Tech Paper. HMI Display Readability During Sinusoidal Vibration

Tech Paper HMI Display Readability During Sinusoidal Vibration

HMI Display Readability During Sinusoidal Vibration Abhilash Marthi Somashankar, Paul Weindorf Visteon Corporation, Michigan, USA James Krier, Wayne Nowicki(Former Visteon) Michigan, USA Abstract Modern automotive cockpit design trends have increased the number of displays and the locations and manner in how they are packaged. One theme in particular is the packaging of the displays in novel locations that may be marginal in terms of dynamic stability during road load vibrations. Examples of this include displays that adjust their position in the vehicle. The image of the display may be partially or fully blurred during vibration events which can produce a poor HMI experience and potential safety issues. This paper will present the results of a HMI study that evaluated the readability of different sizes and contrast ratios of TFT color display graphics via jury evaluation during varying vibration acceleration and frequency levels in a controlled lab environment. The result of this study was identification of minimum natural frequencies and maximum acceleration levels for the display mounting structure as a function of display graphics size and contrast ratios. This information is intended to be used by designers of the cockpit structures that package the displays as well as guidance for the HMI design of the display graphics Author Keywords Display, Vibration, Readability, Angular Subtence 1. Introduction The size and quantity of digital displays are increasing in the vehicle as a general market trend. This trend as a general rule has increased the mass of the displays. This increased mass coupled with lower bracket stiffness can cause a reduction in the mechanical natural frequency of these displays. Reduced natural frequency increases the relative motion the observer sees in the form of blurring the display characters when the vehicle is being driven over rough roads. As displays replace traditional mechanical switches, reduced viewability during vibration can cause safety issues since increased viewing time can distract the driver. In addition to the mechanical aspects of structural stability, the design of display character size and contrast ratio is included in the paper for graphic screen design guidelines. The study varied the contrast ratio and size of display characters at black on white and white on black test screens to examine the impact on viewability. Existing knowledge of display HMI viewability [1] are static in nature. They do not factor the vibration environment that needs to be considered. As stated previously, for historical display applications in clusters and centerstacks that achieve relatively rigid mounting with higher natural frequencies, this is not an issue. This paper presents the results of a jury evaluation of a digital display that was tested for readability at varying display character parameters and vibration levels. This paper will recommend design guidelines for natural frequency and sinusoidal acceleration g-level as a function of display character size, angular subtence and contrast ratio. The information will be presented in the following outline. 1. General description of the test setup 2. Identification of the display type, detailed description of the viewed characters, & illumination levels. 3. Jury information 4. Vibration test parameters 5. Data 6. Conclusions 2. Method An 8 diagonal Color TFT display was used in the study. It is a typical size used in modern passenger vehicles found in the centerstack location. The display utilized had the following characteristics: 1000 cd/m 2 (white) 800 x 480 resolution (117ppi) Normally Black ASV type Contrast Ratio = 2000 (typical) Two test screens were designed for the study, see Figure 2-1 & 2-2. Black letters on lighter backgrounds and white letters on darker backgrounds were selected to study the effect of inverse contrast ratios. 11 rows of text with increasing size were chosen to permit discrimination at varying vibration levels. Contrast ratios were also studied by selecting 4 columns with decreasing contrast ratios by adjusting the background luminance.

Figure 2-1. Black letters on lighter background test screen. Contrast Values, Microsoft Powerpoint Black Letters, Lighter Background Column A Column B Column C Column D White Background 35% Transparancy 15% Transparancy 5% Transparancy Figure 2-2. White letters on darker background test screen Calibri(Body) font was chosen, it is a common office type font with relatively even font width. Text height was measured optically on the display screen and reported in mm. The measured heights of W, e & h are reported in Table 2-1. These letters represent the range in height of the text characters in millimeters. Table 2-1. Character font heights used for test screens Measured Display Letter Height (mm)- "We hold" Font Size Font Type (Point) "W" "e" "h" White Letters, Darker Background Column A Column B Column C Column D 35% Transparancy 15% Transparancy 5% Transparancy The luminance and contrast ratios of the test screens were measured using a Lumicam 1300 video photometer per Table 2-3 & 2-4. To reduce measurement sampling noise, special test screens that permitted a substantial area of measurement for the Lumicam were used instead of attempting to measure only the relatively small text area. These special test screens were comprised of creating black or white squares to represent the text using the same background shading color per Figure 2-3. Figure 2-4 shows the display with the special test screens and the Lumicam. For the actual measurements, a dark hood was used to prevent any ambient light from influencing the camera measurements. Row 1 Calibri (Body) 18 1.7 1.2 1.8 Row 2 Calibri (Body) 20 2 1.7 2.1 Row 3 Calibri (Body) 24 2.4 1.8 2.5 Row 4 Calibri (Body) 28 3 2 3.1 Row 5 Calibri (Body) 32 3.2 2.4 3.4 Row 6 Calibri (Body) 36 3.7 2.9 3.8 Row 7 Calibri (Body) 40 4.3 3.3 4.5 Row 8 Calibri (Body) 44 4.8 3.5 5 Row 9 Calibri (Body) 48 5 3.9 5.4 Row 10 Calibri (Body) 54 5.8 4.8 6.2 Row 11 Calibri (Body) 60 6.5 5.9 7 Background shading was chosen to decrease the contrast ratio of the characters, Microsoft PowerPoint was used to create transparent shading for the columns, and Table 2-2 identifies the used. Table 2-2. Microsoft PowerPoint transparency shading settings used to generate background contrast Figure 2-3. Lumicam special test screens to measure luminance

