半自动驾驶情境下面向整体有用性的HUD界面设计

【作者】 张丽；

【导师】 林丽；

【作者基本信息】 贵州大学 ， 机械， 2023， 硕士

【摘要】 自动驾驶的全面实现将为人们带来极大的便利。但由于网络、机器学习技术、政策等方面的限制,人机共驾是全面实现自动驾驶的必经之路。从手动驾驶到半自动驾驶过渡进程中,驾驶员从单一操作逐步过渡为监管与操作并存。此过程中,认知负荷的降低、被动的监管导致的无聊均会导致驾驶员警惕性下降,从而引发一系列安全问题。合理的HUD能够改善上述问题。但错误设计的HUD可能会使驾驶员产生认知隧道、干扰视觉搜索行为,从而加剧事故风险。由此,本研究以此为出发点,以提高半自动驾驶安全性为目标,以有用性、实用性、可用性原则作为指导方针,开展了三部分研究:HUD功能实用性研究:搜集整理现有商用自动驾驶车辆人机协同相关的交互信息,结合文献研究法,筛选半自动驾驶情境下实用性HUD信息内容;基于眼动追踪、访谈法与驾驶接管绩效量化实施HUD实用性实验验证,经实验结果分析,对实用性存在争议的HUD,提出应改变元素设计形式的建议。HUD显示形式可用性研究:通过信任度问卷联合脑电测试,推导HUD数量与信任度关系;参考现有商用汽车与专家意见设计了上述研究内容涉及的HUD不同显示形式;通过眼动注视分配量化、驾驶接管绩效测试、自动化模式感知能力测量及态势感知能力测量,获得HUD不同显示形式可用性量化数据。基于整体有用性的HUD设计元素优选及界面方案展示:在上述HUD实用性和可用性研究基础上,针对两种经验驾驶员与两类交通流量驾驶环境,利用AHP排序、优选HUD的设计元素,并展示基于HUD设计元素优选下的新型HUD界面方案。研究为半自动驾驶情境下的HUD界面设计提供了初步的信息选择及显示形式设计的依据和指导,对促进人与自动驾驶车辆交流、提高驾驶安全性、提升交互体验具有实践应用价值。更多还原

【Abstract】 The full realization of autonomous driving will bring great convenience to people.However,due to the limitations of network,machine learning technology and policies,human-machine co-driving is a necessary path for the full realization of autonomous driving.In the transition process from manual driving to semi-autonomous driving,the driver gradually transitions from a single operation to both regulation and operation.In this process,the reduction of cognitive load and boredom caused by passive supervision can lead to a decrease in driver alertness,which can lead to a series of safety problems.A properly designed HUD can improve these problems.However,incorrectly designed HUDs may cause cognitive tunneling and interfere with visual search behavior,thereby increasing the risk of accidents.Thus,with this as a starting point and with the goal of improving the safety of semi-autonomous driving,this study conducted a three-part study using the principles of usefulness,utility,and usability as guidelines:HUD functional utility research: collect and organize the existing interaction information related to human-machine cooperation in commercial autonomous driving vehicles,and combine the literature research method to screen the content of practical HUD information in semi-autonomous driving situations;implement HUD for utility experimental verification based on eye-tracking,interview method and quantification of driving takeover performance,and after experimental results analysis,propose that the design of elements should be changed for HUDs with controversial utility Suggestions for changing the design of elements for HUDs with controversial utility.HUD display form usability study: The relationship between the number of HUDs and the trust level was deduced through the trust level questionnaire and the EEG test;different display forms of HUDs were designed with reference to existing commercial vehicles and expert opinions;quantitative usability data of different display forms of HUDs were obtained through eye gaze distribution quantification,driver takeover performance test,automated mode perception capability measurement and situational awareness capability measurement.HUD design element selection and interface solution demonstration based on overall usefulness: Based on the above-mentioned HUD utility and usability study,we use AHP to rank and select HUD design elements for two types of experienced drivers and two types of traffic flow driving environments,and demonstrate a new HUD interface solution based on HUD design element selection.The study provides a preliminary basis and guidance for information selection and display form design for HUD interface design in semi-autonomous driving situations,which has practical application value for promoting human-autonomous vehicle communication,improving driving safety and enhancing interaction experience.更多还原