An Analysis on Auditory Displays for Multiple Warnings under Severe Temporal Constraints
著者
長谷川 暖, 古川 宏(筑波大学大学院システム情報工学研究科リスク工学専攻)
Abstract
Intelligent Transport System (ITS) and Vehicle Information and Communication System (VICS) have ability to simultaneously inform drivers about several different affairs occurred in same time. This situation may drive them into a corner, where they must catch overwhelming information while driving a car. In this paper, we evaluate feasibility and effectiveness of two auditory display methods which can mitigate the difficulty. The first method is auditory overlapping display giving different phrases with different start timings. The second method is using plural speakers in different places. In a cognitive experiment, participants were asked to catch three different auditory phrases and select them from several alternatives written on an answering card. The quantitative evaluations show that the first method can support users to understand multiple phrases displayed in parallel manner. This result may suggest that users with the display can catch multiple information in a shorter time than with the ordinary sequential display method.
キーワード
音声警報, 多重表示, 時間制約下, ITS, 認知的制約
Keywords
auditory warnings, overlapping display, severe temporal constraints, ITS, cognitive constraints
Recently, three-dimensional images rapidly spread in public, they have come to be used in mobile devices. In this study, we evaluated the visibility of the 3D mobile phone LYNX 3D SH-03C. As a concrete item, we examined capabilities of 3D stereognosis according to age, relationship between popup distance and pupil distance, and visibility and readability of solid character in each evaluation, deflection angle, and turning angle. As a result, capabilities of stereognosis didn’t depend on the age, and popup distance and pupil distance followed the theory value. Moreover, about the visibility and the readability of solid character of each angle, it has been understood that stereognosis and reading out become difficult increases of the angle.
キーワード
立体視, 3Dケータイ, パララックスバリア方式, 視認性, 可読性
Keywords
stereoscopic vision, 3D mobile phone, parallax barrier, visibility and readability
We carried out experiments on the readability of iPad using middle aged and elderly subjects. Subjects were tested their readability for positive and negative polarities displayed on iPad screens. The subjects with low cataract cloudiness could read the texts on iPad faster than those with high cataract cloudiness. Positive polarity display required shorter reading time than negative polarity display among the subjects with low cataract cloudiness. Negative polarity display required shorter reading time than positive polarity display among the subjects with high cataract cloudiness.
キーワード
iPad, 可読性, 陽画表示, 陰画表示, 高齢者
Keywords
iPad, readability, positive polarity display, negative polarity display, elderly people
Accommodation and convergence were measured simultaneously while subjects viewed 3D images. The aim was to evaluate legibility of glasses-free mobile 3D display. 3D images (Power3DTM, Olympus Visual Communications, Inc.) were presented vertically and horizontally so that we measured and compared legibility in both cases. Measurements were made twice, 40 seconds each. The result suggested that ocular functions during viewing of 3D images are very similar to those during natural viewing, and there was no significant difference between vertical and horizontal cases. Previously-established and widely-used theories, such that within a VR system eyes must maintain accommodation on the fixed display, may need to be corrected.
キーワード
立体視, 調節と輻輳, 同時計測, 眼精疲労
Keywords
Stereoscopic Vision, Accommodation and Convergence, Simultaneous Measurement, Visual Fatigue
Visual Search of Elderly Driver when Looking at In-vehicle Navigation Display ? Relationship between Map Elements and Detection Time in Visual Search for Landmarks of Surrounding Facility-
著者
小堀 恒(慶應義塾大学大学院理工学研究科), 大門 樹(慶應義塾大学理工学部)
Abstract
This study focuses on the relationship between detecting ability of elderly drivers and map elements of in-vehicle navigation display, when they search a certain landmark on the display during driving. The experimental condition was determined based on the size, color and location of the certain landmark, the density and size of the other landmarks. The participants were required to search the certain landmark on the in-vehicle display and touch it with their fingers. The data obtained in the experiment were analyzed and evaluated between the elderly and the non-elderly participants. As a result, the factor of “landmark size” is more effective than the other experimental conditions on searching a certain landmark.
キーワード
カーナビゲーション, 高齢ドライバー, 視覚探索, ランドマーク, 地図構成要素
Keywords
car-navigation, elderly driver, visual search, landmark and map element
Crash Rate Analysis on Expressways Using Poisson Mixture Models -For Mobile Device-based Situational Safety Information Service-
著者
熊谷 徹, 赤松 幹之(独立行政法人産業技術総合研究所ヒューマンライフテクノロジー研究部門)
Abstract
Crash rates are not constant throughout the roadways, and some places appear to induce more crashes than others. These crashes are related to specific features of a given roadway and its environment. Crash rates would be drastically reduced by providing drivers with information about high-risk places approaching near at hand, by means of a vehicle-installed system. To this end, it is crucial to identify high-risk points on the roadway. Screening of high-risk points using statistical data from crashes is, however, very difficult, except for points inducing a large number of crashes. A considerable number of crashes may occur even if the crash rate is low, as demonstrated in statistical theory. In this study, we proposed to utilize a Poisson mixture model in order to analyze a given roadway into high-risk (and other) points. In the proposed method, we modeled the statistical data from crashes with a Poisson mixture model composed of two Poisson models, using an iterative EM algorithm; and assigned each point on the roadway to one of the two Poisson models, based on posterior probability of each Poisson model. We determined that points belonging to the model with a higher mean value were high-risk points. We applied the proposed method to the Tomei and Meishin expressways. Most roadway points classified as high-risk points were likely to involve sharp curvature, steep slope and/or merging traffic into a busy lane with heavy traffic around the respective point.
キーワード
混合ポアソンモデル, 高速道路, 事故発生率, 車載機器, 事故多発地点
Keywords
Poisson mixture model, expressway, crash rate, vehicle installed system, high-risk point on the roadway