Figure 2-4. Display and Lumicam prior to luminance measurements Table 2-3: Luminance measurements of Figure 1 black text on lighter background Black Text, Lighter Background Column A Column B Column C Column D Black Text (Cd/m2) 2.7 2.1 1.6 1.4 Background (Cd/m2) 615.8 69.7 11.4 2.1 Contrast Ratio (Background/Text) 228.1 33.2 7.1 1.5 Table 2-4: Luminance measurements of Figure 2 white text on darker background White Text, Darker Background Column A Column B Column C Column D White Text (Cd/m2) 602.9 67.4 9.9 1.8 Background (Cd/m2) 1.5 1.1 1 1 Contrast Ratio (Text/Background) 401.9 61.3 9.9 1.8 Some variation in the measurements was recorded, this was attributed to variation in the backlight behind the display glass. A sinusoidal vibration test was designed to cover varying acceleration levels and frequencies. Sinusoidal was chosen over random because although random is a more realistic test condition in the automotive environment, the details of a random vibration test s power spectral density were expected to be more complicated to cover the various packaging locations in the vehicle compared to a simple sinusoidal acceleration vs frequency test. Some preliminary tests were run to set acceleration ( g s ) and frequency (Hz) ranges that would minimize jury evaluation test time. Table 2-5 & Table 2-6 are the that were chosen for the test with calculated maximum displacement and maximum velocity listed, respectively. Table 2-5: Acceleration & frequency versus peak to peak displacement Peak to Peak Displacement (Millimeters): G's vs Hz Frequency (Hz) G Level 10 20 30 40 50 0.25 1.24 0.31 0.14 0.08 0.05 0.5 2.48 0.62 0.28 0.16 0.10 0.75 3.73 0.93 0.41 0.23 0.15 1 4.97 1.24 0.55 0.31 0.20 Table 2-6: Acceleration & frequency versus maximum velocity Max Velocity (Millimeters/Second): G's vs Hz Frequency (Hz) G Level 10 20 30 40 50 0.25 39.02 19.51 13.01 9.75 7.80 0.5 78.04 39.02 26.01 19.51 15.61 0.75 117.06 58.53 39.02 29.26 23.41 1 156.08 78.04 52.03 39.02 31.22 The most difficult viewing conditions would be expected to occur at higher G level and lower frequency since the displacements and velocities are at their greatest that would cause increased blurring. The display was mounted to an electromagnetic vibration table using a fixture that approximated viewing the display normal to the viewer per Figure 2-6. A viewing distance of 27 (686mm) was chosen to represent a taller driver that would represent a worst case viewing distance typical for a centerstack or instrument cluster display location. Figure 2-6. Test viewing condition A jury of 12 people were used for the study. Age and genders are listed per Table 2-7. Table 2-7: Jury age and gender Juror Age Gender 1 23 Male 2 25 Male 3 25 Male 4 38 Female 5 48 Female 6 52 Female 7 55 Male 8 59 Male 9 59 Male 10 61 Male 11 61 Male 12 65 Male The test procedure for the jurors was to examine the white letters with darker background and black letters with lighter background test screens without vibration and identify the smallest text from row D that was readable. No time limit was used to determine this value. The purpose of this static test was to act as a eyetest to discriminate the juror s vision for very difficult low contrast

ratio conditions. The next task was to have the jurors identify the smallest text row number that could be read during vibration at each of the different G and Hz levels. The criteria was readability at a glance to simulate safe driving conditions in the car. If the juror couldn t decide upon two different text row sizes, the larger text value was chosen. During the test, the jurors were instructed to either close their eyes between readings or glance away from the display. 3. Result Figure 3-1 is a graph of the static Eyetest used to discriminate the relative eyesight of each of the jurors. The results represent the smallest text height that could be read without vibration of the lowest contrast column D of the test screens. The chart is shown by age order. White & black text screen results are shown along with a 2 nd degree polynomial curve fit. Figure 3-3: Eyetest histogram of white text, darker background, column D The definition of subtended angle is shown below in Figure 3-4 for reference and the equation (1) used for it [2] is shown below the figure. Subtended angle was selected for reporting of results instead of character height to produce results that are independent of viewing distance and character height. Figure 3-1: Static Eyetest of Column D A histogram of the distribution of the jurors for both test screens is presented in Figure 3-2 and Figure 3-3 Figure 3-4: Subtended Angle definition used to report results in terms of angular subtence Angular Subtence in degrees = 2 arctan ( Angular Subtence in minutes of arc = 60 x 2 arctan ( target height target height ) 2 x viewing distance ) 2 x viewing distance 3438 x target height = ( ) (1) viewing distance Figure 3-2: Eyetest histogram of black text, lighter background, column D Figure 3-5 thru 3-10 are summaries of the average Angular subtense that was legible during the test vibration G levels and frequencies. A separate chart is shown for each column representing a unique contrast ratio for both the black letters on lighter background of Figure 2-1 and the white letters on darker background of Figure 2-2.

