跳到内容
指出 “支持人在环路混合智能的交互设计” 这一类设计问题, 研究人在环路混合智能系统中交互设计的问题, 为相关设计, 技术与应用研究提供索引和参考. 从人在环路混合智能的概念和架构出发, 引出人在环路混合智能的交互设计; 基于对相关文献的整理, 总结常见界面构成和交互方式; 总结整理人在环路混合智能的生命周期. 指明了人在环路混合智能是需要用户交互的智能模型, 介绍了由用户, 人工智能算法, 用户接口构成的系统架构; 总结了针对不同数据类型的现有工作可能的交互方式; 分析了人在环路混合智能完整生命周期中的设计挑战, 根据现有文献提取关键界面构成, 提出了人在环路混合智能系统的设计建议; 提出了从智能系统, 用户, 设计师三方面建立设计方法论, 完善设计工具, 更有效地支持和推动人在环路混合智能系统的应用的建议.[......]继续阅读
智能系统在智能制造、智慧城市、医疗健康、生活服务等各种场景中越来越广泛地存在，为应对人与智能系统的交互中所面临的诸多挑战，从技术和体验的视角分析人机智能协同中的关键问题。对从人机交互到人机智能协同的发展脉络与研究范围进行梳理，提出综合技术视角和体验视角的研究框架；从智能系统的特征出发，梳理出技术视角下人机智能协同所带来的新兴问题；从体验的视角探讨如何推动实现人机智能协同；在此基础上总结人机智能协同的发展趋势。总结了人机交互演进的三个阶段；提出了技术视角下人机智能协同的关键问题，包括人机能动性分配、动态学习和修正、情境自适应及主动响应模式；探讨了体验视角下人机智能协同的可解释性、信任问题、情感化及公平负责等问题；指出了人机智能协同全方位、多类型及体系化的发展趋势。[......]继续阅读
合成孔径雷达(SAR)图像目标识别是实现微波视觉的关键技术之一。尽管深度学习技术已被成功应用于解决SAR图像目标识别问题，并显著超越了传统方法的性能，但其内部工作机理不透明、解释性不足，成为制约SAR图像目标识别技术可靠和可信应用的瓶颈。深度学习的可解释性问题是目前人工智能领域的研究热点与难点，对于理解和信任模型决策至关重要。该文首先总结了当前SAR图像目标识别技术的研究进展和所面临的挑战，对目前深度学习可解释性问题的研究进展进行了梳理。在此基础上，从模型理解、模型诊断和模型改进等方面对SAR图像目标识别的可解释性问题进行了探讨。最后，以可解释性研究为切入点，从领域知识结合、人机协同和交互式学习等方面进一步讨论了未来突破SAR图像目标识别技术瓶颈有可能的方向。[......]继续阅读
With the rapid development and application of autonomous vehicles, human-autonomous vehicle interaction (HAI) has recently gained importance. Simulation is an effective and efficient approach for the research and testing in the HAI domain, and scenarios are crucial to the validity and experience of HAI simulators. However, research on systematically building scenarios is still lacking. This paper proposes the concept of the narrative scenario in the HAI simulation, and a three-stage framework for building narrative scenarios is established. First, key-plot scenarios are collected from the whole journey analysis and deconstructed into elements. After extracted and categorized into a library, the physical and narrative attributes of scenario elements are defined. Finally, scenarios in simulators can be rebuilt in a narrative sequence driven by task goals.[......]继续阅读
In the era of a new generation of artificial intelligence, this work aims to analyze and identify the characteristics and value of artificial intelligence products and service systems, indicates the future development trends, and provides references for related design, technology and application research. Starting from the concept of artificial intelligence (AI), the concepts of AI products and service systems is defined in this paper. Typical AI products and related re- search are reviewed and analyzed, and the key features and supporting technologies of artificial intelligence products are summarized. Then, this paper explores the typical service scenarios of AI products and reviews the current status of relevant research. Finally, the future trends and challenges are predict based on the previous analysis. This paper indicates that the typical characteristics of AI products include context awareness, adaptive learning, autonomous decision-making, proactive interaction and collaboration. Framework of supporting technologies of AI products is described, that data and computing power as base, algorithms as core, and multiple underlying technologies and general technologies supporting applications in various scenarios. The value that the service system of AI products can bring is analyzed in different scenarios. The future trends are predicted as the transformation from tech-driven to design driven, and the switch of perspective from single AI product to service system.
