登录 EN

添加临时用户

基于触控交互测评的智能座舱用户体验改善方法

Method for Improving User Experience in Intelligent Cockpits Based on Touch Interaction Evaluation

作者:曹婷婷
  • 学号
    2021******
  • 学位
    硕士
  • 电子邮箱
    cao******com
  • 答辩日期
    2025.05.20
  • 导师
    王建强
  • 学科名
    工程管理
  • 页码
    75
  • 保密级别
    公开
  • 培养单位
    015 车辆学院
  • 中文关键词
    智能座舱;用户体验;触控交互;实车测评;测评系统
  • 英文关键词
    Intelligent Cockpit; User Experience; Touch Interaction; In-Vehicle Evaluation; Evaluation System

摘要

近年来,汽车智能化发展迅速,众多前沿科技配置被引入智能座舱。然而,配置数量的增加并不等同于用户体验的提升。在激烈的智能化竞争中,精准提升用户体验是产品脱颖而出的关键。因此,如何评估智能座舱是否能满足用户需求和体验,成为当前研究的重要课题。人机交互(Human-Machine Interaction, HMI)理论强调人与机器之间的自然、高效和安全交互。智能座舱作为人机交互的重要场景,其交互设计需要充分考虑用户的操作习惯、认知负荷以及多模态交互的协同性。目前,随着智能汽车的发展,智能座舱标准研究取得了显著进展,国际标准、国家标准和行业标准等多个层面均已发布了汽车智能座舱的测评标准。但现有测评标准主要集中在性能和交互方面,测试方法中主观评价较多,客观性不足,导致产品良莠不齐,用户体验参差不齐。为解决以上问题,完成了以下研究内容:首先,构建了智能座舱触控交互测评模型。针对智能座舱测评标准不够完善的问题,基于各权威机构现有测评体系,整合测评维度和主客观测评项目,计算各测评标准的匹配度,建立了一套综合测评模型。该模型从视觉协调性、操作层级合理性、人机交互便捷性、文本和图标易理解性、功能完整性和安全性六个评价维度,设计了测试任务并搭建了测评环境,为评价智能座舱触控交互方面的用户体验改善效果,提供了理论支撑。其次,基于上述触控交互测评模型,对五种智能汽车进行了实车测评,验证触控交互测评方法对智能座舱用户体验的改善效果。测试过程中,通过眼动仪、高帧摄像头等测评设备记录主客观测试中的动态与静态数据。经定量与定性分析,发现了触控交互中影响用户体验的问题,并提出改善方法。通过主客观综合测评方法量化用户体验,对比改进前后效果,发现触控交互中存在的影响用户体验的问题(如理想L9的灯光设置的功能分类分散等),并通过方案优化提升用户体验。最后,将触控交互测评的结论应用于整个智能座舱。基于现有智能座舱测评体系,构建了智能座舱测评系统和数据库,涵盖车型、功能、测评项目等信息。这些信息通过数据模型进行链接,实现了智能座舱测评数据的可视化。该系统使产品开发人员能够更精准地发现问题,克服了实车测试周期长、数据处理工作量大的问题。通过智能座舱的测评,实现了整个智能座舱用户体验改善的目标。

In recent years, the development of automotive intelligence has been rapid, with numerous cutting-edge technologies being introduced into the intelligent cockpit. However, an increase in the number of configurations does not equate to an improvement in user experience. Amidst fierce competition in the field of automotive intelligence, accurately enhancing user experience is the key to making a product stand out. Therefore, how to assess whether the intelligent cockpit can meet user needs and provide a satisfactory experience has become an important research topic. Human-Computer Interaction (HCI) theory emphasizes natural, efficient, and safe interactions between humans and machines. As an important scenario for human-computer interaction, the design of the intelligent cockpit's interaction must fully consider users' operational habits, cognitive load, and the synergy of multimodal interactions. Currently, with the development of intelligent vehicles, research on intelligent cockpit standards has made significant progress, and evaluation standards for automotive intelligent cockpits have been released at multiple levels, including international, national, and industry standards. However, existing evaluation standards mainly focus on performance and interaction, with a predominance of subjective evaluation methods and insufficient objectivity, leading to inconsistent product quality and varied user experiences. To address these issues, this thesis has completed the following research:Firstly, an evaluation model for touch interaction in the intelligent cockpit was constructed. In response to the imperfections in the existing evaluation standards for intelligent cockpits, this study integrated evaluation dimensions and subjective and objective evaluation items based on existing evaluation systems from authoritative institutions. By calculating the matching degree of each evaluation standard, a comprehensive evaluation model was established. The model evaluates user experience in touch interaction from six dimensions: visual coordination, rationality of operation hierarchy, convenience of human-computer interaction, ease of understanding of text and icons, functional completeness, and safety. Test tasks were designed, and an evaluation environment was built to provide theoretical support for improving user experience in touch interaction within the intelligent cockpit.Secondly, based on the touch interaction evaluation model, practical vehicle evaluations were conducted on five types of intelligent vehicles to verify the effectiveness of the evaluation method in improving user experience in the intelligent cockpit. During the testing process, dynamic and static data from subjective and objective tests were recorded using evaluation devices such as eye trackers and high-frame-rate cameras. Through quantitative and qualitative analyses, issues affecting user experience in touch interaction were identified, and improvement methods were proposed. By quantifying user experience with a comprehensive subjective and objective evaluation method and comparing the effects before and after improvements, problems in touch interaction were discovered (e.g., the分散function classification of lighting settings in the Li Auto L9), and user experience was enhanced through optimized solutions.Lastly, the conclusions drawn from the touch interaction evaluation were applied to the entire intelligent cockpit. Based on the existing evaluation system for intelligent cockpits, an evaluation system and database for the intelligent cockpit were constructed, covering information on vehicle models, functions, and evaluation items. These pieces of information were linked through a data model to achieve visualization of evaluation data for the intelligent cockpit. This system enables product developers to identify problems more accurately, overcoming issues such as long practical vehicle testing cycles and large data processing workloads. Through the evaluation of the intelligent cockpit, the goal of improving overall user experience was achieved.