路径预测的时间,并得到较为准确的预测结果。     

Perceived task complexity as a criterion for information support

Here, the conceptual model of perceived task complexity, which incorporates the user, task, and system into a comprehensive unit of analysis, is used to investigate how, when, and why computer-assisted instruction (CAI) systems provide effective support. The work is based on data provided by 156 experimental subjects; information system (IS) use was found to vary by user groups depending on task performance. In addition, the effects of task presentation on task behavior and performance were found to vary between groups, based on their performance. The findings imply that user, task, and system dynamically interact in providing effective IS support. A comprehensive model of perceived task complexity that includes a user's domain knowledge is developed. This should help in further investigating into how, when, and why IS are effective.     

Modelling surface radioactive,chemical and oil spills in the Strait of Gibraltar

A model that simulates the dispersion of chemical/radioactive and oil spills in the Strait of Gibraltar has been developed. Water currents over the Strait have been obtained from a hydrodynamic model. Computed tides and residual currents have been compared with observations in the area. The dispersion model, based on a particle-tracking technique, is solved off-line. Standard tidal analysis, carried out over results provided by the hydrodynamic model, is applied to obtain currents at any time and position of the Strait. Specific processes for each contaminant (decay of radioactive material, oil evaporation and decomposition) are included and simulated by means of a stochastic method. A Monte Carlo method is applied for turbulent diffusion. The model can deal with instantaneous and continuous releases. MatLab graphic user interfaces have been developed to introduce input data and visualize simulation results. Some dispersion calculations have been carried out. In general, contaminants are flushed towards the east due to the residual currents. Nevertheless, dominant east winds tend to retain contamination in the Strait and to enhance mixing. This is also the case if the release occurs close to the coast, where currents are weaker than in the central part of the Strait.     

基于ESN的污水处理过程优化控制

针对污水处理过程能耗过高的问题,提出了一种基于状态回声网络(ESN)的在线优化控制方法。建立了污水处理过程预测模型,实现性能指标的预测;根据系统的状态以及预测的性能指标,采用ESN实时优化控制变量的设定值;将优化后的设定值传送给底层控制器进行跟踪控制。将ESN优化控制方法在污水处理过程基准仿真模型(BSM1)上进行了验证,实验结果表明,该方法不但能够满足出水水质的要求,而且降低了污水处理过程运行成本。     

面向边缘设备的高能效深度学习任务调度策略

近年来,深度学习在图像和自然语言处理等诸多领域表现出色,与深度学习相关的各类移动应用发展迅速,但由于移动网络状态的不稳定性及网络带宽的限制,基于云计算的深度模型任务可能出现较大响应延迟,严重影响用户体验.与此同时,深度模型对设备的计算及存储能力有较高的要求,无法直接在资源受限的移动设备中进行部署.因此,亟须设计一种新的计算模式,使得基于深度模型的移动应用能够满足用户对快速响应、低能耗及高准确率的期望.本文提出一种面向边缘设备的深度模型分类任务调度策略,该策略通过协同移动设备与边缘服务器,充分利用智能移动终端的便捷性和边缘服务器强大的计算能力,综合考虑分类任务的复杂度和用户期望,完成深度模型在移动设备和边缘服务器中的动态部署,并对推理任务进行动态调度,从而提升任务执行效率,降低深度学习模型推理开销.本文以基于卷积神经网络的图像识别应用为例,实验结果表明,在移动环境中,相比于准确率最高的深度模型,本文提出的高能效调度策略的推理能耗可降低93.2%、推理时间降低91.6%,同时准确率提升3.88%.     

大型多人在线角色扮演游戏的下一地点预测

近年来,大型多人在线角色扮演游戏(MMORPG)已经成为最流行的网络娱乐活动之一。MMORPG在游戏环境中形成虚拟社会,其中每个玩家扮演某个虚构角色,并控制该角色的大多数活动。游戏的迅猛发展累积了海量数据,其中包含游戏虚拟社会的语义和拓扑信息。研究者针对游戏数据开展了一系列研究工作,如玩家退出预测、游戏服务器整合等。游戏角色的下一地点预测对提升游戏体验、改善游戏设计和检测游戏机器人均有十分重要的意义。目前,该项预测任务主要使用统计分析完成。然而,由于游戏数据具有海量特征,因此需要一种自动化的计算方法。文中提出了基于隐马尔科夫模型的游戏角色下一地点预测模型,该模型能够考虑与位置特性相关的不可观测的属性,同时兼顾游戏角色前期行为的影响。实验结果表明,与现有方法相比,该方法具有建模直观的特点,在稠密分布的MMORPG数据中能够得到更准确的下一地点预测结果。     

A review on automatic image annotation techniques

Nowadays, more and more images are available. However, to find a required image for an ordinary user is a challenging task. Large amount of researches on image retrieval have been carried out in the past two decades. Traditionally, research in this area focuses on content based image retrieval. However, recent research shows that there is a semantic gap between content based image retrieval and image semantics understandable by humans. As a result, research in this area has shifted to bridge the semantic gap between low level image features and high level semantics. The typical method of bridging the semantic gap is through the automatic image annotation (AIA) which extracts semantic features using machine learning techniques. In this paper, we focus on this latest development in image retrieval and provide a comprehensive survey on automatic image annotation. We analyse key aspects of the various AIA methods, including both feature extraction and semantic learning methods. Major methods are discussed and illustrated in details. We report our findings and provide future research directions in the AIA area in the conclusions     

基于卷积神经网络的飞机液压系统故障诊断

现代飞机系统的复杂度和集成度均大幅提升,使得其故障诊断存在难度高和准确率低等特点。采用一维卷积神经网络方法对军用飞机液压系统的故障诊断问题进行了研究,构建了满足多传感器数据分析要求的卷积神经网络模型。考虑到神经网络的输入来自不同的传感器数据序列,各数据序列之间的空间关系不明显,因此,即使网络输入是二维形式,而实际的卷积操作均在一维上进行。通过解决某飞机液压系统的故障诊断问题,证明将标准化后的多传感器数据序列及对应故障模式作为训练样本对卷积神经网络模型进行训练时,采用满足训练要求的网络对飞机液压系统进行故障诊断时具有较高的准确率。