[......]继续阅读
Mixed reality (MR) devices blur the boundaries between the virtual world and reality, reshaping the way people work with assistive information. However, there are still strong lim-for information searching tasks in MR glasses. Added capabilities of HoloLens 2, a recently released MR device, bring new possibilities to deal with these issues. Interactive approaches are proposed in this study, including body/hand-locked components , view-locked navigation components, and view-sensitive information layout. Prototypes were developed with and without these interactive approaches, and user studies were conducted to measure the task performance, usability, and presence. Results show that interactive approaches have positive effect in terms of task completion time. Different cognitive and behavioral styles may lead to distinct preferences for different interactive approaches.[......]继续阅读
设计说明: 在这个数据过载的时代, 信息可视化如同天文学家的望远镜和生物学家的显微镜, 是将数据处理成易于人脑理解和吸收形式的工具. 我们接触到的数据, 常常是每个数据项具有两个以上属性值的高维数据; 而我们常用的纸媒和屏幕, 只有物理上的两个显示信息的维度, 对于高维数据的呈现有着一定局限性.[......]继续阅读
通过引入主动式 HMI，车辆可以预测用户的意图并启动功能，从而减少干扰，增强灵性，提高驾驶安全性和用户体验。通过主动式 HMI，使用意图预测模型的准确性机制层面成为影响主动HMI体验质量的关键。然而，缺乏有效的手段来提高用户预测模型的准确性，并且相关研究还不够充分。智能交互是提高工作效率的有效方法机器人的性能。通过将该技术引入到主动响应式的设计中通过交互，可以获得用户的意图，有助于突破当前算法的瓶颈。提出一种基于智能的汽车主动响应式交互设计框架交互并利用智能交互提高预测精度，是预测的关键点还列出了值得关注的地方，并举例说明了具体的设计案例。[......]继续阅读
In this paper, we propose a vision-based hand gesture recognition system for human-computer interaction. The gesture recognition systems are employed in developing a rock-paper-scissors game between human and our robotic hands in realtime. Our task is to predict the gestures as soon as possible by using high-speed cameras. Due to the computational complexity, the standard long-term recurrent convolution network-based action classification system cannot be contented with classification tasks based on high-speed cameras. We propose to address this issue by employing a more efficient network architecture and using a threshold-based method to predict the gesture in advance. We validate our proposed method on the new gesture dataset for the rock-paper-scissors game. The model is able to successfully learn gestures varying in duration and complexity. A comparative analysis of CNN and long-term recurrent convolution network is performed. We report a gesture classification accuracy of 97% and report a near real-time computational complexity of 7 ms per frame.[......]继续阅读
HMI is used to refer to human-vehicle interaction design from the perspective of taking car as a machine. However, with the quick increase of demand for smart cockpit, it would put strong constraints to the design of intelligent interactions and connected services if we still design from the perspective of control-oriented interface with a machine. By switching the concept from Human Machine Interaction (HMI) to Human Robot Interaction (HRI) can instead greatly open up the space of innovation for the development of natural interactions with the car as an intelligent system. This also make it possible to further focus on topics such as adaptive learning of the system through smart interaction. Designing from the perspective of human robot interaction is even more important for autonomous vehicles, which can provide to users a more consistent intelligent experience from driving control to in-vehicle functions and connected services. we introduce in this paper our approach in designing human-vehicle interaction from the HRI perspective, which is further composed of three parts: the intelligent sensing, predicting, and decision-making module, the adaptive user interface module, and the intelligent voice module.[......]继续阅读