Figure 3-8: Column A legible vibration response Figure 3-5: Column A Legible vibration response Figure 3-9: Column B legible vibration response Figure 3-6: Column B legible vibration response Figure 3-10: Column C legible vibration response Figure 3-7: Column C legible vibration response Table 3-1 thru 3-6 is the same average legibility data as Figure 3-5 thru 3-10 but in table form.

Table 3-1: Column A legible vibration angular subtense Dark Letters, Lighter Background Hz Column A - Averages 1/2G 3/4G 1 G 50 7.04 7.29 7.46 40 7.46 7.33 7.67 30 7.71 8.46 9.88 20 8.92 11.08 12.55 10 10.83 13.22 16.95 Table 3-2: Column B legible vibration angular subtense Dark Letters, Lighter Background Column B - Averages Hz 0.5G 0.75G 1.0G 50 7.50 7.50 7.58 40 7.54 7.79 8.00 30 8.21 8.67 9.96 20 9.38 11.08 12.55 10 11.34 13.51 17.88 Table 3-3: Column C legible vibration angular subtense Dark Letters, Lighter Background Column C - Averages Hz 0.5G 0.75G 1.0G 50 7.79 8.08 8.29 40 8.25 8.25 9.04 30 9.04 9.63 10.38 20 10.21 11.01 12.55 10 12.13 14.61 18.99 Table 3-4: Column A legible vibration angular subtense White Letters, Darker Background Hz Column A - Averages 1/2G 3/4G 1 G 50 7.58 7.29 7.54 40 7.79 8.00 7.92 30 8.21 8.88 9.00 20 10.08 11.50 12.76 10 12.79 14.69 18.03 Table 3-5: Column B legible vibration angular subtense White Letters, Darker Background Column B - Averages Hz 0.5G 0.75G 1.0G 50 7.92 7.38 7.63 40 8.17 8.08 8.17 30 8.83 9.38 9.29 20 10.08 11.09 12.54 10 13.35 15.07 18.24 Table 3-6: Column C legible vibration angular subtense White Letters, Darker Background Column C - Averages Hz 0.5G 0.75G 1.0G 50 8.67 7.63 8.67 40 9.46 8.17 9.46 30 10.33 9.29 10.33 20 12.88 12.54 12.88 10 19.66 18.24 19.66 4. Discussion The test provided data for relative viewability of different text sizes and contrast ratios at varying vibration G and frequency levels. In order to extract conclusions from the data, a number of assumptions need to being made. Assumptions: -The jury s eyesight is representative of the population as a whole. -Average jury text size are being reported for a given test. Several of the jurors appeared to have worse eyesight than the rest of the group. This should bias the results to a somewhat

conservative value. -The sinusoidal vibration test generates a fixed observer and moving display producing a given amount of relative motion. Actual vehicle vibration will introduce motion of both the observer and the display which could either increase or decrease the amount of relative motion depending on if the driver and display are moving in phase or out of phase with each other. -Only linear components of vibration in one degree of freedom are considered. 5. Summary and Conclusion The test results show that: - As the frequency is decreased at a given G level, a larger character angular subtense (i.e. larger Character Height) is required to maintain visibility. - As the G level is increased at lower frequencies (10-30Hz), a larger character angular subtense (i.e. larger Character Height) is required to maintain visibility. From the results, it can be concluded that the target display module frequency range should be greater than 25Hz at a given G level. Below this frequency, the angular subtense increases above a normal font size. 6. References [1] INTERNATIONAL STANDARD ISO/FDIS 15008 SAE, Road vehicles Ergonomic aspects of transport information and control systems Specifications and compliance procedures for in-vehicle visual presentation ISO/FDIS Standard 15008:2002(E) [2] INTERNATIONAL STANDARD ISO 13406-2, Ergonomic requirements for work with visual displays based on flat panels -- Part 2: Ergonomic requirements for flat panel displays

White Paper About Visteon Visteon is a global company that designs, engineers and manufactures innovative cockpit electronics products and connected car solutions for most of the world s major vehicle manufacturers. Visteon is a leading provider of instrument clusters, head-up displays, information displays, infotainment, audio systems, telematics and SmartCore cockpit domain controllers. Visteon also supplies embedded multimedia and smartphone connectivity software solutions to the global automotive industry. Headquartered in Van Buren Township, Michigan, Visteon has approximately 10,000 employees at more than 40 facilities in 18 countries. Visteon had sales of $3.16 billion in 2016. Learn more at www.visteon.com. Visteon Corporation One Village Center Dr. Van Buren Township, MI 48188 1-800-VISTEON www.visteon.com Copyright 2017 Visteon Corporation